31
Preprint 12 January 2021 Evolutionary Map of the Universe (EMU): Compact radio sources in the S field towards the Galactic plane S. Riggi 1 ? , G. Umana 1 , C. Trigilio 1 , F. Cavallaro 11,1 , A. Ingallinera 1 , P. Leto 1 , F. Bufano 1 , R.P. Norris 2,3 , A.M. Hopkins 4,2 , M.D. Filipović 2 , H. Andernach 5 , J.Th. van Loon 6 , M. J. Michalowski 7 , C. Bordiu 1,8 , T. An 9 , C. Buemi 1 , E. Carretti 10 , J.D. Collier 11,2 , T. Joseph 12 , B.S. Koribalski 3,2 , R. Kothes 13 , S. Loru 1 , D. McConnell 3 , M. Pommier 14 , E. Sciacca 1 , F. Schillirò 1 , F. Vitello 10 , K. Warhurst 15 and M. Whiting 3 1 INAF-Osservatorio Astrofisico di Catania, Via Santa Sofia 78, 95123 Catania, Italy 2 School of Science, Western Sydney University, Locked Bag 1797, Penrith, NSW 2751, Australia 3 CSIRO Astronomy & Space Science, P.O. Box 76, Epping, NSW 1710, Australia 4 Australian Astronomical Optics, Macquarie University, 105 Delhi Rd, North Ryde, NSW 2113, Australia 5 Depto. de Astronomía, DCNE, Universidad de Guanajuato, Cjón. de Jalisco, Col. Valenciana, Guanajuato, CP 36023, Mexico 6 Lennard-Jones Laboratories, Keele University, ST5 5BG, UK 7 Astronomical Observatory Institute, Faculty of Physics, Adam Mickiewicz University, ul. Sloneczna 36, 60-286 Poznań, Poland 8 Centro de Astrobiología (INTA-CSIC), Ctra. M-108, km. 4, 28850 Torrejón de Ardoz, Madrid, Spain 9 Shanghai Astronomical Observatory, Chinese Academy of Sciences, 80 Nandan Road, Shanghai, 200030, China 10 INAF Istituto di Radioastronomia, Via Gobetti 101, 40129 Bologna, Italy 11 The Inter-University Institute for Data Intensive Astronomy (IDIA), Department of Astronomy, University of Cape Town, Rondebosch, 7701, South Africa 12 School of Physics and Astronomy, University of Manchester, Oxford Road, Manchester, M13 9PL, UK 13 Dominion Radio Astrophysical Observatory, Herzberg Astronomy and Astrophysics, National Research Council Canada, PO Box 248, Penticton, BC V2A 6J9, Canada 14 Univ Lyon, Univ Lyon1, Ens de Lyon, CNRS, Centre de Recherche Astrophysique de Lyon UMR5574, 9 av Charles André F- 69230, Saint-Genis-Laval, France 15 CSIRO Astronomy & Space Science, 33 Onslow St, Geraldton, WA 6530. Australia Accepted 2020 December 28. Received 2020 December 23; in original form 2020 September 30 ABSTRACT We present observations of a region of the Galactic plane taken during the Early Science Program of the Australian Square Kilometre Array Pathfinder (ASKAP). In this context, we observed the S field at 912 MHz with an uncompleted array consisting of 15 commissioned antennas. The resulting map covers a square region of 40 deg 2 , centred on (l , b)=(343.5 ,0.75 ), with a synthesized beam of 24"×21" and a background rms noise of 150-200 μ Jy/beam, increasing to 500-600 μ Jy/beam close to the Galactic plane. A total of 3963 radio sources were detected and characterized in the field using the source finder. We obtained differential source counts in agreement with previously published data after correction for source extraction and characterization uncertainties, estimated from simulated data. The ASKAP positional and flux density scale accuracy were also investigated through comparison with previous surveys (MGPS, NVSS) and additional observations of the S field, carried out with ATCA at 2.1 GHz and 10" spatial resolution. These allowed us to obtain a measurement of the spectral index for a subset of the catalogued sources and an estimated fraction of (at least) 8% of resolved sources in the reported catalogue. We cross-matched our catalogued sources with different astronomical databases to search for possible counterparts, finding 150 associations to known Galactic objects. Finally, we explored a multiparametric approach for classifying previously unreported Galactic sources based on their radio-infrared colors. Key words: radio continuum: general – catalogues – surveys – Galaxy: general – techniques: interferometric – techniques: image processing ? E-mail: [email protected] © 0000 The Authors arXiv:2101.03843v1 [astro-ph.GA] 11 Jan 2021

EvolutionaryMapoftheUniverse(EMU

  • Upload
    others

  • View
    7

  • Download
    0

Embed Size (px)

Citation preview

Page 1: EvolutionaryMapoftheUniverse(EMU

Preprint 12 January 2021

Evolutionary Map of the Universe (EMU)Compact radio sources in the Scorpio field towards the Galacticplane

S Riggi1 G Umana1 C Trigilio1 F Cavallaro111 A Ingallinera1 P Leto1 F Bufano1RP Norris23 AM Hopkins42 MD Filipović2 H Andernach5 JTh van Loon6M J Michałowski7 C Bordiu18 T An9 C Buemi1 E Carretti10 JD Collier112T Joseph12 BS Koribalski32 R Kothes13 S Loru1 D McConnell3 M Pommier14E Sciacca1 F Schillirograve1 F Vitello10 K Warhurst15 and M Whiting3

1INAF-Osservatorio Astrofisico di Catania Via Santa Sofia 78 95123 Catania Italy2School of Science Western Sydney University Locked Bag 1797 Penrith NSW 2751 Australia3CSIRO Astronomy amp Space Science PO Box 76 Epping NSW 1710 Australia4Australian Astronomical Optics Macquarie University 105 Delhi Rd North Ryde NSW 2113 Australia5Depto de Astronomiacutea DCNE Universidad de Guanajuato Cjoacuten de Jalisco Col Valenciana Guanajuato CP 36023 Mexico6Lennard-Jones Laboratories Keele University ST5 5BG UK7 Astronomical Observatory Institute Faculty of Physics Adam Mickiewicz University ul Słoneczna 36 60-286 Poznań Poland8 Centro de Astrobiologiacutea (INTA-CSIC) Ctra M-108 km 4 28850 Torrejoacuten de Ardoz Madrid Spain9Shanghai Astronomical Observatory Chinese Academy of Sciences 80 Nandan Road Shanghai 200030 China10 INAF Istituto di Radioastronomia Via Gobetti 101 40129 Bologna Italy11The Inter-University Institute for Data Intensive Astronomy (IDIA) Department of Astronomy University of Cape Town Rondebosch 7701 South Africa12School of Physics and Astronomy University of Manchester Oxford Road Manchester M13 9PL UK13 Dominion Radio Astrophysical Observatory Herzberg Astronomy and Astrophysics National Research Council Canada PO Box 248 Penticton BC V2A 6J9 Canada14Univ Lyon Univ Lyon1 Ens de Lyon CNRS Centre de Recherche Astrophysique de Lyon UMR5574 9 av Charles Andreacute F- 69230 Saint-Genis-Laval France15 CSIRO Astronomy amp Space Science 33 Onslow St Geraldton WA 6530 Australia

Accepted 2020 December 28 Received 2020 December 23 in original form 2020 September 30

ABSTRACTWe present observations of a region of the Galactic plane taken during the Early ScienceProgram of the Australian Square Kilometre Array Pathfinder (ASKAP) In this contextwe observed the Scorpio field at 912 MHz with an uncompleted array consisting of 15commissioned antennas The resulting map covers a square region of sim40 deg2 centred on(lb)=(3435075) with a synthesized beam of 24times21 and a background rms noise of150-200 microJybeam increasing to 500-600 microJybeam close to the Galactic plane A total of3963 radio sources were detected and characterized in the field using the caesar source finderWe obtained differential source counts in agreement with previously published data aftercorrection for source extraction and characterization uncertainties estimated from simulateddata The ASKAP positional and flux density scale accuracy were also investigated throughcomparison with previous surveys (MGPS NVSS) and additional observations of the Scorpiofield carried out with ATCA at 21 GHz and 10 spatial resolution These allowed us to obtaina measurement of the spectral index for a subset of the catalogued sources and an estimatedfraction of (at least) 8 of resolved sources in the reported catalogue We cross-matched ourcatalogued sources with different astronomical databases to search for possible counterpartsfinding sim150 associations to known Galactic objects Finally we explored a multiparametricapproach for classifying previously unreported Galactic sources based on their radio-infraredcolorsKey words radio continuum general ndash catalogues ndash surveys ndash Galaxy general ndash techniquesinterferometric ndash techniques image processing

E-mail simoneriggiinafitcopy 0000 The Authors

arX

iv2

101

0384

3v1

[as

tro-

phG

A]

11

Jan

2021

2 S Riggi et al

1 INTRODUCTION

The next-generation deep radio continuum surveys such as the Evo-lutionary Map of the Universe (EMU) (Norris et al 2011) plannedat the Australian SKA Pathfinder (ASKAP) telescope (Johnston etal 2008) will open a new era in radio astronomy with potential newdiscoveries expected in several fields from Galaxy evolution andcharacterization to Galactic ScienceIn the full-operational mode the ASKAP telescope is made of 3612-metre antennas installed inWesternAustralia each one equippedwith a Phased Array Feed (PAF) receiver (Schinckel et al 2012)operating with a bandwidth of 288 MHz over the frequency range700-1800 MHz The PAF system forms 36 beams and provides aninstantaneous field of view ofsim30 square degrees allowingASKAPto survey the southern sky with unprecedented speed (sim220 deg2

per hour at a target 1σ rms of 100 microJybeam) and higher resolution(sim10 at 950 MHz) compared to existing surveysThe array has a maximum baseline of 6 km and was completedin mid 2019 During the commissioning phase the ASKAP EMUEarly Science Program (ESP) was launched (October 2017) inwhich several target fields were observed with the aim of validatingthe array operations the observation strategy and the data reductionpipeline During this preparatory phase it became evident that theimaging performance exceeded those of past observations enablingvalid scientific results even with an incomplete array The Scorpiofield was the first Galactic field observed during the ASKAP ESPusing 15 commissioned antennas (Umana et al in prep)The Scorpio survey (Umana et al 2015) started in 2011 with mul-tiple scientific goals The original objectives were the study andcharacterization of different types of Galactic radio sources witha focus on radio stars and circumstellar regions (eg Hii regions)Recently the study and characterization of stellar relics such asGalactic supernova remnants (SNRs) in connection with observa-tions at different wavelengths (infrared and gamma-ray primarily)has become an additional target of interest (Ingallinera et al 2017)The survey also represents an important testbench for the ASKAPdata reduction pipeline in the Galactic plane and for the analysismethods designed for the upcoming EMU surveyThis paper is the second of a series of works planned with ASKAPScorpio Early Science data In the first paper (Umana et al inprep) we discuss the ASKAPrsquos potential for the discovery of differ-ent classes of Galactic objects in comparison with original ATCAobservations and described the data reduction strategies adopted toproduce the final mosaic The goal of this paper is to report a firstcatalogue of the compact sources present in Scorpio This workwill also serve as a validation on real data of the designed sourceextraction algorithms tested so far with simulated data (Riggi et al2016 2019)The paper is organized as follows The ASKAP Scorpio radio ob-servations and data reduction are briefly described in Section 2In Section 3 we describe the source extraction methodologies usedto build the source catalogue and discuss the typical performanceachieved in source detection and characterization (eg complete-ness reliability positional and flux density accuracy) In Section 4we present the analysis conducted on the resulting source cataloguefrom source counts to spectral indices while in Section 5 we reporta comparison with existing astronomical databases and a prelimi-nary study of unclassified sources Finally we report in Section 6 asummary of the results obtained and future prospects

2 OBSERVATIONAL DATA OF THE SCORPIO FIELD

21 ASKAP 912 MHz observations and data reduction

The Scorpio field was observed in January 2018 with 15 anten-nas equipped with the new PAF system version (Mk II) in band 1(from 792 to 1032 MHz) In this array configuration the minimumand maximum baselines were respectively 224 m and 23 km Theformer corresponds to a maximum theoretical largest angular scale(LAS) around 50 arcmin at 912 MHz The total surveyed area cov-ers sim40 square degrees centred on l=3435 b=075 extendingby a factor of sim48 the area surveyed with past Scorpio observa-tions done with the Australian Telescope Compact Array (ATCA)(Umana et al 2015 Ingallinera et al 2019)The calibration and imaging procedures adopted to produce the finalmosaic (shown in Fig 1 and referred to as the ScorpioASKAPmapin the rest of the paper) are described in detail in the Scorpio paper1 (Umana et al in prep) The green contour in Fig 1 delimits thefield region considered for source extraction (see Section 3) whilethe yellow contour denotes the Scorpio region observed with theATCA telescope at 21 GHz (Umana et al 2015) (see Section 22for details) The synthesized beam of the final map in J2000 coor-dinates is 24times21 at a position angle of 89The background level and rms noise were estimatedwith the caesarfinder using parameter values reported in Table A1 The backgroundrms noise was obtained by interpolating the median absolute devia-tion (MAD) of pixel fluxes computed over moving sampling boxesof size 10 times the area of the synthesized beam The backgroundlevel varies considerably across the surveyed area In Fig 2 wereport the estimated background noise in microJybeam as a functionof the Galactic latitude coordinate b averaged over the Galacticlongitude coordinate l We observe a noise level sim200 microJybeamin regions far from the Galactic plane and without bright sourcesClose to the Galactic plane the background noise increases due tothe Galactic diffuse emission and the bright emission from extendedsources filling the beam of the telescopes and increasing the sys-tem temperature In regions free of extended sources we observe abackground noise around 500-600 microJybeam Only 20 of the fieldarea has a 5σ noise level smaller than 1 mJybeam while forsim70of the field the 5σ noise is smaller than 2 mJybeam

22 ATCA 21 GHz observations and data reduction

The Scorpio field was observed with the Australia Telescope Com-pact Array (ATCA) in the 6A and 6D configurations at the referencefrequency of 21 GHz using the 16-cm CABB receiver (observingband from 11 to 31 GHz) (Wilson et al 2011) In this array config-uration the theoretical upper limits for the LAS ranges fromsim43 tosim122 arcmin The observations conducted in different runs from2011 to 2012 the data reduction strategy and the scientific resultsare extensively described elsewhere (Umana et al 2015 Riggi et al2016 Cavallaro et al 2018 Ingallinera et al 2019) ATCA observa-tions cover only a small portion of the Scorpio field observed withASKAP equivalent to 84 square degrees (see Fig1 for a compar-ison of the surveyed area size) In Fig C1 we present the ScorpioATCA mosaic We refer to this as the Scorpio ATCA map Theachieved rms is sim30-40 microJybeam and the synthesized beam inJ2000 coordinates is 98times58 (position angle of -3)The ATCA data obtained with a bandwidth of sim17 GHz weredivided into 7 sub-bands (〈νGHz〉=1449 1681 1844 20652337 2614 2895) and imaging was independently performed oneach of them to produce additional mosaics The 2895-GHz chan-nel map was not considered as it was significantly affected by noise

EMU Compact radio sources in SCORPIO 3

Figure 1 912 MHz mosaic of the Scorpio region observed with ASKAP The purple crosses indicate the beam centre positions for the three interleavedpointings The green contour delimits the mosaic area considered for source catalogue extraction with the caesar source finder The yellow contour denotesthe Scorpio region observed with the ATCA telescope at 21 GHz (see text)

and imaging artefacts The remaining sub-band mosaics (1-6) alongwith the full band mosaic are used throughout the paper as ancillarydata to complement the ASKAP catalogue with value-added infor-mation such as the source spectral indices (see Section 42) or toestimate the expected fraction of extended sources (see Section 41)

23 MOST 843 MHz observations

The Molonglo Galactic Plane Survey 2nd Epoch (MGPS-2) (Mur-phy et al 2007) carried out with the Molonglo Observatory Syn-thesis Telescope (MOST) at a frequency of 843 MHz completelycovers the Scorpio field observed with ASKAP with a lower spa-tial resolution (45times45 cosec|δ |) and a source detection thresholdof sim10 mJy 799 MGPS sources fall in the Scorpio region Theirposition uncertainty is considered better than 1-2 (Murphy et al2007)

24 NVSS 14 GHz observations

The NRAO VLA Sky Survey (NVSS) (Condon et al 1998) coversthe Scorpio region north of DEC=-40 at a frequency of 14 GHzwith an angular resolution of 45 The detection threshold is sim25mJy A number of 853 NVSS sources fall in the Scorpio region

25 TGSS 150 MHz observations

The TIFR GMRT Sky Survey (TGSS) (Intema et al 2017) fullycovers the Scorpio field at the reference frequency of 150 MHzand with an angular resolution of 25times25cos(DECminus 19) anda median rms noise of 35 mJybeam 249 sources from the firstalternative data release (ADR) fall in the Scorpio region

26 GLEAM 200 MHz observations

TheGaLactic and Extragalactic All-sky Murchison Widefield Array(GLEAM) survey (Hurley-Walker et al 2017) partially covers theScorpio field in (1 le |b| le10 345 lt l lt67) at the referencefrequency of 200MHz (bandwidth 60MHz) with an angular resolu-tion ofsim2 arcmin and an rms noise of 10-20 mJybeam 51 sourcesfrom the GLEAM Galactic plane catalogue (Hurley-Walker et al2019) fall in the Scorpio region

27 Supplementary surveys

In this work we will also make use of the following infrared surveysfor source classification studies (see Section 53)

bull AllWISE (Cutri et al 2013) of the Wide-field Infrared Survey

4 S Riggi et al

3minus 2minus 1minus 0 1 2 3

b (deg)0

200

400

600

800

1000

1200

1400

1600

Jyb

eam

)micro

(rm

Figure 2 Estimated background noise of Scorpio mosaic in microJybeam asa function of the Galactic latitude b and averaged over Galactic longitude l

Explorer (WISE Wright et al 2010) The survey is fully coveringthe Scorpio mosaic region with sim93times105 sources detected withSNgt5 in at least one of the four bands at 34 microm (W1) 46 microm(W2) 12 microm (W3) and 22 microm (W4) The angular resolutions are61 64 65 and 12 and the 5σ flux sensitivities for point sourcesare 008 mJy 011 mJy 1 mJy and 6 mJy respectivelybull GLIMPSE (Galactic Legacy Infrared MidPlane Survey Ex-

traordinaire) 80 microm surveys (Churchwell et al 2009) of the SpitzerSpace Telescope (Werner et al 2004) The surveys (GLIMPSE-I andGLIPMSE-3D) are partially covering (sim74) the Scorpio mosaicregion with 82times105 sources detected with SNgt5 The angularresolution is 2 and the 5σ flux sensitivity sim04 mJybull Hi-GAL (Herschel infrared Galactic plane Survey) 70 microm sur-

vey (Molinari et al 2016) of the Herschel Space Observatory (Pil-bratt et al 2010) The survey is partially covering (sim50) theScorpio mosaic region with 5654 sources detected with SNgt5The angular resolution is sim85 and the 1σ flux sensitivity sim20MJysr

3 COMPACT SOURCES IN THE SCORPIO FIELD

31 Source finding

Compact sources were extracted from the Scorpio ASKAP mapwith the caesar source finder (Riggi et al 2016 2019) using aniterative flood-fill algorithm in which the detection threshold isinitially set to 5σ and the flooding threshold to 25σ At each itera-tion the background and noise maps are recomputed excluding thesources found in the previous iteration and the detection thresholdis lowered in steps of 05σ Two iterations were used in this workcorresponding to a final 45σ seed threshold level Algorithm pa-rameter values reported in Table A1 resulted from a fine-tuningprocedure carried out both on the simulated data sample describedin Riggi et al (2019) and on the simulated maps described in Sec-tion 33We detected 5663 islands1 in the map 5413 of these were selected

1 By island (or blob) we denote a group of connected pixels with bright-ness above a merge threshold and around a seed pixel with brightness above

Table 1 Number of sources extracted from the Scorpio ASKAP mosaic at912MHzwith the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 250 0 01 4754 3857 37862 440 308 1733 131 59 4gt3 88 38 0

All 5663 4262 3963

as compact source candidates and successfully fitted with a mixtureof Gaussian componentsTo reject imaging artefacts and sources with a poor quality charac-terization we applied these selection criteria to the extracted sourcesample

bull Source fit converged with a χ2 lt10bull Positive fitted component peak fluxbull Fitted component centroid inside the source island and the

mosaic boundary regionbull Separation between any pair of source components larger than

8 (or 2 pixels)

After the selection 4262 source islands and 4813 fitted source com-ponents are left in the preliminary catalogueThese source counts still include spurious sources mainly due toimaging artefacts and over-deblending of extendeddiffuse emissionsurviving the selection criteria To produce the final catalogue wevisually inspected the entire field labelling each source componentas real or spurious in case of a clear unambiguous identifica-tion 4144 fitted source components were tagged as real in thefinal catalogue Taking into account that the selected survey regionis 377 square degrees a density ofsim110 selected compact sourcesper square degree is obtainedIn Table 1 we report a summary of the number of sources ex-tracted at different selection stages Source numbers labelled withSEL+VIS SEL refer to the number of sources passing both theselection criteria and the visual selection The number of detectedsources classified per number of fitted components is also reportedUsing the same procedure and source selection adopted for theASKAP catalogue we extracted 2227 sources and 2369 fitted com-ponents from the Scorpio ATCA mosaics The source numbersobtained at different selection stages on the full bandwidth map arereported in Table 2

32 Source cross-matching

To validate and complement the ASKAP catalogue we cross-matched it with the ATCA catalogue and with the catalogues ofsupplementary radio (MGPS NVSS TGSS GLEAM) and infrared

a detection threshold Island counts include also islands nested in otherislands

EMU Compact radio sources in SCORPIO 5

Table 2 Number of sources extracted from the Scorpio ATCA mosaic at21 GHz with the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 146 0 01 3021 2104 20962 465 188 1203 158 53 11gt3 102 30 0

All 3892 2375 2227

(AllWISE GLIMPSE Hi-GAL) surveys listed in Section 2 A sum-mary of the match results is reported in Table 3 and details areprovided in the following sections

321 Cross-matching with ATCA source catalogue

731 ASKAP source components (out of 856 components in thecatalogue falling in the ATCA mosaic region) match with at leastone source component from the ATCA catalogue within a searchradius of 24 We visually inspected each cross match rejectingspurious and unclear associations eg cases in which the ASKAPsource flux density measurement is potentially affected by closebackground sources visible in theATCAmap but not inASKAPWefinally selected 648 matches The majority are one-to-one matches(596) while the remaining (52) are one-to-many matches (up tothree components) The number ofmatches purely arising by chancewas estimated by averaging the number of matches found betweenASKAP catalogue and multiple random ATCA catalogues in whichthe measured source positions were uniformly randomized insidethe ATCA mosaic With this procedure we found 38plusmn03 matchesbetween both catalogues We thus concluded that sim995 percent ofthe matches found are likely to be real

322 Cross-matching with supplementary radio surveys

We found 688 MGPS sources potentially associated to ASKAPsources within a match radius of 45 546 of them were finally se-lected for further analysis after excludingmulti-matches and sourceswith potentially unreliable flux densities eg affected by imagingartefacts or closely located to very extended sourcesWithin a match radius of 45 we found 226 NVSS sources matchingto one or more ASKAP source components 189 single matcheswere finally selected after a visual inspection192 out of 217 TGSS source matches can be associated to a singleASKAP source within a match radius of 24 Similarly 32 out of 40GLEAM source matches found within a radius of 45 were selectedfor further analysis after excluding multi-matches or ambiguousassociations

323 Cross-matching with supplementary infrared surveys

For the cross-match we considered the AllWISE survey as the pri-mary dataset reference due to its full coverage To limit the number

Table 3 Number of cross-matches found between ASKAP source compo-nent catalogue and different radio and infrared surveys Column 4 reportsthe total number of sources falling in the Scorpio mosaic for each surveycatalogue In the last two rows we summed up the number of sources fromdifferent catalogues

Surveys radius () matches total

ATCA 24 648 2369MGPS 45 546 799NVSS 45 189 853TGSS 24 192 249GLEAM 45 32 51AllWISE 8 384 123374AllWISE + GLIMPSE 8 225 -AllWISE + GLIMPSE + Hi-GAL 8 41 -dagger

GLIMPSE catalogue has 820112 sources in Scorpiodagger Hi-GAL catalogue has 5654 sources in Scorpio

of spurious associations we selected sources with SNgt5 in boththe W3 and W4 bands with a fraction of saturated pixels smallerthan 10 reducing the number of sources within the ASKAP fieldto sim12times105 A crossmatch with ASKAP source catalogue usinga match radius equal to the ASKAP beam size (24) yields 1319matches About 2900 catalogued sources do not therefore have IRcounterparts with fluxes above W3 and W4 detection thresholdsThe number of source associations arising by chance estimatedusing artificial catalogues with random offsets applied is howevercompatible with the number of obtained matches indicating thatthe vast majority of the matches are spuriousSelecting a smaller radius (eg 8 as in the Galactic object searchanalysis) and requiring flux information in all bands leads to 384associations allowing us to reduce the number of potential spuriousmatches to sim40 The number of associations decreases to 225and 41 respectively if we require triple and quadruple matches withGLIMPSE and Hi-GAL cataloguesThe impact of the 5σ limits requested in W3 and W4 bands wasinvestigated on the stargalaxyquasar dataset provided by Clarke etal (2020)2 The magnitude selection cuts W3lt1132 andW4lt803

were found to remove a very large fraction (sim98) of extragalac-tic sources from the sample but unfortunately also potential radiostars We thus expect the unmatched sources to be mostly extra-galactic with a smaller percentage of IR-quiet (eg pulsars) orfaint mid-infrared Galactic objects

33 Catalogue selection effects and uncertainties

To evaluate the expected source extraction and characterization ac-curacy of the produced catalogue we made use of both simulateddata and cross-matches found with other surveysSimulated samples were drawn from the Scorpio mosaics (bothASKAP and ATCA) using the following approach4 First a resid-ual map was obtained by subtracting all compact sources down toa very low detection threshold (2σ ) Imaging artefacts extendedand diffuse sources were not removed A sample of 50 simulatedmaps was then generated by artificially adding uniformly spaced

2 105281zenodo37683983 See the WISE Explanatory Supplement at httpwise2ipac

caltechedudocsreleaseallskyexpsup4 Simulation tasks (compact source subtraction point-source generation)can be performed using application scripts provided along with caesar

6 S Riggi et al

0

02

04

06

08

1

com

plet

enes

s F

DR

1minus 05minus 0 05 1 15 2 25 3

(SmJy)10

log

Completenessaskapatca

FDRaskapaskap (sclass)atcaatca (sclass)

Figure 3 Completeness (filled markers) and false detection rate (FDR)(open markers) as a function of the injected and measured source flux den-sity respectively obtained on the simulatedASKAP (redmarkers) andATCA(green markers) source catalogues Solid lines represent the fitted complete-ness model of equation 1 Open squares and diamonds (labelled as sclass)represent the false detection rate obtained after applying a neural networkclassifier trained to identify spurious fit components in the catalogue

point sources to the residual mosaic with a density of 100 degminus2

and exponential flux density distribution (propeminusλS with λ=16 asfor real data above the detection threshold) caesar was finally runon the simulated maps and the same set of quality criteria usedfor the Scorpio mosaics were applied to the simulated catalogueSource extraction and characterization metrics were estimated bycross-matching injected sources with extracted sources

331 Completeness and reliability

Catalogue completeness and reliability were studied with simu-lated data as a function of the source flux density Sources injectedin the simulated maps are considered as the true catalogue Ex-tracted sources that do not crossmatch to any injected source arethus counted as false detections and contribute to increase the falsedetection rate (or decrease the reliability) Results are reported inFig 3 for both ASKAP (red markers) and ATCA (green markers)simulated data The completeness C is shown with filled markersand can be parametrized as a function of the flux density S as

C =C0

2

(1+ erf

log10(S)minus log10(S0)

k

)(1)

with parametersC0=099 log10(S0)=minus288 k=040 (ASKAP) andC0=099 log10(S0)=minus358 k=059 (ATCA) For ASKAP (ATCA)observations we expect the catalogue to be gt90 complete above5 mJy (1 mJy)The false detection rate reported in Fig 3 with open dots (ASKAP)and open triangles (ATCA) is found to be of the order of 20-30 slightly decreasing by a few percent when moving away fromthe Galactic plane False detections are largely due to sidelobesaround bright sources (compact or extended) and overdeblendingof extended sources Reliability can be improved by sim10 afterapplying a neural network classifier previously trained to identifygood sources from the data (see Riggi et al 2019 for more de-tails) Results are shown in Fig 3 with open squares (ASKAP) andopen diamonds (ATCA) To further reduce the false detection rate

to acceptable levels (below 1) we visually inspected the producedcatalogue removing spurious sources that were not identified in theautomated procedure (see Section 31)For the future we expect this can be substantially improved in mul-tiple ways eg increasing the number of classifier parameters tun-ing of classifier hyperparameters introducing specialized classifierstrained to identify imaging artefacts (eg around bright sources)which cannot be efficiently removed with the classifier adopted inthis work

332 Source position accuracy

Positional uncertainties in RA and Dec can be expressed as

σα =radic

σ2α f it +σ2

αcal

σδ =radic

σ2δ f it +σ2

δ cal

(2)

where σ f it are the position uncertainties due to the fitting process(including noise and uncertainties in fit parameters) while σcal arethe uncertainties due to the calibration processσ f it is provided by the fitting routine for each source and can becompared with values obtained from simulated data as a functionof the source signal-to-noise ratio (SN) We found that the fit un-certainties are largely underestimated with respect to the expectedvalues found in the simulations The latter are a factor 3-4 larger thanthe semi-analytical estimates provided by Condon (1997)5 This ispartly expected given that the semi-analytical estimate and the esti-mate obtained from the fitting routine are neglecting the effect of thecorrelated noise in the map A correction for this effect for exam-ple using equation 41 of Condon (1997) in the maximal-smoothinglimit would yield an expected analytical uncertainty sim26 largerWe have finally set σ f it to the values obtained from the simulationsand parametrized it as a function of SN as

σASKAPα f it =

986SN

arcsec σASKAPδ f it =

601SN

arcsec

σATCAα f it =

228SN

arcsec σATCAδ f it =

298SN

arcsec(3)

Position fit uncertainties are of the order of 15-18 (05-06) atthe detection threshold and better than 05 (015) with SNgt20 forASKAP (ATCA) dataCalibration uncertainties σcal can be inferred from the posi-tion spread observed in bright sources with respect to a refer-ence catalogue The Radio Fundamental Catalogue (RFC versionrfc_2020b)6 for instance containssim17000 radio sources measuredinmultiple VLBI observations withmilliarcsecond accuracy 5 RFCsources detected in the X band cross-match with ASKAP brightsources (four with SN gt200 and one with SN gt20) and haveposition uncertainty smaller than 002 The standard deviations(sα=05 sδ=04) of the observed position offset provide a firstmeasure of σcal since both RFC and ASKAP fit position uncertain-ties are negligible at high SN Additionally we inferred ASKAPcalibration errors using 344 and 80 sources with SNgt50 match-ing to MGPS and NVSS catalogues respectively The observedstandard deviations (sα=13-15 sδ=13-14) suggest a slightly

5 A discrepancy (a factor sim2) between analytical and measured uncertain-ties was observed also in other analysis carried out with simulated data usingcaesar (Riggi et al 2019) or alternative finders (Hopkins et al 2015)6 httpastrogeoorgrfc

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 2: EvolutionaryMapoftheUniverse(EMU

2 S Riggi et al

1 INTRODUCTION

The next-generation deep radio continuum surveys such as the Evo-lutionary Map of the Universe (EMU) (Norris et al 2011) plannedat the Australian SKA Pathfinder (ASKAP) telescope (Johnston etal 2008) will open a new era in radio astronomy with potential newdiscoveries expected in several fields from Galaxy evolution andcharacterization to Galactic ScienceIn the full-operational mode the ASKAP telescope is made of 3612-metre antennas installed inWesternAustralia each one equippedwith a Phased Array Feed (PAF) receiver (Schinckel et al 2012)operating with a bandwidth of 288 MHz over the frequency range700-1800 MHz The PAF system forms 36 beams and provides aninstantaneous field of view ofsim30 square degrees allowingASKAPto survey the southern sky with unprecedented speed (sim220 deg2

per hour at a target 1σ rms of 100 microJybeam) and higher resolution(sim10 at 950 MHz) compared to existing surveysThe array has a maximum baseline of 6 km and was completedin mid 2019 During the commissioning phase the ASKAP EMUEarly Science Program (ESP) was launched (October 2017) inwhich several target fields were observed with the aim of validatingthe array operations the observation strategy and the data reductionpipeline During this preparatory phase it became evident that theimaging performance exceeded those of past observations enablingvalid scientific results even with an incomplete array The Scorpiofield was the first Galactic field observed during the ASKAP ESPusing 15 commissioned antennas (Umana et al in prep)The Scorpio survey (Umana et al 2015) started in 2011 with mul-tiple scientific goals The original objectives were the study andcharacterization of different types of Galactic radio sources witha focus on radio stars and circumstellar regions (eg Hii regions)Recently the study and characterization of stellar relics such asGalactic supernova remnants (SNRs) in connection with observa-tions at different wavelengths (infrared and gamma-ray primarily)has become an additional target of interest (Ingallinera et al 2017)The survey also represents an important testbench for the ASKAPdata reduction pipeline in the Galactic plane and for the analysismethods designed for the upcoming EMU surveyThis paper is the second of a series of works planned with ASKAPScorpio Early Science data In the first paper (Umana et al inprep) we discuss the ASKAPrsquos potential for the discovery of differ-ent classes of Galactic objects in comparison with original ATCAobservations and described the data reduction strategies adopted toproduce the final mosaic The goal of this paper is to report a firstcatalogue of the compact sources present in Scorpio This workwill also serve as a validation on real data of the designed sourceextraction algorithms tested so far with simulated data (Riggi et al2016 2019)The paper is organized as follows The ASKAP Scorpio radio ob-servations and data reduction are briefly described in Section 2In Section 3 we describe the source extraction methodologies usedto build the source catalogue and discuss the typical performanceachieved in source detection and characterization (eg complete-ness reliability positional and flux density accuracy) In Section 4we present the analysis conducted on the resulting source cataloguefrom source counts to spectral indices while in Section 5 we reporta comparison with existing astronomical databases and a prelimi-nary study of unclassified sources Finally we report in Section 6 asummary of the results obtained and future prospects

2 OBSERVATIONAL DATA OF THE SCORPIO FIELD

21 ASKAP 912 MHz observations and data reduction

The Scorpio field was observed in January 2018 with 15 anten-nas equipped with the new PAF system version (Mk II) in band 1(from 792 to 1032 MHz) In this array configuration the minimumand maximum baselines were respectively 224 m and 23 km Theformer corresponds to a maximum theoretical largest angular scale(LAS) around 50 arcmin at 912 MHz The total surveyed area cov-ers sim40 square degrees centred on l=3435 b=075 extendingby a factor of sim48 the area surveyed with past Scorpio observa-tions done with the Australian Telescope Compact Array (ATCA)(Umana et al 2015 Ingallinera et al 2019)The calibration and imaging procedures adopted to produce the finalmosaic (shown in Fig 1 and referred to as the ScorpioASKAPmapin the rest of the paper) are described in detail in the Scorpio paper1 (Umana et al in prep) The green contour in Fig 1 delimits thefield region considered for source extraction (see Section 3) whilethe yellow contour denotes the Scorpio region observed with theATCA telescope at 21 GHz (Umana et al 2015) (see Section 22for details) The synthesized beam of the final map in J2000 coor-dinates is 24times21 at a position angle of 89The background level and rms noise were estimatedwith the caesarfinder using parameter values reported in Table A1 The backgroundrms noise was obtained by interpolating the median absolute devia-tion (MAD) of pixel fluxes computed over moving sampling boxesof size 10 times the area of the synthesized beam The backgroundlevel varies considerably across the surveyed area In Fig 2 wereport the estimated background noise in microJybeam as a functionof the Galactic latitude coordinate b averaged over the Galacticlongitude coordinate l We observe a noise level sim200 microJybeamin regions far from the Galactic plane and without bright sourcesClose to the Galactic plane the background noise increases due tothe Galactic diffuse emission and the bright emission from extendedsources filling the beam of the telescopes and increasing the sys-tem temperature In regions free of extended sources we observe abackground noise around 500-600 microJybeam Only 20 of the fieldarea has a 5σ noise level smaller than 1 mJybeam while forsim70of the field the 5σ noise is smaller than 2 mJybeam

22 ATCA 21 GHz observations and data reduction

The Scorpio field was observed with the Australia Telescope Com-pact Array (ATCA) in the 6A and 6D configurations at the referencefrequency of 21 GHz using the 16-cm CABB receiver (observingband from 11 to 31 GHz) (Wilson et al 2011) In this array config-uration the theoretical upper limits for the LAS ranges fromsim43 tosim122 arcmin The observations conducted in different runs from2011 to 2012 the data reduction strategy and the scientific resultsare extensively described elsewhere (Umana et al 2015 Riggi et al2016 Cavallaro et al 2018 Ingallinera et al 2019) ATCA observa-tions cover only a small portion of the Scorpio field observed withASKAP equivalent to 84 square degrees (see Fig1 for a compar-ison of the surveyed area size) In Fig C1 we present the ScorpioATCA mosaic We refer to this as the Scorpio ATCA map Theachieved rms is sim30-40 microJybeam and the synthesized beam inJ2000 coordinates is 98times58 (position angle of -3)The ATCA data obtained with a bandwidth of sim17 GHz weredivided into 7 sub-bands (〈νGHz〉=1449 1681 1844 20652337 2614 2895) and imaging was independently performed oneach of them to produce additional mosaics The 2895-GHz chan-nel map was not considered as it was significantly affected by noise

EMU Compact radio sources in SCORPIO 3

Figure 1 912 MHz mosaic of the Scorpio region observed with ASKAP The purple crosses indicate the beam centre positions for the three interleavedpointings The green contour delimits the mosaic area considered for source catalogue extraction with the caesar source finder The yellow contour denotesthe Scorpio region observed with the ATCA telescope at 21 GHz (see text)

and imaging artefacts The remaining sub-band mosaics (1-6) alongwith the full band mosaic are used throughout the paper as ancillarydata to complement the ASKAP catalogue with value-added infor-mation such as the source spectral indices (see Section 42) or toestimate the expected fraction of extended sources (see Section 41)

23 MOST 843 MHz observations

The Molonglo Galactic Plane Survey 2nd Epoch (MGPS-2) (Mur-phy et al 2007) carried out with the Molonglo Observatory Syn-thesis Telescope (MOST) at a frequency of 843 MHz completelycovers the Scorpio field observed with ASKAP with a lower spa-tial resolution (45times45 cosec|δ |) and a source detection thresholdof sim10 mJy 799 MGPS sources fall in the Scorpio region Theirposition uncertainty is considered better than 1-2 (Murphy et al2007)

24 NVSS 14 GHz observations

The NRAO VLA Sky Survey (NVSS) (Condon et al 1998) coversthe Scorpio region north of DEC=-40 at a frequency of 14 GHzwith an angular resolution of 45 The detection threshold is sim25mJy A number of 853 NVSS sources fall in the Scorpio region

25 TGSS 150 MHz observations

The TIFR GMRT Sky Survey (TGSS) (Intema et al 2017) fullycovers the Scorpio field at the reference frequency of 150 MHzand with an angular resolution of 25times25cos(DECminus 19) anda median rms noise of 35 mJybeam 249 sources from the firstalternative data release (ADR) fall in the Scorpio region

26 GLEAM 200 MHz observations

TheGaLactic and Extragalactic All-sky Murchison Widefield Array(GLEAM) survey (Hurley-Walker et al 2017) partially covers theScorpio field in (1 le |b| le10 345 lt l lt67) at the referencefrequency of 200MHz (bandwidth 60MHz) with an angular resolu-tion ofsim2 arcmin and an rms noise of 10-20 mJybeam 51 sourcesfrom the GLEAM Galactic plane catalogue (Hurley-Walker et al2019) fall in the Scorpio region

27 Supplementary surveys

In this work we will also make use of the following infrared surveysfor source classification studies (see Section 53)

bull AllWISE (Cutri et al 2013) of the Wide-field Infrared Survey

4 S Riggi et al

3minus 2minus 1minus 0 1 2 3

b (deg)0

200

400

600

800

1000

1200

1400

1600

Jyb

eam

)micro

(rm

Figure 2 Estimated background noise of Scorpio mosaic in microJybeam asa function of the Galactic latitude b and averaged over Galactic longitude l

Explorer (WISE Wright et al 2010) The survey is fully coveringthe Scorpio mosaic region with sim93times105 sources detected withSNgt5 in at least one of the four bands at 34 microm (W1) 46 microm(W2) 12 microm (W3) and 22 microm (W4) The angular resolutions are61 64 65 and 12 and the 5σ flux sensitivities for point sourcesare 008 mJy 011 mJy 1 mJy and 6 mJy respectivelybull GLIMPSE (Galactic Legacy Infrared MidPlane Survey Ex-

traordinaire) 80 microm surveys (Churchwell et al 2009) of the SpitzerSpace Telescope (Werner et al 2004) The surveys (GLIMPSE-I andGLIPMSE-3D) are partially covering (sim74) the Scorpio mosaicregion with 82times105 sources detected with SNgt5 The angularresolution is 2 and the 5σ flux sensitivity sim04 mJybull Hi-GAL (Herschel infrared Galactic plane Survey) 70 microm sur-

vey (Molinari et al 2016) of the Herschel Space Observatory (Pil-bratt et al 2010) The survey is partially covering (sim50) theScorpio mosaic region with 5654 sources detected with SNgt5The angular resolution is sim85 and the 1σ flux sensitivity sim20MJysr

3 COMPACT SOURCES IN THE SCORPIO FIELD

31 Source finding

Compact sources were extracted from the Scorpio ASKAP mapwith the caesar source finder (Riggi et al 2016 2019) using aniterative flood-fill algorithm in which the detection threshold isinitially set to 5σ and the flooding threshold to 25σ At each itera-tion the background and noise maps are recomputed excluding thesources found in the previous iteration and the detection thresholdis lowered in steps of 05σ Two iterations were used in this workcorresponding to a final 45σ seed threshold level Algorithm pa-rameter values reported in Table A1 resulted from a fine-tuningprocedure carried out both on the simulated data sample describedin Riggi et al (2019) and on the simulated maps described in Sec-tion 33We detected 5663 islands1 in the map 5413 of these were selected

1 By island (or blob) we denote a group of connected pixels with bright-ness above a merge threshold and around a seed pixel with brightness above

Table 1 Number of sources extracted from the Scorpio ASKAP mosaic at912MHzwith the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 250 0 01 4754 3857 37862 440 308 1733 131 59 4gt3 88 38 0

All 5663 4262 3963

as compact source candidates and successfully fitted with a mixtureof Gaussian componentsTo reject imaging artefacts and sources with a poor quality charac-terization we applied these selection criteria to the extracted sourcesample

bull Source fit converged with a χ2 lt10bull Positive fitted component peak fluxbull Fitted component centroid inside the source island and the

mosaic boundary regionbull Separation between any pair of source components larger than

8 (or 2 pixels)

After the selection 4262 source islands and 4813 fitted source com-ponents are left in the preliminary catalogueThese source counts still include spurious sources mainly due toimaging artefacts and over-deblending of extendeddiffuse emissionsurviving the selection criteria To produce the final catalogue wevisually inspected the entire field labelling each source componentas real or spurious in case of a clear unambiguous identifica-tion 4144 fitted source components were tagged as real in thefinal catalogue Taking into account that the selected survey regionis 377 square degrees a density ofsim110 selected compact sourcesper square degree is obtainedIn Table 1 we report a summary of the number of sources ex-tracted at different selection stages Source numbers labelled withSEL+VIS SEL refer to the number of sources passing both theselection criteria and the visual selection The number of detectedsources classified per number of fitted components is also reportedUsing the same procedure and source selection adopted for theASKAP catalogue we extracted 2227 sources and 2369 fitted com-ponents from the Scorpio ATCA mosaics The source numbersobtained at different selection stages on the full bandwidth map arereported in Table 2

32 Source cross-matching

To validate and complement the ASKAP catalogue we cross-matched it with the ATCA catalogue and with the catalogues ofsupplementary radio (MGPS NVSS TGSS GLEAM) and infrared

a detection threshold Island counts include also islands nested in otherislands

EMU Compact radio sources in SCORPIO 5

Table 2 Number of sources extracted from the Scorpio ATCA mosaic at21 GHz with the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 146 0 01 3021 2104 20962 465 188 1203 158 53 11gt3 102 30 0

All 3892 2375 2227

(AllWISE GLIMPSE Hi-GAL) surveys listed in Section 2 A sum-mary of the match results is reported in Table 3 and details areprovided in the following sections

321 Cross-matching with ATCA source catalogue

731 ASKAP source components (out of 856 components in thecatalogue falling in the ATCA mosaic region) match with at leastone source component from the ATCA catalogue within a searchradius of 24 We visually inspected each cross match rejectingspurious and unclear associations eg cases in which the ASKAPsource flux density measurement is potentially affected by closebackground sources visible in theATCAmap but not inASKAPWefinally selected 648 matches The majority are one-to-one matches(596) while the remaining (52) are one-to-many matches (up tothree components) The number ofmatches purely arising by chancewas estimated by averaging the number of matches found betweenASKAP catalogue and multiple random ATCA catalogues in whichthe measured source positions were uniformly randomized insidethe ATCA mosaic With this procedure we found 38plusmn03 matchesbetween both catalogues We thus concluded that sim995 percent ofthe matches found are likely to be real

322 Cross-matching with supplementary radio surveys

We found 688 MGPS sources potentially associated to ASKAPsources within a match radius of 45 546 of them were finally se-lected for further analysis after excludingmulti-matches and sourceswith potentially unreliable flux densities eg affected by imagingartefacts or closely located to very extended sourcesWithin a match radius of 45 we found 226 NVSS sources matchingto one or more ASKAP source components 189 single matcheswere finally selected after a visual inspection192 out of 217 TGSS source matches can be associated to a singleASKAP source within a match radius of 24 Similarly 32 out of 40GLEAM source matches found within a radius of 45 were selectedfor further analysis after excluding multi-matches or ambiguousassociations

323 Cross-matching with supplementary infrared surveys

For the cross-match we considered the AllWISE survey as the pri-mary dataset reference due to its full coverage To limit the number

Table 3 Number of cross-matches found between ASKAP source compo-nent catalogue and different radio and infrared surveys Column 4 reportsthe total number of sources falling in the Scorpio mosaic for each surveycatalogue In the last two rows we summed up the number of sources fromdifferent catalogues

Surveys radius () matches total

ATCA 24 648 2369MGPS 45 546 799NVSS 45 189 853TGSS 24 192 249GLEAM 45 32 51AllWISE 8 384 123374AllWISE + GLIMPSE 8 225 -AllWISE + GLIMPSE + Hi-GAL 8 41 -dagger

GLIMPSE catalogue has 820112 sources in Scorpiodagger Hi-GAL catalogue has 5654 sources in Scorpio

of spurious associations we selected sources with SNgt5 in boththe W3 and W4 bands with a fraction of saturated pixels smallerthan 10 reducing the number of sources within the ASKAP fieldto sim12times105 A crossmatch with ASKAP source catalogue usinga match radius equal to the ASKAP beam size (24) yields 1319matches About 2900 catalogued sources do not therefore have IRcounterparts with fluxes above W3 and W4 detection thresholdsThe number of source associations arising by chance estimatedusing artificial catalogues with random offsets applied is howevercompatible with the number of obtained matches indicating thatthe vast majority of the matches are spuriousSelecting a smaller radius (eg 8 as in the Galactic object searchanalysis) and requiring flux information in all bands leads to 384associations allowing us to reduce the number of potential spuriousmatches to sim40 The number of associations decreases to 225and 41 respectively if we require triple and quadruple matches withGLIMPSE and Hi-GAL cataloguesThe impact of the 5σ limits requested in W3 and W4 bands wasinvestigated on the stargalaxyquasar dataset provided by Clarke etal (2020)2 The magnitude selection cuts W3lt1132 andW4lt803

were found to remove a very large fraction (sim98) of extragalac-tic sources from the sample but unfortunately also potential radiostars We thus expect the unmatched sources to be mostly extra-galactic with a smaller percentage of IR-quiet (eg pulsars) orfaint mid-infrared Galactic objects

33 Catalogue selection effects and uncertainties

To evaluate the expected source extraction and characterization ac-curacy of the produced catalogue we made use of both simulateddata and cross-matches found with other surveysSimulated samples were drawn from the Scorpio mosaics (bothASKAP and ATCA) using the following approach4 First a resid-ual map was obtained by subtracting all compact sources down toa very low detection threshold (2σ ) Imaging artefacts extendedand diffuse sources were not removed A sample of 50 simulatedmaps was then generated by artificially adding uniformly spaced

2 105281zenodo37683983 See the WISE Explanatory Supplement at httpwise2ipac

caltechedudocsreleaseallskyexpsup4 Simulation tasks (compact source subtraction point-source generation)can be performed using application scripts provided along with caesar

6 S Riggi et al

0

02

04

06

08

1

com

plet

enes

s F

DR

1minus 05minus 0 05 1 15 2 25 3

(SmJy)10

log

Completenessaskapatca

FDRaskapaskap (sclass)atcaatca (sclass)

Figure 3 Completeness (filled markers) and false detection rate (FDR)(open markers) as a function of the injected and measured source flux den-sity respectively obtained on the simulatedASKAP (redmarkers) andATCA(green markers) source catalogues Solid lines represent the fitted complete-ness model of equation 1 Open squares and diamonds (labelled as sclass)represent the false detection rate obtained after applying a neural networkclassifier trained to identify spurious fit components in the catalogue

point sources to the residual mosaic with a density of 100 degminus2

and exponential flux density distribution (propeminusλS with λ=16 asfor real data above the detection threshold) caesar was finally runon the simulated maps and the same set of quality criteria usedfor the Scorpio mosaics were applied to the simulated catalogueSource extraction and characterization metrics were estimated bycross-matching injected sources with extracted sources

331 Completeness and reliability

Catalogue completeness and reliability were studied with simu-lated data as a function of the source flux density Sources injectedin the simulated maps are considered as the true catalogue Ex-tracted sources that do not crossmatch to any injected source arethus counted as false detections and contribute to increase the falsedetection rate (or decrease the reliability) Results are reported inFig 3 for both ASKAP (red markers) and ATCA (green markers)simulated data The completeness C is shown with filled markersand can be parametrized as a function of the flux density S as

C =C0

2

(1+ erf

log10(S)minus log10(S0)

k

)(1)

with parametersC0=099 log10(S0)=minus288 k=040 (ASKAP) andC0=099 log10(S0)=minus358 k=059 (ATCA) For ASKAP (ATCA)observations we expect the catalogue to be gt90 complete above5 mJy (1 mJy)The false detection rate reported in Fig 3 with open dots (ASKAP)and open triangles (ATCA) is found to be of the order of 20-30 slightly decreasing by a few percent when moving away fromthe Galactic plane False detections are largely due to sidelobesaround bright sources (compact or extended) and overdeblendingof extended sources Reliability can be improved by sim10 afterapplying a neural network classifier previously trained to identifygood sources from the data (see Riggi et al 2019 for more de-tails) Results are shown in Fig 3 with open squares (ASKAP) andopen diamonds (ATCA) To further reduce the false detection rate

to acceptable levels (below 1) we visually inspected the producedcatalogue removing spurious sources that were not identified in theautomated procedure (see Section 31)For the future we expect this can be substantially improved in mul-tiple ways eg increasing the number of classifier parameters tun-ing of classifier hyperparameters introducing specialized classifierstrained to identify imaging artefacts (eg around bright sources)which cannot be efficiently removed with the classifier adopted inthis work

332 Source position accuracy

Positional uncertainties in RA and Dec can be expressed as

σα =radic

σ2α f it +σ2

αcal

σδ =radic

σ2δ f it +σ2

δ cal

(2)

where σ f it are the position uncertainties due to the fitting process(including noise and uncertainties in fit parameters) while σcal arethe uncertainties due to the calibration processσ f it is provided by the fitting routine for each source and can becompared with values obtained from simulated data as a functionof the source signal-to-noise ratio (SN) We found that the fit un-certainties are largely underestimated with respect to the expectedvalues found in the simulations The latter are a factor 3-4 larger thanthe semi-analytical estimates provided by Condon (1997)5 This ispartly expected given that the semi-analytical estimate and the esti-mate obtained from the fitting routine are neglecting the effect of thecorrelated noise in the map A correction for this effect for exam-ple using equation 41 of Condon (1997) in the maximal-smoothinglimit would yield an expected analytical uncertainty sim26 largerWe have finally set σ f it to the values obtained from the simulationsand parametrized it as a function of SN as

σASKAPα f it =

986SN

arcsec σASKAPδ f it =

601SN

arcsec

σATCAα f it =

228SN

arcsec σATCAδ f it =

298SN

arcsec(3)

Position fit uncertainties are of the order of 15-18 (05-06) atthe detection threshold and better than 05 (015) with SNgt20 forASKAP (ATCA) dataCalibration uncertainties σcal can be inferred from the posi-tion spread observed in bright sources with respect to a refer-ence catalogue The Radio Fundamental Catalogue (RFC versionrfc_2020b)6 for instance containssim17000 radio sources measuredinmultiple VLBI observations withmilliarcsecond accuracy 5 RFCsources detected in the X band cross-match with ASKAP brightsources (four with SN gt200 and one with SN gt20) and haveposition uncertainty smaller than 002 The standard deviations(sα=05 sδ=04) of the observed position offset provide a firstmeasure of σcal since both RFC and ASKAP fit position uncertain-ties are negligible at high SN Additionally we inferred ASKAPcalibration errors using 344 and 80 sources with SNgt50 match-ing to MGPS and NVSS catalogues respectively The observedstandard deviations (sα=13-15 sδ=13-14) suggest a slightly

5 A discrepancy (a factor sim2) between analytical and measured uncertain-ties was observed also in other analysis carried out with simulated data usingcaesar (Riggi et al 2019) or alternative finders (Hopkins et al 2015)6 httpastrogeoorgrfc

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 3: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 3

Figure 1 912 MHz mosaic of the Scorpio region observed with ASKAP The purple crosses indicate the beam centre positions for the three interleavedpointings The green contour delimits the mosaic area considered for source catalogue extraction with the caesar source finder The yellow contour denotesthe Scorpio region observed with the ATCA telescope at 21 GHz (see text)

and imaging artefacts The remaining sub-band mosaics (1-6) alongwith the full band mosaic are used throughout the paper as ancillarydata to complement the ASKAP catalogue with value-added infor-mation such as the source spectral indices (see Section 42) or toestimate the expected fraction of extended sources (see Section 41)

23 MOST 843 MHz observations

The Molonglo Galactic Plane Survey 2nd Epoch (MGPS-2) (Mur-phy et al 2007) carried out with the Molonglo Observatory Syn-thesis Telescope (MOST) at a frequency of 843 MHz completelycovers the Scorpio field observed with ASKAP with a lower spa-tial resolution (45times45 cosec|δ |) and a source detection thresholdof sim10 mJy 799 MGPS sources fall in the Scorpio region Theirposition uncertainty is considered better than 1-2 (Murphy et al2007)

24 NVSS 14 GHz observations

The NRAO VLA Sky Survey (NVSS) (Condon et al 1998) coversthe Scorpio region north of DEC=-40 at a frequency of 14 GHzwith an angular resolution of 45 The detection threshold is sim25mJy A number of 853 NVSS sources fall in the Scorpio region

25 TGSS 150 MHz observations

The TIFR GMRT Sky Survey (TGSS) (Intema et al 2017) fullycovers the Scorpio field at the reference frequency of 150 MHzand with an angular resolution of 25times25cos(DECminus 19) anda median rms noise of 35 mJybeam 249 sources from the firstalternative data release (ADR) fall in the Scorpio region

26 GLEAM 200 MHz observations

TheGaLactic and Extragalactic All-sky Murchison Widefield Array(GLEAM) survey (Hurley-Walker et al 2017) partially covers theScorpio field in (1 le |b| le10 345 lt l lt67) at the referencefrequency of 200MHz (bandwidth 60MHz) with an angular resolu-tion ofsim2 arcmin and an rms noise of 10-20 mJybeam 51 sourcesfrom the GLEAM Galactic plane catalogue (Hurley-Walker et al2019) fall in the Scorpio region

27 Supplementary surveys

In this work we will also make use of the following infrared surveysfor source classification studies (see Section 53)

bull AllWISE (Cutri et al 2013) of the Wide-field Infrared Survey

4 S Riggi et al

3minus 2minus 1minus 0 1 2 3

b (deg)0

200

400

600

800

1000

1200

1400

1600

Jyb

eam

)micro

(rm

Figure 2 Estimated background noise of Scorpio mosaic in microJybeam asa function of the Galactic latitude b and averaged over Galactic longitude l

Explorer (WISE Wright et al 2010) The survey is fully coveringthe Scorpio mosaic region with sim93times105 sources detected withSNgt5 in at least one of the four bands at 34 microm (W1) 46 microm(W2) 12 microm (W3) and 22 microm (W4) The angular resolutions are61 64 65 and 12 and the 5σ flux sensitivities for point sourcesare 008 mJy 011 mJy 1 mJy and 6 mJy respectivelybull GLIMPSE (Galactic Legacy Infrared MidPlane Survey Ex-

traordinaire) 80 microm surveys (Churchwell et al 2009) of the SpitzerSpace Telescope (Werner et al 2004) The surveys (GLIMPSE-I andGLIPMSE-3D) are partially covering (sim74) the Scorpio mosaicregion with 82times105 sources detected with SNgt5 The angularresolution is 2 and the 5σ flux sensitivity sim04 mJybull Hi-GAL (Herschel infrared Galactic plane Survey) 70 microm sur-

vey (Molinari et al 2016) of the Herschel Space Observatory (Pil-bratt et al 2010) The survey is partially covering (sim50) theScorpio mosaic region with 5654 sources detected with SNgt5The angular resolution is sim85 and the 1σ flux sensitivity sim20MJysr

3 COMPACT SOURCES IN THE SCORPIO FIELD

31 Source finding

Compact sources were extracted from the Scorpio ASKAP mapwith the caesar source finder (Riggi et al 2016 2019) using aniterative flood-fill algorithm in which the detection threshold isinitially set to 5σ and the flooding threshold to 25σ At each itera-tion the background and noise maps are recomputed excluding thesources found in the previous iteration and the detection thresholdis lowered in steps of 05σ Two iterations were used in this workcorresponding to a final 45σ seed threshold level Algorithm pa-rameter values reported in Table A1 resulted from a fine-tuningprocedure carried out both on the simulated data sample describedin Riggi et al (2019) and on the simulated maps described in Sec-tion 33We detected 5663 islands1 in the map 5413 of these were selected

1 By island (or blob) we denote a group of connected pixels with bright-ness above a merge threshold and around a seed pixel with brightness above

Table 1 Number of sources extracted from the Scorpio ASKAP mosaic at912MHzwith the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 250 0 01 4754 3857 37862 440 308 1733 131 59 4gt3 88 38 0

All 5663 4262 3963

as compact source candidates and successfully fitted with a mixtureof Gaussian componentsTo reject imaging artefacts and sources with a poor quality charac-terization we applied these selection criteria to the extracted sourcesample

bull Source fit converged with a χ2 lt10bull Positive fitted component peak fluxbull Fitted component centroid inside the source island and the

mosaic boundary regionbull Separation between any pair of source components larger than

8 (or 2 pixels)

After the selection 4262 source islands and 4813 fitted source com-ponents are left in the preliminary catalogueThese source counts still include spurious sources mainly due toimaging artefacts and over-deblending of extendeddiffuse emissionsurviving the selection criteria To produce the final catalogue wevisually inspected the entire field labelling each source componentas real or spurious in case of a clear unambiguous identifica-tion 4144 fitted source components were tagged as real in thefinal catalogue Taking into account that the selected survey regionis 377 square degrees a density ofsim110 selected compact sourcesper square degree is obtainedIn Table 1 we report a summary of the number of sources ex-tracted at different selection stages Source numbers labelled withSEL+VIS SEL refer to the number of sources passing both theselection criteria and the visual selection The number of detectedsources classified per number of fitted components is also reportedUsing the same procedure and source selection adopted for theASKAP catalogue we extracted 2227 sources and 2369 fitted com-ponents from the Scorpio ATCA mosaics The source numbersobtained at different selection stages on the full bandwidth map arereported in Table 2

32 Source cross-matching

To validate and complement the ASKAP catalogue we cross-matched it with the ATCA catalogue and with the catalogues ofsupplementary radio (MGPS NVSS TGSS GLEAM) and infrared

a detection threshold Island counts include also islands nested in otherislands

EMU Compact radio sources in SCORPIO 5

Table 2 Number of sources extracted from the Scorpio ATCA mosaic at21 GHz with the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 146 0 01 3021 2104 20962 465 188 1203 158 53 11gt3 102 30 0

All 3892 2375 2227

(AllWISE GLIMPSE Hi-GAL) surveys listed in Section 2 A sum-mary of the match results is reported in Table 3 and details areprovided in the following sections

321 Cross-matching with ATCA source catalogue

731 ASKAP source components (out of 856 components in thecatalogue falling in the ATCA mosaic region) match with at leastone source component from the ATCA catalogue within a searchradius of 24 We visually inspected each cross match rejectingspurious and unclear associations eg cases in which the ASKAPsource flux density measurement is potentially affected by closebackground sources visible in theATCAmap but not inASKAPWefinally selected 648 matches The majority are one-to-one matches(596) while the remaining (52) are one-to-many matches (up tothree components) The number ofmatches purely arising by chancewas estimated by averaging the number of matches found betweenASKAP catalogue and multiple random ATCA catalogues in whichthe measured source positions were uniformly randomized insidethe ATCA mosaic With this procedure we found 38plusmn03 matchesbetween both catalogues We thus concluded that sim995 percent ofthe matches found are likely to be real

322 Cross-matching with supplementary radio surveys

We found 688 MGPS sources potentially associated to ASKAPsources within a match radius of 45 546 of them were finally se-lected for further analysis after excludingmulti-matches and sourceswith potentially unreliable flux densities eg affected by imagingartefacts or closely located to very extended sourcesWithin a match radius of 45 we found 226 NVSS sources matchingto one or more ASKAP source components 189 single matcheswere finally selected after a visual inspection192 out of 217 TGSS source matches can be associated to a singleASKAP source within a match radius of 24 Similarly 32 out of 40GLEAM source matches found within a radius of 45 were selectedfor further analysis after excluding multi-matches or ambiguousassociations

323 Cross-matching with supplementary infrared surveys

For the cross-match we considered the AllWISE survey as the pri-mary dataset reference due to its full coverage To limit the number

Table 3 Number of cross-matches found between ASKAP source compo-nent catalogue and different radio and infrared surveys Column 4 reportsthe total number of sources falling in the Scorpio mosaic for each surveycatalogue In the last two rows we summed up the number of sources fromdifferent catalogues

Surveys radius () matches total

ATCA 24 648 2369MGPS 45 546 799NVSS 45 189 853TGSS 24 192 249GLEAM 45 32 51AllWISE 8 384 123374AllWISE + GLIMPSE 8 225 -AllWISE + GLIMPSE + Hi-GAL 8 41 -dagger

GLIMPSE catalogue has 820112 sources in Scorpiodagger Hi-GAL catalogue has 5654 sources in Scorpio

of spurious associations we selected sources with SNgt5 in boththe W3 and W4 bands with a fraction of saturated pixels smallerthan 10 reducing the number of sources within the ASKAP fieldto sim12times105 A crossmatch with ASKAP source catalogue usinga match radius equal to the ASKAP beam size (24) yields 1319matches About 2900 catalogued sources do not therefore have IRcounterparts with fluxes above W3 and W4 detection thresholdsThe number of source associations arising by chance estimatedusing artificial catalogues with random offsets applied is howevercompatible with the number of obtained matches indicating thatthe vast majority of the matches are spuriousSelecting a smaller radius (eg 8 as in the Galactic object searchanalysis) and requiring flux information in all bands leads to 384associations allowing us to reduce the number of potential spuriousmatches to sim40 The number of associations decreases to 225and 41 respectively if we require triple and quadruple matches withGLIMPSE and Hi-GAL cataloguesThe impact of the 5σ limits requested in W3 and W4 bands wasinvestigated on the stargalaxyquasar dataset provided by Clarke etal (2020)2 The magnitude selection cuts W3lt1132 andW4lt803

were found to remove a very large fraction (sim98) of extragalac-tic sources from the sample but unfortunately also potential radiostars We thus expect the unmatched sources to be mostly extra-galactic with a smaller percentage of IR-quiet (eg pulsars) orfaint mid-infrared Galactic objects

33 Catalogue selection effects and uncertainties

To evaluate the expected source extraction and characterization ac-curacy of the produced catalogue we made use of both simulateddata and cross-matches found with other surveysSimulated samples were drawn from the Scorpio mosaics (bothASKAP and ATCA) using the following approach4 First a resid-ual map was obtained by subtracting all compact sources down toa very low detection threshold (2σ ) Imaging artefacts extendedand diffuse sources were not removed A sample of 50 simulatedmaps was then generated by artificially adding uniformly spaced

2 105281zenodo37683983 See the WISE Explanatory Supplement at httpwise2ipac

caltechedudocsreleaseallskyexpsup4 Simulation tasks (compact source subtraction point-source generation)can be performed using application scripts provided along with caesar

6 S Riggi et al

0

02

04

06

08

1

com

plet

enes

s F

DR

1minus 05minus 0 05 1 15 2 25 3

(SmJy)10

log

Completenessaskapatca

FDRaskapaskap (sclass)atcaatca (sclass)

Figure 3 Completeness (filled markers) and false detection rate (FDR)(open markers) as a function of the injected and measured source flux den-sity respectively obtained on the simulatedASKAP (redmarkers) andATCA(green markers) source catalogues Solid lines represent the fitted complete-ness model of equation 1 Open squares and diamonds (labelled as sclass)represent the false detection rate obtained after applying a neural networkclassifier trained to identify spurious fit components in the catalogue

point sources to the residual mosaic with a density of 100 degminus2

and exponential flux density distribution (propeminusλS with λ=16 asfor real data above the detection threshold) caesar was finally runon the simulated maps and the same set of quality criteria usedfor the Scorpio mosaics were applied to the simulated catalogueSource extraction and characterization metrics were estimated bycross-matching injected sources with extracted sources

331 Completeness and reliability

Catalogue completeness and reliability were studied with simu-lated data as a function of the source flux density Sources injectedin the simulated maps are considered as the true catalogue Ex-tracted sources that do not crossmatch to any injected source arethus counted as false detections and contribute to increase the falsedetection rate (or decrease the reliability) Results are reported inFig 3 for both ASKAP (red markers) and ATCA (green markers)simulated data The completeness C is shown with filled markersand can be parametrized as a function of the flux density S as

C =C0

2

(1+ erf

log10(S)minus log10(S0)

k

)(1)

with parametersC0=099 log10(S0)=minus288 k=040 (ASKAP) andC0=099 log10(S0)=minus358 k=059 (ATCA) For ASKAP (ATCA)observations we expect the catalogue to be gt90 complete above5 mJy (1 mJy)The false detection rate reported in Fig 3 with open dots (ASKAP)and open triangles (ATCA) is found to be of the order of 20-30 slightly decreasing by a few percent when moving away fromthe Galactic plane False detections are largely due to sidelobesaround bright sources (compact or extended) and overdeblendingof extended sources Reliability can be improved by sim10 afterapplying a neural network classifier previously trained to identifygood sources from the data (see Riggi et al 2019 for more de-tails) Results are shown in Fig 3 with open squares (ASKAP) andopen diamonds (ATCA) To further reduce the false detection rate

to acceptable levels (below 1) we visually inspected the producedcatalogue removing spurious sources that were not identified in theautomated procedure (see Section 31)For the future we expect this can be substantially improved in mul-tiple ways eg increasing the number of classifier parameters tun-ing of classifier hyperparameters introducing specialized classifierstrained to identify imaging artefacts (eg around bright sources)which cannot be efficiently removed with the classifier adopted inthis work

332 Source position accuracy

Positional uncertainties in RA and Dec can be expressed as

σα =radic

σ2α f it +σ2

αcal

σδ =radic

σ2δ f it +σ2

δ cal

(2)

where σ f it are the position uncertainties due to the fitting process(including noise and uncertainties in fit parameters) while σcal arethe uncertainties due to the calibration processσ f it is provided by the fitting routine for each source and can becompared with values obtained from simulated data as a functionof the source signal-to-noise ratio (SN) We found that the fit un-certainties are largely underestimated with respect to the expectedvalues found in the simulations The latter are a factor 3-4 larger thanthe semi-analytical estimates provided by Condon (1997)5 This ispartly expected given that the semi-analytical estimate and the esti-mate obtained from the fitting routine are neglecting the effect of thecorrelated noise in the map A correction for this effect for exam-ple using equation 41 of Condon (1997) in the maximal-smoothinglimit would yield an expected analytical uncertainty sim26 largerWe have finally set σ f it to the values obtained from the simulationsand parametrized it as a function of SN as

σASKAPα f it =

986SN

arcsec σASKAPδ f it =

601SN

arcsec

σATCAα f it =

228SN

arcsec σATCAδ f it =

298SN

arcsec(3)

Position fit uncertainties are of the order of 15-18 (05-06) atthe detection threshold and better than 05 (015) with SNgt20 forASKAP (ATCA) dataCalibration uncertainties σcal can be inferred from the posi-tion spread observed in bright sources with respect to a refer-ence catalogue The Radio Fundamental Catalogue (RFC versionrfc_2020b)6 for instance containssim17000 radio sources measuredinmultiple VLBI observations withmilliarcsecond accuracy 5 RFCsources detected in the X band cross-match with ASKAP brightsources (four with SN gt200 and one with SN gt20) and haveposition uncertainty smaller than 002 The standard deviations(sα=05 sδ=04) of the observed position offset provide a firstmeasure of σcal since both RFC and ASKAP fit position uncertain-ties are negligible at high SN Additionally we inferred ASKAPcalibration errors using 344 and 80 sources with SNgt50 match-ing to MGPS and NVSS catalogues respectively The observedstandard deviations (sα=13-15 sδ=13-14) suggest a slightly

5 A discrepancy (a factor sim2) between analytical and measured uncertain-ties was observed also in other analysis carried out with simulated data usingcaesar (Riggi et al 2019) or alternative finders (Hopkins et al 2015)6 httpastrogeoorgrfc

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 4: EvolutionaryMapoftheUniverse(EMU

4 S Riggi et al

3minus 2minus 1minus 0 1 2 3

b (deg)0

200

400

600

800

1000

1200

1400

1600

Jyb

eam

)micro

(rm

Figure 2 Estimated background noise of Scorpio mosaic in microJybeam asa function of the Galactic latitude b and averaged over Galactic longitude l

Explorer (WISE Wright et al 2010) The survey is fully coveringthe Scorpio mosaic region with sim93times105 sources detected withSNgt5 in at least one of the four bands at 34 microm (W1) 46 microm(W2) 12 microm (W3) and 22 microm (W4) The angular resolutions are61 64 65 and 12 and the 5σ flux sensitivities for point sourcesare 008 mJy 011 mJy 1 mJy and 6 mJy respectivelybull GLIMPSE (Galactic Legacy Infrared MidPlane Survey Ex-

traordinaire) 80 microm surveys (Churchwell et al 2009) of the SpitzerSpace Telescope (Werner et al 2004) The surveys (GLIMPSE-I andGLIPMSE-3D) are partially covering (sim74) the Scorpio mosaicregion with 82times105 sources detected with SNgt5 The angularresolution is 2 and the 5σ flux sensitivity sim04 mJybull Hi-GAL (Herschel infrared Galactic plane Survey) 70 microm sur-

vey (Molinari et al 2016) of the Herschel Space Observatory (Pil-bratt et al 2010) The survey is partially covering (sim50) theScorpio mosaic region with 5654 sources detected with SNgt5The angular resolution is sim85 and the 1σ flux sensitivity sim20MJysr

3 COMPACT SOURCES IN THE SCORPIO FIELD

31 Source finding

Compact sources were extracted from the Scorpio ASKAP mapwith the caesar source finder (Riggi et al 2016 2019) using aniterative flood-fill algorithm in which the detection threshold isinitially set to 5σ and the flooding threshold to 25σ At each itera-tion the background and noise maps are recomputed excluding thesources found in the previous iteration and the detection thresholdis lowered in steps of 05σ Two iterations were used in this workcorresponding to a final 45σ seed threshold level Algorithm pa-rameter values reported in Table A1 resulted from a fine-tuningprocedure carried out both on the simulated data sample describedin Riggi et al (2019) and on the simulated maps described in Sec-tion 33We detected 5663 islands1 in the map 5413 of these were selected

1 By island (or blob) we denote a group of connected pixels with bright-ness above a merge threshold and around a seed pixel with brightness above

Table 1 Number of sources extracted from the Scorpio ASKAP mosaic at912MHzwith the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 250 0 01 4754 3857 37862 440 308 1733 131 59 4gt3 88 38 0

All 5663 4262 3963

as compact source candidates and successfully fitted with a mixtureof Gaussian componentsTo reject imaging artefacts and sources with a poor quality charac-terization we applied these selection criteria to the extracted sourcesample

bull Source fit converged with a χ2 lt10bull Positive fitted component peak fluxbull Fitted component centroid inside the source island and the

mosaic boundary regionbull Separation between any pair of source components larger than

8 (or 2 pixels)

After the selection 4262 source islands and 4813 fitted source com-ponents are left in the preliminary catalogueThese source counts still include spurious sources mainly due toimaging artefacts and over-deblending of extendeddiffuse emissionsurviving the selection criteria To produce the final catalogue wevisually inspected the entire field labelling each source componentas real or spurious in case of a clear unambiguous identifica-tion 4144 fitted source components were tagged as real in thefinal catalogue Taking into account that the selected survey regionis 377 square degrees a density ofsim110 selected compact sourcesper square degree is obtainedIn Table 1 we report a summary of the number of sources ex-tracted at different selection stages Source numbers labelled withSEL+VIS SEL refer to the number of sources passing both theselection criteria and the visual selection The number of detectedsources classified per number of fitted components is also reportedUsing the same procedure and source selection adopted for theASKAP catalogue we extracted 2227 sources and 2369 fitted com-ponents from the Scorpio ATCA mosaics The source numbersobtained at different selection stages on the full bandwidth map arereported in Table 2

32 Source cross-matching

To validate and complement the ASKAP catalogue we cross-matched it with the ATCA catalogue and with the catalogues ofsupplementary radio (MGPS NVSS TGSS GLEAM) and infrared

a detection threshold Island counts include also islands nested in otherislands

EMU Compact radio sources in SCORPIO 5

Table 2 Number of sources extracted from the Scorpio ATCA mosaic at21 GHz with the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 146 0 01 3021 2104 20962 465 188 1203 158 53 11gt3 102 30 0

All 3892 2375 2227

(AllWISE GLIMPSE Hi-GAL) surveys listed in Section 2 A sum-mary of the match results is reported in Table 3 and details areprovided in the following sections

321 Cross-matching with ATCA source catalogue

731 ASKAP source components (out of 856 components in thecatalogue falling in the ATCA mosaic region) match with at leastone source component from the ATCA catalogue within a searchradius of 24 We visually inspected each cross match rejectingspurious and unclear associations eg cases in which the ASKAPsource flux density measurement is potentially affected by closebackground sources visible in theATCAmap but not inASKAPWefinally selected 648 matches The majority are one-to-one matches(596) while the remaining (52) are one-to-many matches (up tothree components) The number ofmatches purely arising by chancewas estimated by averaging the number of matches found betweenASKAP catalogue and multiple random ATCA catalogues in whichthe measured source positions were uniformly randomized insidethe ATCA mosaic With this procedure we found 38plusmn03 matchesbetween both catalogues We thus concluded that sim995 percent ofthe matches found are likely to be real

322 Cross-matching with supplementary radio surveys

We found 688 MGPS sources potentially associated to ASKAPsources within a match radius of 45 546 of them were finally se-lected for further analysis after excludingmulti-matches and sourceswith potentially unreliable flux densities eg affected by imagingartefacts or closely located to very extended sourcesWithin a match radius of 45 we found 226 NVSS sources matchingto one or more ASKAP source components 189 single matcheswere finally selected after a visual inspection192 out of 217 TGSS source matches can be associated to a singleASKAP source within a match radius of 24 Similarly 32 out of 40GLEAM source matches found within a radius of 45 were selectedfor further analysis after excluding multi-matches or ambiguousassociations

323 Cross-matching with supplementary infrared surveys

For the cross-match we considered the AllWISE survey as the pri-mary dataset reference due to its full coverage To limit the number

Table 3 Number of cross-matches found between ASKAP source compo-nent catalogue and different radio and infrared surveys Column 4 reportsthe total number of sources falling in the Scorpio mosaic for each surveycatalogue In the last two rows we summed up the number of sources fromdifferent catalogues

Surveys radius () matches total

ATCA 24 648 2369MGPS 45 546 799NVSS 45 189 853TGSS 24 192 249GLEAM 45 32 51AllWISE 8 384 123374AllWISE + GLIMPSE 8 225 -AllWISE + GLIMPSE + Hi-GAL 8 41 -dagger

GLIMPSE catalogue has 820112 sources in Scorpiodagger Hi-GAL catalogue has 5654 sources in Scorpio

of spurious associations we selected sources with SNgt5 in boththe W3 and W4 bands with a fraction of saturated pixels smallerthan 10 reducing the number of sources within the ASKAP fieldto sim12times105 A crossmatch with ASKAP source catalogue usinga match radius equal to the ASKAP beam size (24) yields 1319matches About 2900 catalogued sources do not therefore have IRcounterparts with fluxes above W3 and W4 detection thresholdsThe number of source associations arising by chance estimatedusing artificial catalogues with random offsets applied is howevercompatible with the number of obtained matches indicating thatthe vast majority of the matches are spuriousSelecting a smaller radius (eg 8 as in the Galactic object searchanalysis) and requiring flux information in all bands leads to 384associations allowing us to reduce the number of potential spuriousmatches to sim40 The number of associations decreases to 225and 41 respectively if we require triple and quadruple matches withGLIMPSE and Hi-GAL cataloguesThe impact of the 5σ limits requested in W3 and W4 bands wasinvestigated on the stargalaxyquasar dataset provided by Clarke etal (2020)2 The magnitude selection cuts W3lt1132 andW4lt803

were found to remove a very large fraction (sim98) of extragalac-tic sources from the sample but unfortunately also potential radiostars We thus expect the unmatched sources to be mostly extra-galactic with a smaller percentage of IR-quiet (eg pulsars) orfaint mid-infrared Galactic objects

33 Catalogue selection effects and uncertainties

To evaluate the expected source extraction and characterization ac-curacy of the produced catalogue we made use of both simulateddata and cross-matches found with other surveysSimulated samples were drawn from the Scorpio mosaics (bothASKAP and ATCA) using the following approach4 First a resid-ual map was obtained by subtracting all compact sources down toa very low detection threshold (2σ ) Imaging artefacts extendedand diffuse sources were not removed A sample of 50 simulatedmaps was then generated by artificially adding uniformly spaced

2 105281zenodo37683983 See the WISE Explanatory Supplement at httpwise2ipac

caltechedudocsreleaseallskyexpsup4 Simulation tasks (compact source subtraction point-source generation)can be performed using application scripts provided along with caesar

6 S Riggi et al

0

02

04

06

08

1

com

plet

enes

s F

DR

1minus 05minus 0 05 1 15 2 25 3

(SmJy)10

log

Completenessaskapatca

FDRaskapaskap (sclass)atcaatca (sclass)

Figure 3 Completeness (filled markers) and false detection rate (FDR)(open markers) as a function of the injected and measured source flux den-sity respectively obtained on the simulatedASKAP (redmarkers) andATCA(green markers) source catalogues Solid lines represent the fitted complete-ness model of equation 1 Open squares and diamonds (labelled as sclass)represent the false detection rate obtained after applying a neural networkclassifier trained to identify spurious fit components in the catalogue

point sources to the residual mosaic with a density of 100 degminus2

and exponential flux density distribution (propeminusλS with λ=16 asfor real data above the detection threshold) caesar was finally runon the simulated maps and the same set of quality criteria usedfor the Scorpio mosaics were applied to the simulated catalogueSource extraction and characterization metrics were estimated bycross-matching injected sources with extracted sources

331 Completeness and reliability

Catalogue completeness and reliability were studied with simu-lated data as a function of the source flux density Sources injectedin the simulated maps are considered as the true catalogue Ex-tracted sources that do not crossmatch to any injected source arethus counted as false detections and contribute to increase the falsedetection rate (or decrease the reliability) Results are reported inFig 3 for both ASKAP (red markers) and ATCA (green markers)simulated data The completeness C is shown with filled markersand can be parametrized as a function of the flux density S as

C =C0

2

(1+ erf

log10(S)minus log10(S0)

k

)(1)

with parametersC0=099 log10(S0)=minus288 k=040 (ASKAP) andC0=099 log10(S0)=minus358 k=059 (ATCA) For ASKAP (ATCA)observations we expect the catalogue to be gt90 complete above5 mJy (1 mJy)The false detection rate reported in Fig 3 with open dots (ASKAP)and open triangles (ATCA) is found to be of the order of 20-30 slightly decreasing by a few percent when moving away fromthe Galactic plane False detections are largely due to sidelobesaround bright sources (compact or extended) and overdeblendingof extended sources Reliability can be improved by sim10 afterapplying a neural network classifier previously trained to identifygood sources from the data (see Riggi et al 2019 for more de-tails) Results are shown in Fig 3 with open squares (ASKAP) andopen diamonds (ATCA) To further reduce the false detection rate

to acceptable levels (below 1) we visually inspected the producedcatalogue removing spurious sources that were not identified in theautomated procedure (see Section 31)For the future we expect this can be substantially improved in mul-tiple ways eg increasing the number of classifier parameters tun-ing of classifier hyperparameters introducing specialized classifierstrained to identify imaging artefacts (eg around bright sources)which cannot be efficiently removed with the classifier adopted inthis work

332 Source position accuracy

Positional uncertainties in RA and Dec can be expressed as

σα =radic

σ2α f it +σ2

αcal

σδ =radic

σ2δ f it +σ2

δ cal

(2)

where σ f it are the position uncertainties due to the fitting process(including noise and uncertainties in fit parameters) while σcal arethe uncertainties due to the calibration processσ f it is provided by the fitting routine for each source and can becompared with values obtained from simulated data as a functionof the source signal-to-noise ratio (SN) We found that the fit un-certainties are largely underestimated with respect to the expectedvalues found in the simulations The latter are a factor 3-4 larger thanthe semi-analytical estimates provided by Condon (1997)5 This ispartly expected given that the semi-analytical estimate and the esti-mate obtained from the fitting routine are neglecting the effect of thecorrelated noise in the map A correction for this effect for exam-ple using equation 41 of Condon (1997) in the maximal-smoothinglimit would yield an expected analytical uncertainty sim26 largerWe have finally set σ f it to the values obtained from the simulationsand parametrized it as a function of SN as

σASKAPα f it =

986SN

arcsec σASKAPδ f it =

601SN

arcsec

σATCAα f it =

228SN

arcsec σATCAδ f it =

298SN

arcsec(3)

Position fit uncertainties are of the order of 15-18 (05-06) atthe detection threshold and better than 05 (015) with SNgt20 forASKAP (ATCA) dataCalibration uncertainties σcal can be inferred from the posi-tion spread observed in bright sources with respect to a refer-ence catalogue The Radio Fundamental Catalogue (RFC versionrfc_2020b)6 for instance containssim17000 radio sources measuredinmultiple VLBI observations withmilliarcsecond accuracy 5 RFCsources detected in the X band cross-match with ASKAP brightsources (four with SN gt200 and one with SN gt20) and haveposition uncertainty smaller than 002 The standard deviations(sα=05 sδ=04) of the observed position offset provide a firstmeasure of σcal since both RFC and ASKAP fit position uncertain-ties are negligible at high SN Additionally we inferred ASKAPcalibration errors using 344 and 80 sources with SNgt50 match-ing to MGPS and NVSS catalogues respectively The observedstandard deviations (sα=13-15 sδ=13-14) suggest a slightly

5 A discrepancy (a factor sim2) between analytical and measured uncertain-ties was observed also in other analysis carried out with simulated data usingcaesar (Riggi et al 2019) or alternative finders (Hopkins et al 2015)6 httpastrogeoorgrfc

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 5: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 5

Table 2 Number of sources extracted from the Scorpio ATCA mosaic at21 GHz with the caesar finder at different selection stages and per numberof fitted components The no sel column reports the number of sourcesextracted by the source finder without any quality cuts applied The selcolumn reports the number of sources obtained after applying the selectioncriteria described in the text The sel+vis sel column reports the numberof sources passing both the quality selection and the visual inspection

components Selection

no sel sel sel+vis sel

0 146 0 01 3021 2104 20962 465 188 1203 158 53 11gt3 102 30 0

All 3892 2375 2227

(AllWISE GLIMPSE Hi-GAL) surveys listed in Section 2 A sum-mary of the match results is reported in Table 3 and details areprovided in the following sections

321 Cross-matching with ATCA source catalogue

731 ASKAP source components (out of 856 components in thecatalogue falling in the ATCA mosaic region) match with at leastone source component from the ATCA catalogue within a searchradius of 24 We visually inspected each cross match rejectingspurious and unclear associations eg cases in which the ASKAPsource flux density measurement is potentially affected by closebackground sources visible in theATCAmap but not inASKAPWefinally selected 648 matches The majority are one-to-one matches(596) while the remaining (52) are one-to-many matches (up tothree components) The number ofmatches purely arising by chancewas estimated by averaging the number of matches found betweenASKAP catalogue and multiple random ATCA catalogues in whichthe measured source positions were uniformly randomized insidethe ATCA mosaic With this procedure we found 38plusmn03 matchesbetween both catalogues We thus concluded that sim995 percent ofthe matches found are likely to be real

322 Cross-matching with supplementary radio surveys

We found 688 MGPS sources potentially associated to ASKAPsources within a match radius of 45 546 of them were finally se-lected for further analysis after excludingmulti-matches and sourceswith potentially unreliable flux densities eg affected by imagingartefacts or closely located to very extended sourcesWithin a match radius of 45 we found 226 NVSS sources matchingto one or more ASKAP source components 189 single matcheswere finally selected after a visual inspection192 out of 217 TGSS source matches can be associated to a singleASKAP source within a match radius of 24 Similarly 32 out of 40GLEAM source matches found within a radius of 45 were selectedfor further analysis after excluding multi-matches or ambiguousassociations

323 Cross-matching with supplementary infrared surveys

For the cross-match we considered the AllWISE survey as the pri-mary dataset reference due to its full coverage To limit the number

Table 3 Number of cross-matches found between ASKAP source compo-nent catalogue and different radio and infrared surveys Column 4 reportsthe total number of sources falling in the Scorpio mosaic for each surveycatalogue In the last two rows we summed up the number of sources fromdifferent catalogues

Surveys radius () matches total

ATCA 24 648 2369MGPS 45 546 799NVSS 45 189 853TGSS 24 192 249GLEAM 45 32 51AllWISE 8 384 123374AllWISE + GLIMPSE 8 225 -AllWISE + GLIMPSE + Hi-GAL 8 41 -dagger

GLIMPSE catalogue has 820112 sources in Scorpiodagger Hi-GAL catalogue has 5654 sources in Scorpio

of spurious associations we selected sources with SNgt5 in boththe W3 and W4 bands with a fraction of saturated pixels smallerthan 10 reducing the number of sources within the ASKAP fieldto sim12times105 A crossmatch with ASKAP source catalogue usinga match radius equal to the ASKAP beam size (24) yields 1319matches About 2900 catalogued sources do not therefore have IRcounterparts with fluxes above W3 and W4 detection thresholdsThe number of source associations arising by chance estimatedusing artificial catalogues with random offsets applied is howevercompatible with the number of obtained matches indicating thatthe vast majority of the matches are spuriousSelecting a smaller radius (eg 8 as in the Galactic object searchanalysis) and requiring flux information in all bands leads to 384associations allowing us to reduce the number of potential spuriousmatches to sim40 The number of associations decreases to 225and 41 respectively if we require triple and quadruple matches withGLIMPSE and Hi-GAL cataloguesThe impact of the 5σ limits requested in W3 and W4 bands wasinvestigated on the stargalaxyquasar dataset provided by Clarke etal (2020)2 The magnitude selection cuts W3lt1132 andW4lt803

were found to remove a very large fraction (sim98) of extragalac-tic sources from the sample but unfortunately also potential radiostars We thus expect the unmatched sources to be mostly extra-galactic with a smaller percentage of IR-quiet (eg pulsars) orfaint mid-infrared Galactic objects

33 Catalogue selection effects and uncertainties

To evaluate the expected source extraction and characterization ac-curacy of the produced catalogue we made use of both simulateddata and cross-matches found with other surveysSimulated samples were drawn from the Scorpio mosaics (bothASKAP and ATCA) using the following approach4 First a resid-ual map was obtained by subtracting all compact sources down toa very low detection threshold (2σ ) Imaging artefacts extendedand diffuse sources were not removed A sample of 50 simulatedmaps was then generated by artificially adding uniformly spaced

2 105281zenodo37683983 See the WISE Explanatory Supplement at httpwise2ipac

caltechedudocsreleaseallskyexpsup4 Simulation tasks (compact source subtraction point-source generation)can be performed using application scripts provided along with caesar

6 S Riggi et al

0

02

04

06

08

1

com

plet

enes

s F

DR

1minus 05minus 0 05 1 15 2 25 3

(SmJy)10

log

Completenessaskapatca

FDRaskapaskap (sclass)atcaatca (sclass)

Figure 3 Completeness (filled markers) and false detection rate (FDR)(open markers) as a function of the injected and measured source flux den-sity respectively obtained on the simulatedASKAP (redmarkers) andATCA(green markers) source catalogues Solid lines represent the fitted complete-ness model of equation 1 Open squares and diamonds (labelled as sclass)represent the false detection rate obtained after applying a neural networkclassifier trained to identify spurious fit components in the catalogue

point sources to the residual mosaic with a density of 100 degminus2

and exponential flux density distribution (propeminusλS with λ=16 asfor real data above the detection threshold) caesar was finally runon the simulated maps and the same set of quality criteria usedfor the Scorpio mosaics were applied to the simulated catalogueSource extraction and characterization metrics were estimated bycross-matching injected sources with extracted sources

331 Completeness and reliability

Catalogue completeness and reliability were studied with simu-lated data as a function of the source flux density Sources injectedin the simulated maps are considered as the true catalogue Ex-tracted sources that do not crossmatch to any injected source arethus counted as false detections and contribute to increase the falsedetection rate (or decrease the reliability) Results are reported inFig 3 for both ASKAP (red markers) and ATCA (green markers)simulated data The completeness C is shown with filled markersand can be parametrized as a function of the flux density S as

C =C0

2

(1+ erf

log10(S)minus log10(S0)

k

)(1)

with parametersC0=099 log10(S0)=minus288 k=040 (ASKAP) andC0=099 log10(S0)=minus358 k=059 (ATCA) For ASKAP (ATCA)observations we expect the catalogue to be gt90 complete above5 mJy (1 mJy)The false detection rate reported in Fig 3 with open dots (ASKAP)and open triangles (ATCA) is found to be of the order of 20-30 slightly decreasing by a few percent when moving away fromthe Galactic plane False detections are largely due to sidelobesaround bright sources (compact or extended) and overdeblendingof extended sources Reliability can be improved by sim10 afterapplying a neural network classifier previously trained to identifygood sources from the data (see Riggi et al 2019 for more de-tails) Results are shown in Fig 3 with open squares (ASKAP) andopen diamonds (ATCA) To further reduce the false detection rate

to acceptable levels (below 1) we visually inspected the producedcatalogue removing spurious sources that were not identified in theautomated procedure (see Section 31)For the future we expect this can be substantially improved in mul-tiple ways eg increasing the number of classifier parameters tun-ing of classifier hyperparameters introducing specialized classifierstrained to identify imaging artefacts (eg around bright sources)which cannot be efficiently removed with the classifier adopted inthis work

332 Source position accuracy

Positional uncertainties in RA and Dec can be expressed as

σα =radic

σ2α f it +σ2

αcal

σδ =radic

σ2δ f it +σ2

δ cal

(2)

where σ f it are the position uncertainties due to the fitting process(including noise and uncertainties in fit parameters) while σcal arethe uncertainties due to the calibration processσ f it is provided by the fitting routine for each source and can becompared with values obtained from simulated data as a functionof the source signal-to-noise ratio (SN) We found that the fit un-certainties are largely underestimated with respect to the expectedvalues found in the simulations The latter are a factor 3-4 larger thanthe semi-analytical estimates provided by Condon (1997)5 This ispartly expected given that the semi-analytical estimate and the esti-mate obtained from the fitting routine are neglecting the effect of thecorrelated noise in the map A correction for this effect for exam-ple using equation 41 of Condon (1997) in the maximal-smoothinglimit would yield an expected analytical uncertainty sim26 largerWe have finally set σ f it to the values obtained from the simulationsand parametrized it as a function of SN as

σASKAPα f it =

986SN

arcsec σASKAPδ f it =

601SN

arcsec

σATCAα f it =

228SN

arcsec σATCAδ f it =

298SN

arcsec(3)

Position fit uncertainties are of the order of 15-18 (05-06) atthe detection threshold and better than 05 (015) with SNgt20 forASKAP (ATCA) dataCalibration uncertainties σcal can be inferred from the posi-tion spread observed in bright sources with respect to a refer-ence catalogue The Radio Fundamental Catalogue (RFC versionrfc_2020b)6 for instance containssim17000 radio sources measuredinmultiple VLBI observations withmilliarcsecond accuracy 5 RFCsources detected in the X band cross-match with ASKAP brightsources (four with SN gt200 and one with SN gt20) and haveposition uncertainty smaller than 002 The standard deviations(sα=05 sδ=04) of the observed position offset provide a firstmeasure of σcal since both RFC and ASKAP fit position uncertain-ties are negligible at high SN Additionally we inferred ASKAPcalibration errors using 344 and 80 sources with SNgt50 match-ing to MGPS and NVSS catalogues respectively The observedstandard deviations (sα=13-15 sδ=13-14) suggest a slightly

5 A discrepancy (a factor sim2) between analytical and measured uncertain-ties was observed also in other analysis carried out with simulated data usingcaesar (Riggi et al 2019) or alternative finders (Hopkins et al 2015)6 httpastrogeoorgrfc

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 6: EvolutionaryMapoftheUniverse(EMU

6 S Riggi et al

0

02

04

06

08

1

com

plet

enes

s F

DR

1minus 05minus 0 05 1 15 2 25 3

(SmJy)10

log

Completenessaskapatca

FDRaskapaskap (sclass)atcaatca (sclass)

Figure 3 Completeness (filled markers) and false detection rate (FDR)(open markers) as a function of the injected and measured source flux den-sity respectively obtained on the simulatedASKAP (redmarkers) andATCA(green markers) source catalogues Solid lines represent the fitted complete-ness model of equation 1 Open squares and diamonds (labelled as sclass)represent the false detection rate obtained after applying a neural networkclassifier trained to identify spurious fit components in the catalogue

point sources to the residual mosaic with a density of 100 degminus2

and exponential flux density distribution (propeminusλS with λ=16 asfor real data above the detection threshold) caesar was finally runon the simulated maps and the same set of quality criteria usedfor the Scorpio mosaics were applied to the simulated catalogueSource extraction and characterization metrics were estimated bycross-matching injected sources with extracted sources

331 Completeness and reliability

Catalogue completeness and reliability were studied with simu-lated data as a function of the source flux density Sources injectedin the simulated maps are considered as the true catalogue Ex-tracted sources that do not crossmatch to any injected source arethus counted as false detections and contribute to increase the falsedetection rate (or decrease the reliability) Results are reported inFig 3 for both ASKAP (red markers) and ATCA (green markers)simulated data The completeness C is shown with filled markersand can be parametrized as a function of the flux density S as

C =C0

2

(1+ erf

log10(S)minus log10(S0)

k

)(1)

with parametersC0=099 log10(S0)=minus288 k=040 (ASKAP) andC0=099 log10(S0)=minus358 k=059 (ATCA) For ASKAP (ATCA)observations we expect the catalogue to be gt90 complete above5 mJy (1 mJy)The false detection rate reported in Fig 3 with open dots (ASKAP)and open triangles (ATCA) is found to be of the order of 20-30 slightly decreasing by a few percent when moving away fromthe Galactic plane False detections are largely due to sidelobesaround bright sources (compact or extended) and overdeblendingof extended sources Reliability can be improved by sim10 afterapplying a neural network classifier previously trained to identifygood sources from the data (see Riggi et al 2019 for more de-tails) Results are shown in Fig 3 with open squares (ASKAP) andopen diamonds (ATCA) To further reduce the false detection rate

to acceptable levels (below 1) we visually inspected the producedcatalogue removing spurious sources that were not identified in theautomated procedure (see Section 31)For the future we expect this can be substantially improved in mul-tiple ways eg increasing the number of classifier parameters tun-ing of classifier hyperparameters introducing specialized classifierstrained to identify imaging artefacts (eg around bright sources)which cannot be efficiently removed with the classifier adopted inthis work

332 Source position accuracy

Positional uncertainties in RA and Dec can be expressed as

σα =radic

σ2α f it +σ2

αcal

σδ =radic

σ2δ f it +σ2

δ cal

(2)

where σ f it are the position uncertainties due to the fitting process(including noise and uncertainties in fit parameters) while σcal arethe uncertainties due to the calibration processσ f it is provided by the fitting routine for each source and can becompared with values obtained from simulated data as a functionof the source signal-to-noise ratio (SN) We found that the fit un-certainties are largely underestimated with respect to the expectedvalues found in the simulations The latter are a factor 3-4 larger thanthe semi-analytical estimates provided by Condon (1997)5 This ispartly expected given that the semi-analytical estimate and the esti-mate obtained from the fitting routine are neglecting the effect of thecorrelated noise in the map A correction for this effect for exam-ple using equation 41 of Condon (1997) in the maximal-smoothinglimit would yield an expected analytical uncertainty sim26 largerWe have finally set σ f it to the values obtained from the simulationsand parametrized it as a function of SN as

σASKAPα f it =

986SN

arcsec σASKAPδ f it =

601SN

arcsec

σATCAα f it =

228SN

arcsec σATCAδ f it =

298SN

arcsec(3)

Position fit uncertainties are of the order of 15-18 (05-06) atthe detection threshold and better than 05 (015) with SNgt20 forASKAP (ATCA) dataCalibration uncertainties σcal can be inferred from the posi-tion spread observed in bright sources with respect to a refer-ence catalogue The Radio Fundamental Catalogue (RFC versionrfc_2020b)6 for instance containssim17000 radio sources measuredinmultiple VLBI observations withmilliarcsecond accuracy 5 RFCsources detected in the X band cross-match with ASKAP brightsources (four with SN gt200 and one with SN gt20) and haveposition uncertainty smaller than 002 The standard deviations(sα=05 sδ=04) of the observed position offset provide a firstmeasure of σcal since both RFC and ASKAP fit position uncertain-ties are negligible at high SN Additionally we inferred ASKAPcalibration errors using 344 and 80 sources with SNgt50 match-ing to MGPS and NVSS catalogues respectively The observedstandard deviations (sα=13-15 sδ=13-14) suggest a slightly

5 A discrepancy (a factor sim2) between analytical and measured uncertain-ties was observed also in other analysis carried out with simulated data usingcaesar (Riggi et al 2019) or alternative finders (Hopkins et al 2015)6 httpastrogeoorgrfc

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 7: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 7

larger calibration uncertainty of σcalα=10-11 and σcalδ=08-11 after subtracting (in quadrature) the ASKAP fit errors and theMGPS (sim09 from Murphy et al 2007) or NVSS (sim1 from Con-don et al 1998) uncertaintiesAs no ATCA-RFC matches were found to infer calibration errorsfor ATCA we considered 38 sources detected with SNgt50 andmatched to MGPS sources In this case we did not use NVSSdata as no matches were found with ATCA above the consideredsignificance level From the offset standard deviations (sα=21sδ=25) we obtained a calibration uncertainty of σcalα=19 andσcalδ=24 after subtracting (in quadrature) the ATCA fit errorsand the MGPS uncertainties Such estimates should be regarded asupper limits if MGPS positional uncertainties are underestimatedSystematic offsets possibly introduced by the fitting procedure wereinvestigated with simulated data for both ASKAP and ATCA Nobias (eg median offsets smaller than 004) was found and henceno corrections were applied to the cataloguesTo assess the absolute positional accuracy of the ASKAP ESP datawe considered the median offsets found with respect to the matchedsources in RFC ATCA MGPS and NVSS catalogues We consid-ered only the ASKAP sources with SNgt50 From the comparisonwe found that the ASKAP Dec offsets are negligible (smaller than005) while the ASKAP RA is systematically higher than thatof the reference catalogues 〈α〉=14 (RFC) 〈α〉=14 (ATCA)〈α〉=16 (NVSS) 〈α〉=17 (MGPS)7 The RA offset is consistentas a function of the ASKAP source flux density above SNgt20 (egvarying by sim02 at most from bin to bin) and slightly larger thanthe total position uncertainties estimated above The astrometricoffset affecting also other ASKAP Early Science observations isdue to a known bug in the ASKAPSoft software version used toprocess Scorpio ASKAP15 data Recent analysis carried out onnewer observations done with the full ASKAP array and improvedversions of the reduction software and calibration procedure do notreport significant astrometric offsets Indeed we were not able todetect comparable offsets from the preliminary ScorpioASKAP36maps For this analysis we have decided to account for the observedsystematic offset by increasing the calibration uncertainty to 15

333 Source flux density accuracy

Uncertainties on the measured flux density S are mainly due to theuncertainties of the source fitting process in the presence of noise(σ f it expressed here as a percentage of the flux density) and the fluxscale calibration uncertainties (σcal usually given as a percentageof the flux density) Total relative uncertainties σ can be thereforeexpressed as

σ =radic

σ2f it +σ2

cal (4)

σ f it values for each source are obtained by error propagation us-ing source parameter errors provided by the fitting routine Thesewere found a factor sim3 smaller with respect to the expected valuesobtained from simulated data Similarly to what was done in Sec-tion 332 we have set the fit uncertainties for both ASKAP andATCA data to the following parametrized values derived from thesimulations as a function of the source signal-to-noise ratio (SN)

σASKAPf it = 014times [log10 (SN)]minus272

σATCAf it = 014times [log10 (SN)]minus274

(5)

7 Themedian offset was found to be 27 but we corrected for the systematicoffset of -1 reported in Murphy et al 2007 with respect to NVSS data

1 10 210 310 (mJy)843S

1

10

210

310

(m

Jy)

912

S

data843=S912S

1 10 210 310 (mJy)843S

03minus02minus01minus

00102

-1gt

843

S91

2lt

S

=0α=-07α=-09α

Figure 4 Top Correlation between flux densities S912 and S843 of cross-matched sources found in ASKAP (912 MHz) and MGPS (843 MHz)catalogues The dashed black line corresponds to a one-to-one relation(S912=S843) Bottom Relative difference between ASKAP and MGPS fluxdensities Black dashed and dotted lines represent the difference expectedfrom S prop να for three different spectral indices (0-07-09)

Fit uncertainties aresim35 at the detection threshold and better than5 with SNgt30It is now known from simulation studies (Hopkins et al 2015 Riggiet al 2019) that many source finders produce a biased flux densitymeasurement as they approach the detection threshold caesar likeother widely used finders (eg selavy aegean pybdsm) tends tooverestimate the flux density at low SN As discussed in Hopkinset al (2015) this is likely due to a poorly constrained Gaussianfit Systematic flux density offsets 〈∆SS〉 f it were therefore charac-terized using simulated data and parametrized as a function of thesource SN as follows

〈∆SS〉ASKAPf it = 005times [log10 (SN)]minus497

〈∆SS〉ATCAf it = 006times [log10 (SN)]minus422

(6)

Systematic fit biases are sim25-30 at the detection threshold andsmaller than sim1 with SNgt25 Measured flux densities in bothASKAP and ATCA catalogues were corrected using the aboveparametrizationsTo assess the absolute flux density scale reliability we carried outa comparison of the measured ASKAP source flux densities withthose reported in previous catalogues at a nearby frequency suchas the MGPS source catalogue The correlation between ASKAP(S912) and MGPS (S843) source flux densities is reported in thetop panel of Fig 4 The dashed black line corresponds to a perfectcorrelation (S912=S843) The bottom panel represents the medianrelative flux density difference observed as a function of the MGPSflux density Error bars correspond to the semi-interquartile rangeof the data in each bin If all sources were to have a single power-law spectrum S prop να we would ideally expect to observe a relativedifference of (912843)αminus1 between the two catalogues Expectedoffsets are shown in Fig 4 with black dashed and dotted lines forthree choices of expected spectral indices α=minus07 for extragalacticsources α=0 for Galactic sources with thermal-dominated emis-

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 8: EvolutionaryMapoftheUniverse(EMU

8 S Riggi et al

sion and an average α=minus09 as observed in Section 42 or in otherradio surveys in the Galactic plane (Cavallaro et al 2018 Wanget al 2018) For sources brighter than sim70 mJy less affected bypossible flux density reconstruction bias on both catalogues weobserve that ASKAP flux densities are on average larger by sim3with respect to MGPS fluxes Provided that the source sample isnot completely dominated by Galactic thermal sources we wouldinfer that ASKAP fluxes are overestimated by 9-10 compared tothose expected from a mixed population of sources with an averagespectral index of minus09We also compared the ASKAP flux densities of selected sourceswith those expected from a power-law fit obtained with at least fourspectral points using MGPS NVSS and ATCA data 45 sources (28with SNgt50) are well-modelled by a power-law (χ2lt2 spectralindices ranging fromminus07 tominus09) over the entire frequency rangeand thus can be used as additional flux calibrators ASKAP fluxdensities for these sourceswere on average found in excess bysim10(SNlt50) and sim5 (SNgt50) with respect to the predicted valuesThe standard deviation of the observed offset of bright sources(SNgt50) is sim5 and can be used as a measure of σcal These re-sults suggest a flux density scale inconsistency in the ASKAP datacalibration or imaging process that has to be investigated with thefull array and improved releases of the data reduction pipeline Theorigin of the observed shift is in fact not fully understood at thisearly stage stage of ASKAP observations in which the calibrationprocess is not yet optimized For this reason we have decided notto correct for it and assume a larger calibration uncertainty (10)to carry out the analysis described in Section 4 For the ATCA datawe have instead assumed σcal=3 following previous works on theScorpio field (Cavallaro et al 2018)

4 ANALYSIS

41 Estimation of the fraction of resolved sources

A widely used method (Franzen et al 2015 2019) to identify nonpoint-like sources in the catalogue is based on the ratio SSpeakbetween the integrated and peak flux density which is expected tobe larger than 1 for extended sources In the XXL survey by Butleret al (2018) sources were classified as resolved if they fulfil thefollowing empirical relation

SSpeak gt a+b

SN(7)

with a =108 to account for possible calibration errors and b=203defined so as to keep 90 of the data having SSpeaklt1 above thecurve SSpeak= a - b

SN We tuned the value of b for ASKAP Scorpiosurvey using simulated data (see Section 33) To have less than 5of misclassified simulated point-sources the value of b has to beincreased to sim34 In Fig 5(a) we report the SSpeak as a functionof SN for Scorpio source components The red line represents theempirical function used in theXXL surveywith parameters a=108b=34 We report in Fig 5(b) the fraction of catalogued sources thatwould be classified as extended if thresholded in SSpeak as a func-tion of the applied threshold on the parameter a (solid black line)assuming b fixed to 34 Adopting the same criterion as in the XXLsurvey (a=108) we would classifysim20 of the catalogued sourcesas resolved This is comparable to values reported in other surveyscarried out in the Galactic plane (eg sim20 in the THOR survey(Wang et al 2018)) or far from the Galactic plane and with dif-ferent spatial resolutions (eg see Butler et al 2018 and references

1 2 3 4 56 10 20 30 210 210times2

rmsσpeakS

1minus10times5

1

2

3

4

567

10

peak

SS

(a) Integrated to peak ratio vs SN

1 12 14 16 18 2

a0

02

04

06

08

1fr

actio

n of

res

olve

d so

urce

(b) Resolved source fraction

Figure 5 Upper panel Ratio between integrated S and peak Speak fluxdensities for catalogued sources as a function of their signal to noise ratioSN The red line indicates the curve SSpeak=a+ b

SN (a=108 b=34) usedin the XXL survey above which sources are labelled as resolved Lowerpanel Fraction of catalogued sources that would be classified as resolvedas a function of the applied threshold on the parameter a (with b fixed to34) shown with a solid black line The dashed black line represents thefraction of truly resolved sources ε (found by comparison with ATCA data)that would be classified as resolved as a function of the applied threshold onthe parameter a

therein)8Using the Scorpio ATCA higher resolution image it is possible toestimate the number of truly resolved sources at least for a portionof the Scorpio field In Section 321 we found that 52648 ASKAPsources match to more than one ATCA source eg sim8 of theASKAP catalogued sources have a genuine extended nature Usingthis limited source sample we can compute the fraction of trulyresolved sources that would be classified as extended ε accordingto the SSpeak criterion described above We report the results in

8 For comparison the angular resolution of XXL and THOR survey are48 and 25 respectively

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 9: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 9

Table 4 Number of ASKAP sources detected at one or multiple radiofrequencies after cross-matching them with MGPS ATCA (also includingsub-band data) and NVSS catalogues

frequencies sources percentage()

1 2986 7212 504 122gt2 654 1583 119 294 25 065 52 136 63 157 222 548 152 379 21 05

Fig 5(b) with a dashed black line As can be seen the SSpeak pa-rameter is not sensitive enough to allow a perfect identification ofknown resolved sources (eg for a=108 we obtained an identifica-tion efficiency of sim60)We therefore repeated the analysis including additional parametersFollowing Riggi et al (2019) we considered in particular the ratioEEbeam between the source fitted ellipse eccentricity E and thebeam ellipse eccentricity Ebeam We found no significant improve-ments in the extended source identification capabilities comparedto the case of a single discriminant parameter In light of these re-sults we will conservatively assume sim8 as a lower limit in thefraction of extended sources expected in the reported catalogueFuture observations of the Scorpio field with the complete arrayand at higher ASKAP frequencies will provide an improved angu-lar resolution (by a factor of 3-4) enabling a direct identification ofadditional extended sources over the entire field

42 Spectral indices

Following the convention S prop να (where S is the integrated sourceflux) we estimated the spectral index α of ASKAP cataloguedsources using the cross-matches found in Section 32with theATCA(including also sub-band data)MGPS andNVSS source cataloguesThe number of ASKAP sources detected at one or multiple radiofrequencies are reported in Table 4 More than 70 of the sourcesdo not cross-match with any of the considered catalogues and thusno spectral index information can be reported for them 12 of thesources have flux information only at two frequencies For them onlya two-point spectral index can be reported These are to be consid-ered as first-order estimates and might not represent a good estimateparticularly for sources in which a turnover is present between thetwo frequencies For example this is the case for some of the Giga-hertz Peaked Spectrum (GPS) and Compact Steep Spectrum (CSS)radio sources (about 40 or sim20 in the compilation presented inJeyakumar 2016) For spectral indices obtained only from ASKAPand ATCA data taking into account the source extraction perfor-mance in both catalogues we expect to obtain an unbiased estimateof α above 10 mJy at 912 MHz with uncertainties smaller thansim02For sim16 of the sources we have flux information in at least threedifferent frequencies These data were fittedwith a power-lawmodelto determine the spectral indices When the fit does not convergeor does not pass a minimum quality criterion (χ2lt10) eg due tothe presence of outliers in one or more frequency bands a robustlinear regression is performed excluding data points with larger fit

05minus 0 05 1 15 2 25 3 35(SmJy)

10log

4minus

2minus

0

2

4

α

fitαaskap-atcaαaskap-nvssα

05minus 0 05 1 15 2 25 3 35

(SmJy)10

log

15minus

1minus

05minus

0

05

gtαlt

Figure 6 Top panel Spectral indices of ASKAP 912 MHz source compo-nents as a function of ASKAP flux density Blue dots represent the spectralindices obtained from a power-law fit of available ASKAP-ATCA-NVSS-MGPS data while the red squares and black triangles indicate the spectralindices obtained from only ASKAP-ATCA and ASKAP-NVSS frequenciesrespectively Bottom panel Median spectral index for the three samples as afunction of ASKAP flux density Error bars are the interquartile range (IQR)in each flux density bin

residuals Without improvement (eg the χ2 of the robust fit is stilllarger than the quality threshold) the reported source spectral indexis finally set to the value found using only two frequencies ASKAP-ATCA or ASKAP-NVSS (if no ATCA information is available)In Fig 6 (upper panel) we report the spectral indices obtainedfrom the radio spectrum fitting procedure (635 sources blue dots)on available ASKAP-ATCA-NVSS-MGPS matches and those ob-tained using only ASKAP-ATCA (49 sources red squares) andASKAP-NVSS (96 black triangles) matches Some sources (sim1of the sample) have a rather extreme spectral index (αltminus3 orαgt25) Although there is a chance they are pulsars (αltminus3) orhyper-ultra-compact Hii regions with an optically thick free-freeemission (αgt25) their measured indices are somewhat question-able We note that the majority of the potentially unreliable indicesare indeed obtained from two nearby frequencies (ASKAP-NVSS)for which a change in flux density (eg due to errors) of sim10would lead to a sim05 variation in the measured spectral indexThese spectral indices should therefore be treated with caution andbe re-estimated with data at additional frequencies Five sourcesin particular present a very steep spectrum (αltminus3) obtained byfitting ASKAP MGPS and NVSS data Spectral index values are inthis case mainly determined by NVSSmeasurements From a visualinspection we were not able to spot possible issues in the ASKAP orMGPS flux estimate eg due to a complex background or imagingartefacts NVSS data for two sources (J171435-392543 J171448-394756) are instead considerably affected by artefacts and thus thereported flux densities may be not accurate Future ASKAP Scor-pio observations bridging the frequency gap between MGPS andNVSS measurements will allow us to determine the spectral slopemore reliablyMedian spectral indices are reported in the bottom panel of Fig 6 asa function of the ASKAP source flux density The median value of

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 10: EvolutionaryMapoftheUniverse(EMU

10 S Riggi et al

4minus10

3minus10

2minus10

1minus10

1

1minus 0 1 2 3 4mJy)

true(S

10log

1minus

0

1

2

3

4

mJy

)m

eas

(S10

log

(a) Response matrix

1

10

210

310

)-1

sr1

5 x

dN

dS

(Jy

25

S

0 05 1 15 2 25 3 35(SmJy)

10log

ASKAP 912 MHz datameasuredunfolded

14 GHz data shifted to 912 MHz=-09)αKatgert+98 (=-09)αHopkins+03 (

=[-11-07]α

(b) Differential source counts

Figure 7 Top Response matrix used to unfold source detection effects(detection efficiency flux bias and uncertainty) from raw source counts Thecolor represents the probability r ji for a source with true flux density in bini to be detected with measured flux density in bin j Bottom Differentialuncorrected source counts of the Scorpio survey catalogue normalized tostandard Euclidean counts (prop Sminus25) shown with filled black dots Openblack dots represents the unfolded differential source counts obtained withthe procedure described in the text and using the response matrix aboveThe shaded area represents the existing extragalactic source count data at14 GHz parametrized by Katgert (1988) Hopkins et al (2003) and shiftedto a frequency of 912 MHz assuming a spectral index α ranging from minus11to minus07 (central value minus09) Values relative to the median spectral indexminus09 are shown as dashed and dotted lines respectively

fitted spectral indices for brighter sources (gt10mJy) is 〈α〉=minus092(IQR=068) A robust least squares linear fit to the data of Fig 6 (toppanel) yields similar values These estimates are consistent withinthe quoted uncertainties with those obtained in a completely inde-pendent analysis (Cavallaro et al 2018) carried out on a portion ofthe Scorpio ATCA field using only ATCA sub-band data

43 Source counts

In Fig 7(b) (filled black dots) we report the differential sourcecounts dNdS (number of sources N per unit area per unit fluxdensity S) obtained by dividing the number of sources found ineach flux density bin by the survey area (377 deg2 sim115times10minus3

sr) and flux density bin width Counts were not corrected for ex-pected source detection efficiency and flux density reconstructionaccuracy (bias and uncertainty) and were normalized by Sminus25 (theEuclidean slope) (Condon 1984) as conventionally done in otherstudies of source counts Error bars denote the statistical Poissonianuncertainties on the obtained countsTo compare it with other surveys we need to unfold the source de-tection effects from the measured counts The former have beenmodeled through a response matrix R = [r ji] expressing the proba-bility for a source with true flux density xequiv log10 (Strue) in bin i tobe detected with measured flux density y equiv log10 (Smeas) in bin jr ji is computed for each flux density bin (i j) as

r ji =int y j+∆y j2

y jminus∆y j2

int xi+∆xi2

ximinus∆xi2

ε(x)εreso(x)radic2πσ(x)

exp

minus (yminus x)2

2σ2(x)

dx dy

(8)

where xi ∆xi and y j ∆y j are the true and measured flux densitybin centres and widths σ and ε are the flux density uncertaintyand the source detection efficiency respectively9 Both have beenparametrized as a function of the true flux density using simulateddata (see Sections 333 and 331) εreso is introduced to correct forthe catalogue incompleteness effect related to the source angular sizeand known as the resolution bias Resolved sources with the sameintegrated flux density of point sources have in fact lower peak fluxdensities and thus a higher chance to be missed in the catalogue asfalling below the source detection threshold The response matrixwas finally normalized such that the probabilities for a true fluxdensity bin i sum up to the overall efficiency in bin i eg εiε

resoi

εreso was set equal to 1minush(gt φlim) where h(gt φlim) is the fractionof sources above a maximum angular size φlim being missed asresolved h(gt φlim) was estimated as a function of the flux densityas described in Windhorst et al (1990) Prandoni et al (2001)Thorat et al (2013) Retana-Montenegro et al (2018)

εreso = 1minush(gt φlim) (9)

h(gt φlim) = exp [minus ln2(φlimφmedian)062] (10)

φmedian =

2rdquo S14GHz le 1mJy2rdquotimesS03

14GHz S14GHz gt 1mJy (11)

φlim = max(φ1φ2) (12)

φ1 =radic

bma jbmintimes (SSthrminus1) (13)

φ2 =

radicbma jbmintimes (a+

bSσrms

) (14)

where bma j=24 and bmin=21 are the beam sizes σrms=300microJybeam is the minimum noise value in the map Sthr=15 mJyis the source minimum detection threshold10 S14GHz is the inte-

9 No bias term is included in equation 8 as we already corrected our datafor measurement bias10 The minimum detection threshold value corresponds here to the 5σ

threshold with respect to the minimum noise (σrms=300 microJybeam) in themap not to the effective detection threshold averaged over the mosaicwhich amounts to sim5 mJy

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 11: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 11

grated flux density at 14 GHz computed using a spectral indexof α=minus09 according to the results obtained in Section 42 and inCavallaro et al (2018) a=108 and b=203 are the parameters usedin Section 41 for the identification of resolved sourcesThe resulting matrix values are reported in Fig 7(a) Above 5 mJythe correction applied to the source counts is negligible as the cata-logue isgt98 complete as shown in Fig 3 and the flux uncertaintyis typically below 1 Below 5 mJy the flux uncertainty rapidlyincreases (up to 30 at 1 mJy) while the completeness rapidlydegrades to zero Source detection inefficiency thus causes truecounts to be underestimated by a factor ε while finite flux den-sity uncertainty cause sources to migrate from true to measuredflux density bins As a result the observed counts in a given bin iscontaminated by upward fluctuations from the adjacent lower binsand downward fluctuations from the adjacent upper bins If the truesource counts is steeply falling with flux density fluctuations fromthe lower bins dominate and the net effect is an overestimation ofthe countsResolution bias also causes an underestimation of the true countsComputed correction factors were found of the order of 5 at 5mJy and smaller than 1 above 50 mJyTo derive the corrected counts micromicromicro from uncorrected counts ννν know-ing the response matrix R we need to solve the following system ofequations

ν1

νM

︸ ︷︷ ︸

ννν

=

r11 r12 middot middot middot r1N

rM1 rM2 middot middot middot rMN

︸ ︷︷ ︸

R

times

micro1

microN

︸ ︷︷ ︸

micromicromicro

M ge N(15)

where M is the number of flux bins for the uncorrected countsand N is the number of flux bins for the corrected counts To de-termine the corrected counts micromicromicro we made use of a forward-foldingprocedure First we considered a model f (log10 StrueΘ) with pa-rameters Θ describing the true source counts micromicromicromodel as a function oflog10 Strue The model folded with the response matrix R was fittedto the observed counts ννν using a maximum likelihood approachThe best fit provides the expected true model counts micromicromicromodel and thecorresponding observed model counts νννmodel Corrected counts arefinally computed as

micromicromicro = ννν lowast micromicromicromodelνννmodel

(16)

We modeled the true source counts with a power-law functionalform This was found to provide a good fit (χ2ndf=44545) to oursource flux density data above the detection thresholdThe unfolded differential source counts normalized by Sminus25 arereported in Fig 7(b) with open black dots while numerical valuesare reported in Table 5 The error bars represent the total uncertain-ties σtot=

radicσ2stat+σ2

syst on the unfolded counts where σstat are thestatistical uncertainties obtained from the Poissonian errors on theuncorrected counts andσsyst=

radicσ2fit+σ2

matrix+σ2cal are the system-

atic uncertainties σfit and σmatrix are obtained by propagating thelikelihood fit parameter errors and the response matrix uncertaintiesin the unfolded counts respectively Matrix uncertainties have beencomputed by taking the variance of several response matrices ran-domized around parametrization errors for ε and σ Additionallyfollowing Windhorst et al (1990) a 10 uncertainty is considered

Table 5 912 MHz source counts in Scorpio field The columns repre-sent (1) flux density bin interval ∆S in mJy (2) flux density bin centre Sin mJy (3) number of catalogued sources N in each flux density bin (4)corrected number of sources Ncorr (5) uncorrected normalized differen-tial source counts (S25dNdS) (6) corrected normalized differential sourcecounts (S25dNdS|corr) with statistical (7) and total (8) uncertainties

∆S S N Ncorr S25dNdS S25dNdS|corr σstat σtot(mJy) (mJy) (Jy15srminus1) (Jy15srminus1)

023 050 111 164015 023 345 033 058037 079 409 218364 172 916 043 110058 126 623 154404 522 1293 050 147093 200 700 98856 1169 1651 056 183147 316 600 68318 2000 2277 087 259233 501 507 54179 3372 3603 155 422369 794 354 37061 4697 4917 251 607585 1259 287 29743 7598 7874 446 1009927 1995 204 21018 10776 11103 782 15671469 3162 146 14981 15388 15789 1248 23682329 5012 84 8592 17665 18068 1950 32963690 7943 60 6122 25176 25689 3338 53755849 12589 55 5602 46046 46898 6384 1017415774 22387 46 4677 60218 61227 9002 1412631473 44668 16 1625 59032 59937 14902 21910335011 158489 8 811 65756 66673 24178 34837

in the resolution bias correction εreso To estimate σcal we shiftedthe measured flux densities by plusmn10 (the calibration uncertaintyquoted in Section 333) repeated the unfolding procedure on theresulting shifted spectra and compared the obtained counts withunshifted countsExisting source count data at 14 GHz as parametrized by Katgert(1988) Hopkins et al (2003) and shifted to a frequency of 912MHz (assuming a spectral index α ranging from minus11 to minus07)are shown for comparison as a shaded blue area The shaded areaalso accounts for the count fluctuations (sim20) reported in pastworks comparing source counts from different extragalactic fields(Windhorst et al 1990 Hopkins et al 2003) or from different re-gions of the same survey (Prandoni et al 2001 Retana-Montenegroet al 2018) The observed spread can be due to either systematicuncertainties different correction factors applied or to the largescale structure of the universe (ie cosmic variance)As can be seen the measured source counts as a function of thesource flux density are consistent with the trend reported in othersurveys carried out far from the Galactic plane (eg see Katgert1988 Hopkins et al 2003 and references therein) The discrep-ancies observed assuming a spectral index α=minus07 ranging from15 to 20 are within the quoted systematic uncertainties Thepresence of our Galaxy is also expected to play a role in this compar-ison Galactic sources indeed contribute to the overall source countsalthoughwith a smaller fraction (sim5 at least from the analysis pre-sented in Section 51) An overdensity effect (sim10-15) reportedfor example by Cavallaro et al (2018) can not however be clearlyidentified being of the same order of the reported uncertaintiesFor the sake of completeness we derived the source counts in aregion outside the Galactic plane (|b|gt2) following the same pro-cedure described above and using an updated response matrixparametrized on the considered mosaic region No significant dif-ferences (within few percent) were found with respect to the sourcecounts obtained over the full mosaic suggesting that the majorityof the unclassified sources have an extragalactic origin

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 12: EvolutionaryMapoftheUniverse(EMU

12 S Riggi et al

5 SOURCE CLASSIFICATION

51 Search for known Galactic objects

We cross-matched the ASKAP Scorpio source catalogue with dif-ferent astronomical databases to search for possible associationswith known Galactic objects We restricted the search to the follow-ing types of objects

bull Stars associations searched in the SIMBAD AstronomicalDatabase11 in the Galactic Wolf-Rayet Star Catalogue12 (Ross-lowe amp Crowther 2015) and in the Gaia data release 2 catalogue(Brown et al 2018)bull Pulsars associations searched in theATNFPulsarCatalogue13

(Manchester et al 2005) (version 163)bull Planetary Nebulae (PNe) associations searched in the

Hong KongAAOStrasbourg H-alpha (HASH) Planetary NebulaDatabase14 (Parker et al 2016)bull Hii regions associations searched in the WISE Catalogue of

Galactic Hii regions15 (Anderson et al 2014)

Extended objects such as the supernova remnants (SNRs)listed in the Galactic SNR catalogue16 by Green (Green 2019)were not considered in this analysis Possible associations and newdetections will be reported in a future work dedicated to Scorpioextended sourcesWe summarize the cross-match results in Table 6 The number ofassociations found for each catalogue is reported in column 4 whilethe expected number of false matches is given in column 5 Thesewere estimatedwith the sameprocedure described in Section 32 ieevaluating the number of matches found within the chosen radius inseveral randomized catalogues The search radius considered forthe matching reported in column 3 corresponds to the maximumstatistical significance of the match signal above the backgroundThe number of objects labelled as confirmed in the astronomicaldatabases is reported in column 6Only 146 objects were associated to known classes of Galacticobjects corresponding tosim4 of the total number of Scorpio cat-alogued sources The vast majority of the catalogued sources arethus labelled as not classified A 2D map showing the positions ofboth classified and unclassified sources is reported in Fig D1

511 Stars

Inside the Scorpio ASKAP region we selected 10628 stars in theSIMBAD database and 19 Wolf-Rayet stars in the Rosslowe ampCrowther (2015) catalogue These were cross-matched to ASKAPsource components within a match radius of 4 in sky coordinatesSince the Gaia catalogue is densely populated (more than 9times106

entries found in the Scorpio region) we lowered the search radius to2 comparable to the positional uncertainties obtained in ASKAPWe found 20 associations in SIMBAD among them 7 YSO (1confirmed 6 candidates) and no associations withWolf-Rayet starsup to a matching radius of 32 SIMBAD matches are expected tobe real as the estimated number of chance matches is 23plusmn03 (seeTable 6) The associations found with Gaia DR2 (933) are instead

11 httpsimbadu-strasbgfrsimbad12 httppacrowtherstaffshefacukWRcatindexphp13 httpswwwatnfcsiroauresearchpulsarpsrcat14 http2021891171018999gpnedbMainPagephp15 httpastrophyswvueduwise16 httpwwwmraocamacuksurveyssnrs

Table 6 Number of sources from different astronomical catalogues (seetext) associated to ASKAP Scorpio radio sources

Obj Type Catalogue rmatch nmatches nrandommatches Confirmed

() objects

StarSIMBAD 4 20 23plusmn03 13WR Cat 4 0 0 0GAIA DR2 2 933 8742plusmn96 0

Pulsar ATNF Cat 8 21 0 21

PNe HASH 8 38 0 27

Hii regions WISE 32 67 0 35

All 146 96

GAIA matches not included in the final count

dominated by random matches and will not be further consideredA list of the associated objects found is reported in Table E2 Thereported classification was investigated with infrared and opticaldata Details are reported below

(i) SSTGLMC G3437018+000861 invisible in optical andnear-IR clear but not particularly red in the mid-IR Not immersedin a region of star formation YSO candidate classification is pre-mature Could well be extragalactic

(ii) IRAS 16495-4140 likely a YSO It is a small clump of starsof which two dominate in the near-IR (2MASS) Together they arebright and red in the mid-IR (WISE not covered by Spitzer)

(iii) 2MASS J17062471-4156536 classified as a YSO candi-date with star-forming activity (and dark cloud) nearby Could beextragalactic Invisible in the optical but bright (and red) in 2MASSNot particularly red in Spitzer and WISE

(iv) SSTGLMC G3442155-007460 only detected in the mid-IR faintly There is nothing to suggest otherwise that this is abackground AGN It is not near star-forming complexes and it isnext to darker areas of extinction ie viewed through a relativelymore transparent region The YSO classification is questionable

(v) IRAS 16534-4123 likely extragalactic It is exceedinglyfaint in optical and very bright in WISE

(vi) [MHL2007] G3450052+018209 1 Class I protoclusterClump of red stars in 2MASS no detection in optical embeddedwithin a diffuse background emission seenwith Spitzer (GLIMPSE)and saturating in WISE

(vii) HD 326586 F8 star bright GALEX (UV) source(viii) SSTGLMC G3426544-003827 faint red Detected in

optical and near-IR not particularly bright or red in mid-IR Notnecessarily a YSO and no indications that it should be radio-loudCould well be a background AGN

(ix) IRAS 16472-4401 related by SIMBAD with an IRASsource undetected in 2MASS Close to the sourcersquos sky position(asymp6 arcsec) a YSO was detected by the ATLASGAL survey at 870microm (Urquhart et al 2018)(x) 2MASS J16504054-4328122 (IRAS 16470-4323) while

SIMBAD classes it as an AGB star there is no direct evidencefor this It is strangely yellow in the DECaPS (five-band optical andnear-infrared survey of the southern Galactic plane with the DarkEnergy Camera (DECam) at Cerro Tololo) unlike any other objectaround it It is red in 2MASS but again yellow in Spitzer andWISEThe Gaia data are marginal so we cannot completely exclude anextragalactic nature

(xi) HD 326392 classified as B8 star asymp830 pc far (Brown et

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 13: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 13

al 2018) At the stellar distance the measured flux of HD 326392corresponds to a radio luminosity ofasymp6times1017 ergs sminus1 Hzminus1 morethan one order of magnitude higher than CU Vir (asymp3times 1016 ergssminus1 Hzminus1 Leto et al 2006) a magnetic late B type star well studiedat radio regime The radio luminosity of HD 326392 is instead morecomparable to those of strong magnetic stars of B2 spectral typelevel asymp10 kG and luminosity close to 1018 ergs sminus1 Hzminus1 (Letoet al 2017 2018) This suggests HD 326392 as a possible strongmagnetic star(xii) 2MASS J17122205-4230414 fairly red but not too faint in

the optical bright in themid-IR As such it is isolated the post-AGBclassification seems more likely than a YSO (as listed in SIMBAD)It is found to be slightly extended and classified as a planetary nebulaby (Suarez et al 2006) with high confidence though their opticalspectrum shows no trace of [OIII](xiii) TYC 7872-1355-1 Anonymous star but in a reflection

nebula(xiv) MSX6C G3464809+001320 YSO Class 0 (invisible in

near-IR) with a 67-GHz methanol maser detection by Gaylard ampMacLeod (1993) and a single 100 polarized 1665-MHzOHmaserpeak by Caswell amp Haynes (1983)(xv) 2MASS J17074166-4031240 likely a YSO It is invisible

in the optical but clear and red in 2MASS however the mid-IRemission comes mainly from the bright rim (photo-dissociationregion) of a cloudHII region immediately to its side most likelythe origin of the radio emission(xvi) CD-38 11343 Classified as double star (er

(M3Ve+M4Ve)) in SIMBAD Eruptive pair of M dwarfs at15 pc with likely expected variability also in radio(xvii) IRAS 17056-3930 very faint red optical but bright(ish)

red near-IR and mid-IR source However it is very isolated there isnothing like it within at least 5 arcminute radius in a field of starswith the nearest sign of star formation activity about 8 arcminutesaway Further investigations are needed to address its nature(xviii) IRAS 17056-3916 no detection in optical or near-IR

but very red mid-IR source possibly unresolved It is located rightat the edge of a dark cloud which could suggest it is a YSO orexplain why it is not seen in the optical Further investigations areneeded to confirm if of extragalactic origin(xix) Cl NGC 6318 PCA 7229 Classified as star in cluster

in SIMBAD This might in fact be the AGAL G347919-00762

YSO detected in Spitzer data(xx) TYC 7873-953-1 radio source generically classified by

the SIMBAD database as star This is associated with a visiblesource about 12 kpc far (Brown et al 2018) Further this wasdiscovered as a variable radio source at 14 GHz from NVSS data(Ofek amp Frail 2011)

We compared the matches found with those obtained in pilotobservations of the Scorpio field with ATCA at 21 GHz Only2 of the 10 star associations reported in Umana et al (2015) areretrieved also in the ASKAP map Due to a lack of sensitivity noradio sources are detected in the direction of the two Wolf-Rayetstars (HD 151932 HD 152270) previously detected in ATCAFor 6 associated objects shown in Fig 8 and reported as topentries in Table E2 we were able to derive a spectral indexmeasurement (column 14) using Scorpio ASKAP and ATCAMGPS and NVSS data (see Section 42) Among them we have 4Young Stellar Object (YSO) candidates with fitted spectral index α021plusmn006 (SSTGLMC G3437018+000861) 002plusmn015 (2MASSJ17062471-4156536) 017plusmn009 (SSTGLMC G3442155-

007460) 043plusmn008 ([MHL2007] G3450052+018209 1)

01minus 0 01 02 03 04 05GHz)ν(

10log

05

1

15

2

25

(Sm

Jy)

10lo

g

SSTGLMC G3437018+000861006)plusmn=021α=136

2χsimfit (

IRAS 16495-4140008)plusmn=058α=069

2χsimfit (2MASS J17062471-4156536

015)plusmn=002α=422 2χsimfit (

SSTGLMC G3442155-007460009)plusmn=017α=259

2χsimfit (

IRAS 16534-4123007)plusmn=-006α=056

2χsimfit (

G3450052+018209008)plusmn=043α=021

2χsimfit (

Figure 8 Spectral data of Scorpio sources associated to stars in the SIM-BAD database obtained using ASKAP and ATCA data (filled markers) andprevious MGPS and NVSS observations (open markers) The power-law fitsare reported with coloured lines

Taking into account the reported uncertainties these values aregenerally consistent with the spectral indices expected in opticallythin free-free emission processes from ionized gas Accordingto Scaife (2012) Ainsworth et al (2012) Anglada et al (2018)the free-free radio emission at centimetre wavelengths is ascribedto outflow processes (eg thermal jets) causing the required gasionization particularly in the earliest protostellar stages (Class0 and Class I) The resulting YSO radio spectral indices α areexpected in the range minus01ltαlt11 but their values depend onthe protostar evolution For example in collimated outflowstypical of early protostar stages a radio spectral index α sim025 isfavoured while for standard conical jets spectral indices around06 are expected (Reynolds 1986 Anglada et al 1998) Existingmeasurements mostly fall in the above range although in somecases the observed spectral indices (ltminus05) suggest a contributionfrom non-thermal processes (Ainsworth et al 2012) Our spectralindex measurements are found within the range expected fromfree-free emission Additional radio data at different frequenciesare however needed to further constrain the dominant emissionmechanism

512 Pulsars

The ATNF Pulsar database has 58 catalogued pulsars inside theScorpio region We carried out a search for possible associationswith Scorpio source components finding 21 associations within amatching radius of 8 and a number of chance matches compatiblewith zeroIn Table E3 we report the full list of matched objects Four of these(J1654-4140 B1703-40 J1702-4128 J1702-4217) shown astop entries in the table have spectral index information (column 14)that can be obtained from ASKAP and ATCA data only Column12 represents the spectral index reported in the ATNF cataloguefrom measurements at 08 14 and 30 GHz (Jankowski et al 20182019 Kramer et al 2003 Johnston and Kerr 2018 Johnston etal 1992) Flux density measurements available in the literaturefor three of them (J1654-4140 B1703-40 J1702-4128) are re-

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 14: EvolutionaryMapoftheUniverse(EMU

14 S Riggi et al

01minus 0 01 02 03 04 05 06GHz)ν(

10log

15minus

1minus

05minus

0

05

1

15

2

(Sm

Jy)

10lo

g

PSR J1654-4140

this analysis

past data (kbm+03jvk+18)

016)plusmn=-200α=394 2χsimfit (

PSR B1703-40this analysispast data (jk18jvk+18jlm+92jbv+19)

011)plusmn=-282α=318 2χsimfit (

PSR J1702-4128

this analysis

past data (jvk+18)

010)plusmn=-041α=375 2χsimfit (

Figure 9 Spectral data for J1654-4140 (open black dots) B1703-40 (openred squares) and J1702-4128 (open blue triangles) pulsars reported in theATNF pulsar catalogue kbm+03 (Kramer et al 2003) jvk+18 (Jankowskiet al 2018) jk18 (Johnston and Kerr 2018) jlm+92 (Johnston et al 1992)jbv+19 (Jankowski et al 2019) ASKAP and ATCA spectral data obtained inthis work are shown with filled markers Single power-law fits are reportedwith solid dashed and dotted lines for the three sample pulsars

ported in Fig 9 with open markers New measurements obtainedin this work are shown with filled markers Both ASKAP andATCA fluxes compare well to existing data allowing us to derivea new spectral index measurement for all three pulsars Fit resultsare shown in the figure with solid dashed and dotted lines Thesteep spectral index values found for J1654-4140 and B1703-40reported in Table E3 (column 14) are compatible with the distri-bution of pulsar spectral indices compiled in Maron et al (2000)(〈α〉=minus18plusmn02) The flat spectrum of J1702-4128 confirmed byour ASKAP and ATCA observations supports the hypothesis of apulsar wind nebula (PWN) as the origin of the radio emission Thiswas investigated by Chang et al (2008) to explain the nature ofthe X-ray emission from Chandra CXOU J1702524-412848 but noconclusive evidence was reached Future ASKAP EMU and POS-SUM data might give further evidence for a PWN if this source islinearly polarized

513 Planetary Nebulae

Within our surveyed area HASH has 60 PNe17 that were cross-matched to ASKAP catalogued sources at different matching radiiA number of 38 associations were found assuming a radius of 827 of the 38 matches correspond to confirmed PN objects Thenumber of chance matches found is compatible with zero Wevisually inspected the 5 confirmed PNe (MPA J1715-3903 MPAJ1717-3945 MPA J1654-3845 MPA J1644-4002 PHR J1709-

3931) that were not cross-matched to any ASKAP source In the

17 This number refers to the number of PNe (32 True PN 6 Likely PN 7Possible PN 15NewCandidates) listed in the onlineHASHdatabase version46 (http2021891171018999gpnedbMainPagephp) at thetime of writing of the paper and found inside the source finding region Ifwe also consider the mosaic edges the PN count increases to 67 (36 TruePN 7 Likely PN 8 Possible PN 16 New Candidates)

01minus 0 01 02 03 04 05 06GHz)ν(

10log

02minus

0

02

04

06

08

1

12

(Sm

Jy)

10lo

g

IRAS 16515-4050

sr)-11021)x10plusmn=(240Ωpc -6 cm6

076)x10plusmn=040 EM=(6532χsimFF fit (

MPA J1654-4041

010)plusmn=-058α=166 2χsimPL fit (

PHR J1654-4143

017)plusmn=-103α=070 2χsimPL fit (

Pre 11

010)plusmn=-016α=286 2χsimPL fit (

DGPK 2

032)plusmn=064α=194 2χsimPL fit (

Figure 10 Radio continuum spectra for five PNe obtained using ASKAPand ATCA observations

direction of the first four PNe there are no radio sources in themap A faint radio source is instead found in the direction of PHRJ1709-3931 but it is detected at an SN of only 41A full list of associated PNe is reported in Table E4 With ASKAPwe detected the 9 HASH PNe studied with ATCA 21 GHz ob-servation in Ingallinera et al (2019) ATCA flux measurement ishowever reported in this study for only 5 PNe (1 confirmed IRAS16515-4050 4 candidates MPA J1654-4041 PHR J1654-4143Pre 11 DGPK 2) shown in Fig 10 and listed as top entries inTable E4 Four PNe were fitted with simple power-law models (la-belled as PL in Fig 10) as their radio spectra do not show evidenceof spectral curvature in the studied frequency range Although the fitmodels are not highly statistically significant the results constitutea first measurement of the average spectral index for these objectsas no other measurements are available in the literatureThe radio spectrum of IRAS 16515-4050 PN cannot be describedby a single power-law model and was instead fitted with a thermalfree-free emission model S f f

S f f (ν TeEMΩ) = F(ν Te)(1minus eminusτ(ν TeEM))Ω (17)

τ(ν TeEM) = 3014times10minus2Tminus15e

GHz

)minus2g f f (ν Te) EM (18)

g f f (ν Te) = ln[

4955times10minus2(

ν

GHz

)minus1]+15lnTe (19)

where F(ν Te) is the Planck function Te is the electron tempera-ture EM is the emission measure and Ω is the source solid angleA good fit of the data was obtained with Te fixed to 104 K and theother parameters free to vary18 EM=653plusmn076times106 cmminus6pc andΩ=240plusmn021times10minus11 sr (angular size θ sim1)For the rest of catalogued PNe we were able to estimate the spec-tral index by combining our measurements with existing ones at843 MHz (Murphy et al 2007) (MOST) 14 GHz (Condon et al1998 Condon and Kaplan 1998) (VLA) 48 GHz (van de Steene

18 We considered an average value with respect to measured PN electrontemperatures found in the range 5000-15000 K (eg see Zhang et al 2014for results on a sample of 48 Galactic PNe)

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 15: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 15

and Pottasch 1993) (ATCA) Some of them obtained using 843and 912 MHz measurements only are clearly unreliable due to thelimitations discussed in Section 333The spectral indices obtained with data at higher frequencies arefound in general agreement with the expected nature of the emissionin PNe ie thermal free-free radiation due to electron-ion interac-tions in the nebula shell (Kwok 2000) The expected spectral indiceshowever vary considerably depending on the thermal and electrondensity scenario roughly sim minus01 in an optically thin regime andpositive up to sim2 (Pottasch 1984) for optically thick PNe The firstscenario is likely in place in MPA J1654-4041 PHR J1654-4143Pre 11 Vd 1-5 and Vd 1-6 for which we observed negative spec-tral indices The second scenario seems favoured for some otherdetected PNe eg IRAS 16515-4050 DGPK 2 PM 1-119 and PM1-131 Due to the limited number of radio observations availablethe precision achieved on the spectral indices does not allow usto constrain the PN nature and corresponding emission mechanismDeeper analysis possibly in combination with IR data as in Fragkouet al (2018) Ingallinera et al (2019) will be therefore performed inthe future once new ASKAP observations with different frequencybands become available

514 Hii regions

Within the Scorpio region used for source finding we found 356Galactic Hii regions catalogued in Anderson et al (2014)19 256of these were classified by these authors as known (ie objectsconfirmed by Anderson et al 2014 or in previous studies includingradio quiet Hii regions) with the rest as candidates (including thoseclosely located to a known Hii region) Their sizes range from 02to sim23 arcmin with a median of sim1 arcmin Within a matchingradius of 32 we found 67 Hii regions associations20 The matchedHii regions have a reported radius between 12 and 100 (averageradius sim52) From a visual inspection we were able to confirmsim90 of the matches found in the automated analysisThirteen Hii regions were detected in both the ASKAP and ATCAmaps We were able to determine a first measurement of the radiospectral index for 9 of them (6 confirmed objects and 3 candidates)The remaining ATCA sources have either an unreliable spectral fitdue to noisy subband data or an unclear association with ASKAPdata For instance ASKAP source associated to G344993-00265is located close to another more compact Hii region (G344989-00269) The higher resolution of ATCA observations allows usto distinguish the two sources when estimating their flux densitiesThis is not the case for ASKAP data in which the two sourcesare blended Future ASKAP observations will provide the requiredsensitivity resolution and additional spectral data to refine and ex-tend this analysis Finally an additional spectral index measurementwas obtained for G347921-00763 using MGPS and NVSS dataMore details for each source are reported in Table E5

52 Extragalactic sources

A large number of the extracted sources (sim3800 source islandsgt95 of the catalogue) are not classified or associated to a known

19 The number of Hii regions falling in the full ASKAP mosaic includingborders is 382 as reported in paper 1 (Umana et al in prep)20 The number of Hii regions detected in ASKAP is 261 as reported inpaper 1 (Umana et al in prep) if we also consider the mosaic borders andthe extended sources that were excluded from this catalogue

Figure 11 A sample radio galaxy classified with high-confidence in boththe ASKAP Scorpio map (colour scale and solid black contour) and inthe VLASS survey (superimposed solid green contours) The ellipse at thebottom left indicates the ASKAP synthesized beam size

astrophysical object Following the findings reported in previousradio surveys carried out in the Galactic plane (Wang et al 2018Cavallaro et al 2018) and the comparison with extragalactic sourcecounts reported in Section 43 we may reasonably expect thatthe majority of them are extragalactic objects A search in theNASAIPAC Extragalactic Database (NED)21 returned only 20known objects classified as galaxies (G) Two of them (2MASSJ17172771-4306573 2MASX J16463421-3903086) are foundassociated to Scorpio catalogued sources Another one (2MASSJ17162433-4225102) is likely associated to a faint non-cataloguedASKAP radio source (SN=37) For the rest there are no indica-tions for a co-spatial radio emission at the current map sensitivityWe report here a preliminary analysis to increase the number ofclassified sources and provide the basis for more advanced studiesto be done in the future Indeed some of the unclassified sourceislands are found to have a bipolar morphology resembling those ofradio galaxy lobes or can be visually associated to neighbor islandsconnected by a radio jet-like emission We can therefore visuallyinspect the ASKAP map to search for this kind of object and labelthem as candidate radio galaxy Since a precise identification usingmorphological considerations only is currently severely limited bythe resolution of our ASKAP map we also considered the ATCAmap which provided the needed resolution to address some of theunclear identifications Results are reported in Table 7A number of 277 candidate radio galaxies were identified None ofthem are associated to known NED extragalactic objects The ma-jority was found associated to one island detected in the catalogue(eg other components are either not visible or below the detectionthreshold) and only a few associated to two or three islands (whichmay correspond to a single source) The number of source islandsassociated to the candidate galaxy do not always have a 1-to-1 cor-respondence to the physical components (eg core lobes) In manycases the latter are in fact visible inside a single islandFor almost half of the cases there were no strong indications whether

21 The NASAIPAC Extragalactic Database (NED) (httpnedipaccaltechedu) is operated by the Jet Propulsion Laboratory CaliforniaInstitute of Technology under contract with the National Aeronautics andSpace Administration

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 16: EvolutionaryMapoftheUniverse(EMU

16 S Riggi et al

Table 7 Number of radio galaxies identified in the Scorpio ASKAP mosaicper number of islands and labelled according to the degree of confidencereached in the identification (see text)

islands Degree of confidence

LOW HIGH

1 103 1312 10 283 0 5

All 113 164

the radio morphology would suggest a radio galaxy These havebeen labelled as candidates with lower degree of confidence in Ta-ble 7 About 70 radio galaxy candidates lying north of Dec=minus40deg were visually inspected in the QuickLook data release of theVLASS 2-4 GHz survey (Lacy et al 2020) Only two of them (oneis reported in Fig 11) were firmly confirmed as radio galaxies fromtheir morphology while the rest are resolved out or only showing thecore component Given the resolution limitations already discussedfuture ASKAP observations at higher resolution will be needed toconfirm the nature of these candidatesSpectral information measured only for 74 source componentsfound in candidate radio galaxy islands may provide additionalhints for these studies at least for a subset of the sources Followingthe classification scheme reported in Zajaček et al (2019) 62 ofthem have a lobe-like steep index (α lt minus07) 6 are found with ajet-like index (minus07le α leminus04) and 6 with a flatter core-like index(α gtminus04)

53 Unclassified sources

The remaining unclassified sources (3588) which still constitutethe major part of the catalogue (sim87) cannot be classified on thebasis of their morphology given that the majority (sim95) does notpresent any internal structure at the resolution of ASKAP map Forthese the radio spectral index and the correlation with observationsat different wavelengths can provide valid hints The radio-infraredcorrelation in particular was suggested by many authors (eg seeIngallinera et al 2014 Fragkou et al 2018 Akras et al 2019) as apowerful tool for the identification of different classes of Galacticobjects Radio continuum emission traces the ionized part of thenebula whereas infrared traces excited gas and the dust In thiscontext the combination of infrared color indices obtained withdifferent wavelength filters (from few microm to hundreds of microm) hasbeen a widely used technique to probe IR emission from differentcomponents (Nikutta et al 2014 Anderson et al 2012) For ex-ample the correlation of far-infrared to mid-infrared colors allowsto compare emission from cold and warmer dust components Theratio between IR fluxes at 12 microm and 8 microm is instead sensitive tothe emission from PAHs (polycyclic aromatic hydrocarbons)As discussed in Section 323 about 2500 unclassified sources(sim70) do not have infrared counterparts within a search radiusequal to the ASKAP synthesized beam size These are expectedto be mostly extragalactic sources and pulsars For 607 of thesewe obtained a radio spectral index reported in Fig 13 (solid blackhistogram) that will be combined with additional observables infuture classification studies Bearing the cross-match limitations inmind (highlighted in Section 323) we will carry out the follow-ing analysis to the other 284 of the unclassified sources found with

infrared counterparts within 8 of which 67 have both the radiospectral index information and measured radio-infrared correlationparameters

531 Infrared-radio colors

We report in Fig 12 a series of correlation plots of selected infrared-infrared and infrared-radio color indices for different IRwavelengths(34 microm 46 microm 8 microm 12 microm 22 microm 70 microm) Color ci j is definedas the magnitude difference between measured fluxes Si and S j inband i and j where λ jgtλi eg ci j = 25log10(S jSi) As only asubset of the available colors are independent we selected the mosteffective ones in terms of source discrimination power reported inthe literature (Nikutta et al 2014 Anderson et al 2012) c3446c4612 c1222 c812 c2270 An additional color (c22radio) was con-sidered to include the radio information Unclassified objects areshown with open black dots while pre-classified sources found inthe Scorpio mosaic with colored markers (Hii blue crosses PNegreen diamonds stars red stars YSOs orange asterisks galaxiespurple dots)We superimposed data for different classes of objects taken fromprevious radio surveys crossmatched to the aforementioned infraredsurveys Radio star data (shown as red stars) at 843 MHz and 14GHz are taken fromKimball and Ivezić (2008)22 and fromWendker(1995) Data for galaxy and QSO with point-like morphology at 14GHz (shown respectively with purple dots and cyan triangles) aretaken fromKimball and Ivezić (2008) Additionally we included in-frared data for stars galaxies and quasars from Clarke et al (2020)taking only sources reported with classification probability equalto 1 and requiring infrared fluxes above the AllWISE 5σ limits forall bands (Cutri et al 2013) Hii regions (blue crosses) and PNe(green diamonds) at 843 MHz and 14 GHz data are taken from theAnderson et al (2014) and HASH (Parker et al 2016) databases re-spectively Radio data at 5 GHz for YSO (orange asterisks) are takenfrom Urquhart et al (2009) Pulsar data from the ATNF database(Manchester et al 2005) were also considered but since no matchwas found with IR data they were not reported in Fig 12From Fig 12 we observe that most color correlations provide valu-able information to separate Galactic and extragalactic sourcesLonger wavelength bands are found more sensitive to object iden-tity information for classification purposes For example knownGalactic objects tend to have larger S1222S3446radio flux ratios onaverage compared to galaxies From the data shown in the bottomright panel onemay be tempted to conclude that the sole presence ofan associated 70 microm emission is a clear indication of the sourcersquosGalactic origin This is rather due to a lack of reference data forextragalactic background objects in this portion of the parameterspace Indeed far infrared emission (also at 70 microm) from severalnearby galaxies was reported in different works (eg see Dale et al2012) and expected from our calculations23Reference data have some limitations to be considered for further

22 Catalogue data available athttpwwwaocnraoedu~akimballradiocat_20shtml23 Following Silva et al (1998) Michałowski et al (2008 2010) Iglesias-Paacuteramo et al (2007)we computed the expected 70 microm emission as a functionof redshift for different SED models and compared it with the Hi-GALdetection threshold of sim05 mJy (Molinari et al 2016) For radio fluxesclose to the catalogue detection threshold (sim5 mJy) only galaxies (z lt06)with specific SED models can be detected Above the ASKAP detectionthreshold we expect galaxies to be detectable at all redshifts and thus topopulate the corresponding parameter space in Fig 12

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 17: EvolutionaryMapoftheUniverse(EMU

EMU Compact radio sources in SCORPIO 17

Figure 12 Scatter plots of selected infrared-infrared or infrared-radio colors for different infrared bands (34 microm 46 microm 12 microm 22 microm 70 microm) Unclassifiedobjects are shown with gray smaller dots while pre-classified sources with colored markers (Hii regions blue crosses PNe green diamonds stars red starsYSOs orange asterisks galaxies purple dots QSO cyan triangles)

analysis First the currently available radio data are obtained atslightly different frequencies (843 MHz 14 GHz 5 GHz) withrespect to the ASKAP Scorpio data (912 MHz) The introducedspread in the IRradio colors is expected to be negligible for Galac-tic sources with thermal emission or flat spectra A correction forsources with steeper spectra (expected of the order of 05 and 18

at 14 and 5 GHz respectively with α=minus1) is however not pos-sible as the spectral index information is only available for veryfew sources A second limitation is that the source parameter datahave many missing values for multiple reasons that can not easilybe distinguished from one another without an in-depth revision ofthe external catalogue data This prevented us from encoding and

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 18: EvolutionaryMapoftheUniverse(EMU

18 S Riggi et al

3minus 2minus 1minus 0 1 2 3

α1minus10

1

10

210

uncl

assi

fied

sour

ces

no IRradio dataIRradio - galactic classIRradio - extragal class

Figure 13Measured spectral index for unclassified sources The solid blackline histogram includes unclassified sources forwhich no combined infrared-radio parameter data are available Filled red and shaded blue histogramsinclude unclassified sources that were classified asGalactic and extragalacticsources respectively in the combined infrared-radio analysis

imputing missing information in an effective way For examplesome catalogues do not always provide match information for allIR bands eg WISE W1 and W2 information is missing in the Hiiregion and PN catalogues but we may expect for a subset of thesesources an emission at these wavelengths24 Flux information mayalso bemissing for some sources either because the reference surveycovers a limited portion of the sky (eg radio galaxy data obtainedin extragalactic surveys do not have 8 and 70 microm information) orbecause the source is not emitting or below the detection thresh-old We eventually removed sources with missing data (eg usinglistwise deletion) reducing the sample size usable for analysis

532 Classification analysis

We attempted to perform a simple source classification analysisto identify Galactic source candidates using four color parameters(c3446 c4612 c1222 c22radio) for which we have the largest avail-ability of data for all bands The resulting training sample (sim1650sources) is unbalanced by a factor 10 towards extragalactic ob-jects A decision tree classifier was then trained on this dataset andtested on the unclassified sample (284 sources) The optimal treemaximum depth size (3) was determined by comparing classifi-cation performances on multiple random validation samples Onthe full training sample we obtained these classification metrics(recallprecisionF1-score) 099097098 for extragalactic sourcesand 069089078 for Galactic sources As expected from Fig 12

24 We visually inspected some of the catalogued Hii regions in all WISEbands The infrared emission in bands W1 and W2 was effectively foundabsent or not correlated with the W3 and W4 emission but we cannotexclude such a correlation for the rest of catalogued objects that were notinspected In fact in Assef et al (2018) the authors conservatively removedthe areas of the sky around WISE Hii regions to eliminate the Galacticsource background when searching for potential AGN inW1 andW2 bandsAn emission at these shorter wavelengths albeit smaller compared to W3andW4 bands can be therefore present and was indeed observed in some ofthe Hii regions catalogued by CORNISH (Purcell et al 2013) (for examplein G0196062-009018)

stars are the most misclassified type of object (sim70 of the wrongidentifications) From the decision rules it can be seen that one ofthe color parameter (c1222) is not effectively used for the classifi-cation Decision rules can be expressed in pseudocode as follows

if c22radio leminus405 thenGalactic

elseif c4612 leminus051 then

if c3446 leminus068 thenextragalactic

elseGalactic

endelse if c4612 gtminus051 amp c4612 le297 then

extragalacticelse

Galacticend

end

The trained tree classifies 99 sources as Galactic objects 75 ofthem with classification probability larger than 90 In Fig 13 wereport the measured spectral indices for a subset (30) of them (filledred histogram) in comparison with sources classified as extragalac-tic (shaded blue histogram)Unfortunately we do not have enough training data to classify dif-ferent types of Galactic objects Nevertheless it is interesting toinspect the predictions of other classifiers (employing additional IRcolors) on our data Akras et al (2019) for example obtained a PNclassification efficiency ofsim50 and a contamination frommimics(Hii stars YSO) ranging from 30 to 40 using these color cuts

JminusH lt 110 W1minusW4ge 787 orJminusH lt 131 KminusW3ge 642

(20)

No unclassified source was found to satisfy both cuts In particularthe second condition (involving WISE colors) of both cuts was notfulfilled by any source passing the first conditionAnderson et al (2012) reported an Hii region classification ef-ficiency better than 95 with a PN contamination around 20requiring the following criteria (at least one of them)

log10(S12S8)lt 03

log10(S22S8)lt 1

log10(S70S12)gt 13

log10(S70S22)gt 08

(21)

We found that 47 sources previously classified as Galactic objectssatisfy the first two criteria while only 2 are passing all criteria(requiring emission at 70 microm) These are likely new Hii region can-didates to be explored and confirmed in follow-up studiesTo make further advances we will need to individually assess thereliability of each radio-IR association (currently quoted in Sec-tion 323 assim60with a match radius of 8) and possibly increasethe sample of pre-classified objects to constrain the parameter spacewith a higher degree of confidence Data validation efforts have al-ready started and will be reported in a future paper

6 SUMMARY

We have analyzed ASKAP Early Science observations of the Scor-pio field (sim40 deg2 σrmsgt200 microJybeam 24times21 angular resolu-tion) at 912 MHz and produced a first catalogue of compact radiosources and their components using the caesar source finder Wesummarize below the main results

(i) The catalogue contains 4144 source components with fluxdensities ranging from 39 Jy down to 03 mJy About sim87 were

EMU Compact radio sources in SCORPIO 19

detected at a significance higher than 5σ From simulations weestimated a catalogue completeness of at least 90above 5mJy Thedifferential source counts are compatible with literature data at 14GHz obtained outside the Galactic plane after taking into accountthe different observation frequency and analysis systematics Thissuggests a majority of extragalactic sources in the catalogue(ii) Through comparison with different radio surveys including

our ATCA observations at 14 - 26 GHz we estimated the spectralindices for a subset of the catalogued sources (780) finding amedianvalue of 〈α〉=minus092 (IQR=068) in agreement with previous workswithin the discussed uncertainties The higher resolution ATCAdata also allowed us to derive a lower limit (sim8) on the fractionof the extended sources present in the catalogue(iii) We cross-matched the source catalogue with different as-

tronomical databases (SIMBAD HASH WISE Hii regions Wolf-Rayet catalogue) to search for possible associations to knownGalac-tic objects Including candidate objects we have found 20 stars (in-cluding 7 YSO) 21 pulsars 38 PNe 67 Hii regions providing newspectral index measurements for a subsample of these detected inboth ASKAP and ATCA observations(iv) The vast majority (gt96) of the catalogued sources are

unclassified likely dominated by extragalactic background sourcesAbout 300 sources have been identified as radio galaxy candidatesfrom morphological considerations The remaining sources wereanalyzed in correlation with existing mid-infrared data in particularwith the AllWISE survey The majority of the unclassified sourcesdo not have IR counterparts at 12 and 22 microm above IRflux sensitivitylevels (sim6mJy)within theASKAPbeam radius About one hundredsources found with IR counterparts were preliminarily classifiedas Galactic objects on the basis of their radio-infrared colors Asummary of the obtained results is reported in Table E1

The source catalogue will be updated before the beginning of theASKAP Galactic plane survey once that the scheduled Early Sci-ence data of the ASKAP Scorpio field with 36 antennas at threedifferent frequencies (920 MHz 1296 MHz 1630 MHz) are avail-able and fully reduced The multi-frequency data and the expectedboost in sensitivity and spatial resolution will enable us to measurethe spectral index for all Scorpio sources including those locatedoutside the ATCA survey and near the current ASKAP detectionthreshold enabling further advances in our classification studies

ACKNOWLEDGEMENTS

TheAustralian SKAPathfinder is part of theAustralia TelescopeNa-tional Facility which is managed by CSIRO Operation of ASKAPis funded by the Australian Government with support from the Na-tionalCollaborativeResearch Infrastructure Strategy Establishmentof the Murchison Radio-astronomy Observatory was funded by theAustralian Government and the Government of Western AustraliaThis work was supported by resources provided by the Pawsey Su-percomputing Centre with funding from the Australian Governmentand the Government ofWestern Australia We acknowledge theWa-jarri Yamatji people as the traditional owners of the ObservatorysiteWe thank the authors of the following software tools and librariesthat have been extensively used for data reduction analysis andvisualization ASKAPsoft (Whiting et al 2019) caesar (Riggi etal 2016 2019) astropy (Astropy Collaboration et al 2013 2018)Root (BrunampRademakers 1996) Topcat (Taylor 2005 2011) ds9(Joye amp Mandel 2003) APLpy (Robitaille amp Bressert 2012)This publication makes use of data products from the Wide-field

Infrared Survey Explorer which is a joint project of the Univer-sity of California Los Angeles and the Jet Propulsion Labora-toryCalifornia Institute of Technology funded by the NationalAeronautics and Space Administration Additionally this researchhas made use of the SIMBAD database operated at CDS Stras-bourg France (Wenger et al 2000) and the VizieR catalogue ac-cess tool CDS Strasbourg France (DOI 1026093cdsvizier) Theoriginal description of the VizieR service was published in Ochsen-bein et al (2000)Some of the authors benefited from grant No 863448 (NEANIAS)of the Horizon 2020 European Commission programme HA ben-efited from grant CIIC 902020 of Universidad deGuanajuatoMex-ico MJM acknowledges the support of the National Science Cen-tre Poland through the SONATABIS grant 201830EST900208

DATA AVAILABILITY

Source tables are provided as online supplementary materialThe following catalogue data products are available in the Zenodorepository25 at httpdoiorg105281zenodo4386692

(i) Source catalogue in tabular format Two asciiFITS tablefileswith a series of summary parameters for each catalogued sourceislands and fitted components respectively Table format (numberof data columns and column description) is detailed in the caesardocumentation at httpscaesar-docreadthedocsio Ad-ditionally we provide an added-value source component cataloguetable (ascii and FITS formats) with extra-information (correctedfluxes radioinfrared cross-match info spectral indices etc) Itsformat is described in Table B1 (available in the online version ofthis paper)

(ii) Source catalogue in Root format A Root26 file storingthe list of catalogued sources and relative components as a cae-sar Source C++ object For each source the summary parame-ters plus detailed information at pixel level are available The de-tailed format is described in the caesar API documentation athttpscaesar-docreadthedocsio

(iii) Source list in region format Two ds927 region files withthe list of catalogued source islands and fitted components respec-tively reported as labelled polygons or ellipses

(iv) Background maps in FITS format Two FITS files withbackground and noise maps obtained in the source finding process

REFERENCES

Anderson C et al [ACES and SDP teams] ASKAP Science ObservationGuide CSIRO version 11 July 2019

Anglada G et al 1998 AJ 116 2953Anglada G et al 2018 Astron Astrophys Rev 26 3Ainsworth R E et al (AMI Consortium) 2012 MNRAS 423 1089Akras S et al 2019 MNRAS 488 3238Anderson L D et al 2012 AampA 537 A1Anderson L D et al 2014 APJSS 212 1AssefR J et al 2018 ApJS 234 23Astropy Collaboration Robitaille T P Tollerud E J Greenfield P Droet-

tboom M Bray E Aldcroft T et al 2013 AampA 558 A33Astropy Collaboration Price-Whelan A M Sipőcz B M Guumlnther H M

Lim P L Crawford S M Conseil S et al 2018 AJ 156 123

25 httpszenodoorg26 httpsrootcernch27 httpds9siedu

20 S Riggi et al

Brown A GA et al (Gaia collaboration) 2018 AampA 616 A1Brun R and Rademakers F Proc AIHENPrsquo96 Workshop Lausanne Sep

1996 Nucl Inst amp Meth in Phys Res A 389 (1997) 81Butler A et al 2018 AampA 620 A3Caswell J L Haynes R F 1983 AuJPh 36 361Cavallaro F et al 2018 MNRAS 473 1685Chang C et al 2008 ApJ 682 1177Churchwell E et al 2009 PASP 121 877Clarke A O et al 2020 AampA 639 A84Condon J J 1997 PASP 109 166Condon J J 1984 AJ 287 461Condon J J et al 1998 AJ 115 1693Condon J J Kaplan D L 1998 ApJS 117 361Cutri R MWright E L Conrow T et al 2013 Explanatory Supplement

to the AllWISE Data Release Products Tech repDale D A et al 2012 ApJ 745 95Franzen T M O et al 2015 MNRAS 453 4020Franzen T M O et al 2019 PASA 36 E004Fragkou V et al 2018 MNRAS 480 2916Gaylard M J MacLeod G C 1993 MNRAS 262 43Green D A 2019 Journal of Astrophysics and Astronomy 40 36Han J et al 2016 Res Astron Astrophys 16 159Hopkins A M et al 2015 PASA 32 E037Hopkins A M et al 2003 AJ 125 465Hurley-Walker N et al 2017 MNRAS 464 1146Hurley-Walker N et al 2019 PASA 36 E047Iglesias-Paacuteramo J et al 2007 ApJ 670 279Ingallinera A et al 2014 MNRAS 437 3626Ingallinera A et al 2017 Proc of the International Astronomi-

cal Union 12 (S331) 345-350 doi httpsdoiorg101017S1743921317004574

Ingallinera A et al 2019 MNRAS 490 5063Intema H T et al 2017 AampA 598 A78Jankowski F et al 2018 MNRAS 473 4436Jankowski F et al 2019 MNRAS 484 3691Jeyakumar S 2016 MNRAS 458 3786Johnston S Kerr M 2018 MNRAS 474 4629Johnston S et al 2008 Experimental Astronomy 22 151Johnston S et al 1992 MNRAS 255 401Joye W A Mandel E 2003 in Astronomical Data Analysis Software and

Systems XII eds H E Payne R I Jedrzejewski amp R N Hook ASPConf Ser 295 489

Katgert P Oort M J A amp Windhorst R A 1988 AampA 195 21Kimball A E and Ivezić Z 2008 AJ 136 684Kramer M et al 2003 MNRAS 342 1299Kwok S 2000 The Origin and Evolution of Planetary Nebulae Cambridge

University PressLacy M et al 2020 PASP 132 035001Leto P et al 2006 AampA 458 831Leto P et al 2017 MNRAS 467 2820Leto P et al 2018 MNRAS 476 562Manchester R N Hobbs G B Teoh A amp Hobbs M 2005 AJ 129

1993Maron O et al 2000 AampA 147 195Molinari S et al 2016 AampA 591 A149Murphy T et al 2007 MNRAS 382 382McConnell D et al 2016 PASA 33 e042Michałowski M J et al 2008 ApJ 672 817Michałowski M et al 2010 AampA 514 A67Milne D K Aller L H 1975 AampA 38 183Milne D K Aller L H 1982 AampAS 50 209Nikutta R et al 2014 MNRAS 442 3361Norris R P et al 2011 PASA 28 215Ochsenbein F Bauer P Marcout J 2000 AampAS 143 23Ofek E O Frail D A 2011 ApJ 737 45Parker Q A et al 2016 J Phys Conf Ser 728 032008Pilbratt G L et al 2010 AampA 518 L1Pottasch S R Planetary nebulae - A study of late stages of stellar evolution

by SR Pottasch Dordrecht D Reidel Publishing Co (Astrophysics andSpace Science Library Volume 107) 1984 335 p

Prandoni I et al 2001 AampA 365 392PurcellC R et al 2013 ApJS 205 1Retana-Montenegro E et al 2018 AampA 620 A74Reynolds S P 1986 ApJ 304 713Riggi S et al 2016 MNRAS 460 1486Riggi S et al 2019 PASA 36 E037Robitaille T Bressert E 2012 APLpy Astronomical Plotting Library in

Python Astrophysics Source Code Library (ascl1208017)Rosslowe C K Crowther P A 2015 MNRAS 447 2322Scaife AM M 2012 Astronomical Review 7 26Schinckel A E et al 2012 in Ground-based and Airborne Telescopes IV

p 84442A doi10111712926959Silva L et al 1998 ApJ 509 103Suarez O et al 2006 AampA 458 174Taylor M B 2005 in Shopbell P Britton M Ebert R eds Astronomical

Society of the Pacific Conference Series Vol 347 Astronomical DataAnalysis Software and Systems XIV p 29

Taylor M 2011 TOPCAT Tool for OPerations on Catalogues And TablesAstrophysics Source Code Library (ascl1101010)

Thorat K et al 2013 ApJ 762 16Umana G et al 2015 MNRAS 454 902Umana G et al 2020 in prepUrquhart J S et al 2009 AampA 501 539Urquhart J S et al 2018 MNRAS 473 1059van de Steene G C M Pottasch S R 1993 AampA 274 895Wang Y et al 2018 AampA 619 A124Wendker H J 1995 AampAS 109 177March 2001 update of the catalogue

CDS VIII99Wenger M et al 2000 AampAS 143 9Werner M W et al 2004 ApJS 154 1Wilson W E et al 2011 MNRAS 416 832Windhorst R et al 1990 in Evolution of the Universe of Galaxies Edwin

Hubble Centennial Symposium ed R G Kron ASP Conf Ser 10389

Whiting M et al ASKAP Science Data Processor software httpsbitbucketcsiroauscmcasssoftaskapsoftgit

Wright E L et al 2010 AJ 140 1868Zajaček M et al 2019 AampA 630 A83Zhang Y et al 2004 MNRAS 351 935Zhao R S et ak 2019 ApJ 874 64

This paper has been typeset from a TEXLATEX file prepared by the author

APPENDIX A SOURCE FINDER PARAMETERS

Table A1 Main parameters used in the caesar source finder to build the source catalogue More details are available in Riggi et al 2019 and in the caesaronline documentation at httpscaesar-docreadthedocsioenlatest

Stage Par Name Par Value Description

Background method local Compute local background and noise mapsfinder bkgEstimator median Estimator used to compute background level

rmsEstimator MAD Estimator used to compute background noise rmsbkgBoxSize 10timesbeam Size of sampling box for local bkg calculationbkgGridSize 02timesbox size Size of grid for local bkg interpolation

Source σseed 5 Seed threshold in blob finding given as number of sigmas above backgrounddetection σmerge 25 Merge threshold in blob finding given as number of sigmas above background

npix 5 Minimum number of pixels to consider a blob as source candidate∆σ 05 Decrease step applied to the seed threshold at each iteration Effective only when nitersgt1niters 2 Number of iterations to be performed in compact source searchσnested

seed 5 Seed threshold for nested blob searchσnested

merge 25 Merge threshold for nested blob searchnestedBlobMinScale 1timesbeam Nested blob search min scalenestedBlobMaxScale 2timesbeam Nested blob search max scalenestedParentSourceMinDist 2 Minimum distance in pixels between nested and parent blob centroids below which

nested source is skippedmaxMatchingPixFraction 05 Max fraction of matching pixels between nested and parent blob above which

nested source is skipped

Source method peak search+ Search blended components around detected peaks using a multiscale blob detectiondeblending blob detection algorithm based on a Watershed segmentation of the Laplacian of Gaussian (LoG)

filtered mappeakMinKernSize 3 Min dilation kernel size (in pixels) used to detect peakspeakMaxKernSize 7 Max dilation kernel size (in pixels) used to detect peakspeakKernMultiplicityThr 1 Requested peak multiplicity across different dilation kernelspeakShiftTolerance 2 Shift tolerance (in pixels) used to compare peaks in different dilation kernelspeakZThrMin 1 Minimum peak flux significance below which peak is skipped

Source method N-gaus fit Fit a mixture of 2D gaussians to source islandsfitting nBeamsMaxToFit 100 Threshold above which source fitting is not performed

nMaxComponents 5 Max number of fit components

APPENDIX B CATALOGUE FORMAT

Table B1 Format of the added-value catalogue for ASKAP Scorpio source components (one component per table row 135 data columns per component)More details of each data column are available in the caesar online documentation at httpscaesar-docreadthedocsioenlatest

Col Num Name Unit Description1 sname - Source island name2 npix - Number of pixels in island3 compId - Component id4 sname_iau - Component name in IAU format5-6 (xy) - Component x y position7-8 (∆x∆y) - Component x y position error9-10 (radec) deg Component RA and Dec11-12 (∆ra∆dec) deg Component RA and Dec errors13 ν GHz Data frequency from map header

14-15 (Speak ∆Speak) Jybeam Component peak brightness and its error16-17 (S∆S) Jy Component (corrected) flux density and its total error18-19 (Sisland ∆Sisland ) Jybeam Island flux brightness and its error20-22 (abθ ) --deg Component ellipse pars28

23-25 (∆a∆b∆θ ) - Component ellipse par errors26-28 (abθ )wcs deg Component ellipse pars in J2000 coords29-31 (∆a∆b∆θ )wcs deg Component ellipse par errors in J2000 coords32-34 (abθ )wcs

deconv deg Component ellipse pars in J2000 coords deconvolved by beam35 EEbeam - Ratio of fitted and beam ellipse eccentricities36 AAbeam - Ratio of fitted and beam ellipse areas37 ψ deg Angle between fitted major axis and beam major axis38 microbkg Jybeam Background averaged over island pixels39 σrms Jybeam Background noise averaged over island pixels

40-41 χ2ndf - Source island fit chisquare and degrees of freedom42 F1 - Source fit quality flag 0=BAD 1=LOW 2=MEDIUM 3=HIGH43 F2 - Component sourceness flag 1=REAL 2=CANDIDATE 3=FAKE44 F3 - Component morphology flag 1=COMPACT 2=POINT-LIKE 3=EXTENDED 4=COMPACT-EXTENDED45 sname_atca - Name of matched source island in ATCA full band map46 compId_atca - ATCA component id47 npix_atca - Number of pixels in ATCA island

48-49 (ra_atcadec_atca) deg ATCA component RA and Dec50-51 (∆ra_atca∆dec_atca) deg ATCA component RA and Dec errors52-53 (Satca

peak ∆Satcapeak) Jybeam ATCA component peak brightness and its error

54 microatcabkg Jybeam Background averaged over ATCA island pixels

55 σatcarms Jybeam Background noise averaged over ATCA island pixels

56-57 (Satca∆Satca) Jy ATCA component (corrected) flux density and its total error58-71 (Satca

k ∆Satcak ) Jy Component (corrected) flux densities and their total errors for ATCA subbands (k=1 7)

72 matchFlag_atca - ASKAP-ATCA match flag -1=BADNO-MATCH 0=UNCERTAIN 1=SINGLE 2=DOUBLE73 sname_mgps - Matched source name in MGPS

74-75 (ra_mgpsdec_mgps) deg MGPS source RA and Dec76-77 (Smgps∆Smgps) Jy MGPS flux and its error78 matchFlag_mgps - ASKAP-MGPS match flag79 sname_nvss - Matched source name in NVSS

80-81 (ra_nvssdec_nvss) deg NVSS source RA and Dec82-83 (Snvss∆Snvss) Jy NVSS flux and its error84 matchFlag_nvss - ASKAP-NVSS match flag85 sname_tgss - Matched source name in TGSS

86-87 (ra_tgssdec_tgss) deg TGSS source RA and Dec88-89 (Stgss∆Stgss) Jy TGSS flux and its error90 matchFlag_tgss - ASKAP-TGSS match flag91 sname_gleam - Matched source name in GLEAM

92-93 (ra_gleamdec_tgss) deg GLEAM source RA and Dec94-95 (Sgleam∆Sgleam) Jy GLEAM flux and its error96 matchFlag_gleam - ASKAP-GLEAM match flag97 dwise Distance (rlt24) between ASKAP and matched AllWISE source98 sname_wise - Matched source name in AllWISE

99-100 (ra_wisedec_wise) deg AllWISE source RA and Dec101-108 (Swise

k ∆Swisek ) Jy AllWISE source flux and errors in all bands (k=W1W2W3W4)

Continued on next page

28 Position angle is measured counterclock-wise from North

Table B1 ndash continued from previous pageCol Num Name Unit Description109-114 (S2MASS

k ∆S2MASSk ) Jy 2MASS source flux densities and their errors in all bands (k=JHK)

115 sname_glimpse - Matched source name in GLIMPSE116-117 (ra_glimpsedec_glimpse) deg GLIMPSE source RA and Dec118-119 (Sglimpse∆Sglimpse) Jy GLIMPSE 8 microm flux and its error120 sname_higal - Matched source name in Hi-GAL

121-122 (ra_higaldec_higal) deg Hi-GAL source RA and Dec123-124 (Shigal ∆Shigal ) Jy Hi-GAL 70 microm flux and its error125 resolvedFlag - Resolved flag 0=POINT-SOURCE 1=RESOLVED 2=ATCA-RESOLVED

126-127 (α ∆α) - Spectral index and its error128-129 (χ2ndf )sed - Radio spectrum fit χ2 and degrees of freedom130 sedRobustFitFlag - Spectral data fit flag 0=STANDARD 1=ROBUST131 objClassStrId - Matched astro object class string id132 objClassId - Matched astro object class integer id

0=UNKNOWN -1=MULTI-ID 1=STAR 2=GALAXY 3=PN 6=HII 23=PULSAR 24=YSO133 objClassSubId - Matched astro object class integer sub-category id134 objConfirmedFlag - Matched astro object flag 0=CANDIDATE 1=CONFIRMED135 objName - Matched astro object name(s) (separated by semicolons for multiple matches)

APPENDIX C SCORPIO ATCA MAP

Figure C1 21 GHzmosaic of the Scorpio region observed with ATCA The yellow contour delimits the mosaic area considered for source catalogue extractionwith caesar

APPENDIX D SOURCE CATALOGUE MAP

Figure D1 Scorpio ASKAP region in J2000 sky coordinates (solid black line contour) Colored markers represent the ASKAP source components that wereclassified on the basis of their morphology or match association to objects reported in astronomical databases (see text) The sample includes both confirmedand candidate objects Small gray dots represent the unclassified source components

APPENDIX E SOURCE TABLES

Table E1 Summary of source component classification information Column 3 reports the total number of source components (ntot ) classified in a givencategory including number of confirmed (ncon f irmed ) and candidate (ncandidate) objects which are reported in columns 4 and 5 respectively The number ofsource components (nα ) found with spectral index information is reported in column 6 Column 7 reports the number of source components (nIR) found withinfrared counterparts in all four AllWISE bands Column 8 reports the number of source components (nα+IR) found with both measured spectral indices andIR data

Type (1) Class (2) ntot (3) ncon f irmed (4) ncandidates (5) nα (6) nIR (7) nα+IR (8)

GALACTIC

STAR 9 8 1 5 7 5WR STAR 0 minus minus minus minus minusYSO 6 2 4 4 5 3PULSAR 21 21 0 3 0 0PN 36 26 10 11 29 10Hii REGIONS 64 32 32 7 34 3UNKNOWN 104 5 99 32 101 31

EXTRAGALACTIC RG 415 262 153 74 23 5UNKNOWN 185 0 185 37 185 37

UNCLASSIFIED minus 3304 minus minus 607 0 0

ALL minus 4144 356 484 780 384 94

TableE2

Listof

SIMBA

Dstellaro

bjectsfoun

dassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arlySc

iencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5representthe

spectra

ltyp

e(oro

bjectclassificatio

n)and

objectnamerepo

rtedin

theSIMBA

Dcatalogu

eMissing

entries

incolumn4aregeneric

ally

classifie

das

rsquostarrsquoin

SIMBA

DC

olum

n6indicatesifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nin

SIMBA

Disrepo

rtedin

columns

7-8in

J200

0coordinatesColum

n9representsthedistance

betweentheob

ject

andASK

APmeasuredsource

positio

nsin

arcsecC

olum

n10

-13arethesource

fluxdensity

and

totalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

n14

-15representsthemeasuredspectra

lind

exandits

erroro

btainedfrom

ASK

APATC

AN

VSS

andMGPS

data(w

henavailable)

Source

name(1)

Island

(2)

Com

p(3)

Type

(4)

Objnam

e(5)

C(6)

RA(7)

Dec

(8)

d(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1659354-421931

S2609

1YSO

SSTGLMC

G3437018+000861

02548971

minus423250

12

983

127

1291

042

021

006

J1653029-414508

S3496

1-

IRAS

16495-4140

12532622

minus417524

06

439

062

766

025

058

008

J1706248-415654

S1694

1YSO

2MASSJ17062471-4156536

02566030

minus419482

17

341

064

295

020

002

015

J1704513-422541

S1951

1YSO

SSTGLMC

G3442155-007460

02562133

minus424281

15

686

082

757

029

017

009

J1656547-412824

S2931

1-a

IRAS

16534-4123

12542278

minus414737

11

1257

131

1185

038

minus006

007

J1656400-401334

S2933

1YSO

[MHL2007]

G3450052+018209

10

2541658

minus402261

30

12532

1282

20442

633

043

008

J1658474-434741

S2740

1-

HD326586

12546974

minus437943

21

2320

239

--

--

J1658031-432615

S2820

1YSO

SSTGLMC

G3426544-003827

12545129

minus434375

05

379

068

--

--

J1650543-440704

S3841

1-c

IRAS

16472-4401

12527268

minus441173

21

2296

244

--

--

J1650406-432812

S3867

1AB

2MASSJ16504054-4328122

02526690

minus434701

11

272

067

--

--

J1652382-424925

S3587

1B8

HD326392

12531596

minus428243

23

075

032

--

--

J1712222-423041

S968

1YSO

b2MASSJ17122205-4230414

02580919

minus425115

28

298

038

--

--

J1656419-401520

S2934

1-

TYC

7872-1355-1

12541740

minus402552

23

781

255

144

057

minus20

09

J1708226-400527

S1348

1YSO

MSX6CG3464809+001320

12570948

minus400906

20

669

233

--

--

J1707417-403125

S1455_N1

1-d

2MASSJ17074166-4031240

12569236

minus405233

08

36530

3839

--

--

J1703437-400350

S2012

1-

TYC

7873-953-1

12559318

minus400648

35

3581

366

--

--

J1656487-390541

S2886

1er(M

3Ve+

M4V

e)CD-3811343

12542024

minus390939

34

356

073

--

--

J1709101-393433

S1212

1-

IRAS

17056-3930

12572918

minus395753

18

2786

296

--

--

J1709040-391949

S1216

1-

IRAS

17056-3916

12572664

minus393301

11

710

087

--

minus165

082

J1716324-392718

S324

1Cle

Cl

NGC

6318

PCA

7229

12591343

minus394549

29

14869

1527

--

minus074

010

aTh

issource

matches

also

with

acandidatePN

(Pre

38)

bTh

issource

matches

also

with

aconfi

rmed

PN(PM1-131)

cTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G341314+00190)

dTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G346056-00020)

eTh

issource

matches

also

with

aconfi

rmed

Hiiregion

(G347921-00763)

TableE3

Listof

pulsarsintheATN

Fdatabase

associated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquo

stands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicate

theisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4indicatestheob

ject

nameas

repo

rtedin

thedatabaseT

heob

ject

positio

nfrom

thepu

lsar

database

isrepo

rtedin

columns

5-6in

J200

0coordinatesColum

n7representsthedistance

betweentheATN

Fpu

lsar

database

positio

nandtheASK

APsource

centroid

inarcsec

Colum

ns8-11

arethesource

fluxdensity

andtotalerror

inmJy

at91

2MHzand21GHzrespectiv

elyColum

ns12

-13representsthespectra

lind

exandits

errorreportedin

theATN

Fdatabase

(whenavailable)

and

takenfrom

measurementsat

5GHz(Zhaoet

al2

019)

728

1382

and31

00MHz(Jankowskie

tal20

18)a

nd14GHz(H

anet

al2

016)C

olum

ns14

-15representsthespectra

lind

exandits

errorm

easuredusing

ASK

APATC

Aandliteraturedataat08

14and30GHzrepo

rtedin

theATN

Fcatalogu

e(Jankowskietal20

18K

ramer

etal2

003

John

ston

andKerr2

018

John

ston

etal1

992)

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

RA(5)

Dec

(6)

d(7)

S 912

(8)

∆S 9

12(9)

S 210

0(10)

∆S 2

100(11)

αat

nf(12)

∆α

atn

f(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1654238-414026

S3298

1J1654-4140

2535979

minus416733

51

233

033

036

003

--

minus200

016

J1707218-405355

S1521

1B1703-40a

2568405

minus408989

18

2902

331

333

042

minus200

010

minus282

011

J1702526-412848

S2180

1J1702-4128

2557188

minus414801

18

191

044

129

008

minus020

020

minus041

010

J1702363-421708

S2246

1J1702-4217

2556518

minus422837

68

713

085

--

minus120

019

-b-

J1649207-434923

S4085

1J1649-4349

2523351

minus438228

43

259

052

--

--

--

J1711105-432256

S1142

1J1711-4322

2577940

minus433814

26

083

041

--

--

--

J1709413-440115

S1371

1J1709-4401

2574225

minus440198

40

072

025

--

--

--

J1707403-434112

S1608

2J1707-4341

2569171

minus436867

22

141

027

--

minus185

050

--

J1705357-433112

S1887

1J1705-4331

2563996

minus435204

31

114

024

--

--

--

J1702529-442805

S2279

1J1702-4428

2557192

minus444675

48

154

047

--

--

--

J1646554-430802

S4499

1J1646-4308

2517304

minus431353

56

127

029

--

--

--

J1653402-424904

S3421

1J1653-4249

2534176

minus428176

03

293

051

--

100

060

--

J1651488-424611

S3699

1B1648-42c

2529533

minus427697

06

5722

582

--

minus203

010

--

J1706044-431020

S1804

1J1706-4310

2565188

minus431725

14

043

019

--

--

--

J1702269-431042

S2288

1J1702-4310

2556123

minus431778

23

131

032

--

minus110

030

--

J1650133-412636

S3915

1J1650-4126

2525549

minus414427

29

071

028

--

--

--

J1705299-395057

S1736

1J1705-3950

2563743

minus398497

23

238

044

--

minus084

030

--

J1717524-410315

S284

1B1713-40d

2594676

-410547

30

285

055

--

minus140

050

--

J1716423-400525

S339

1J1716-4005

2591753

minus400908

38

378

155

--

--

--

J1715411-403421

S459

1J1715-4034

2589208

minus405728

18

423

070

--

minus220

020

--

J1655383-384414

S3051

1J1655-3844

2539111

minus387358

75

097

041

--

--

--

aAlternativenameJ1707-4053

bSp

ectra

lindex

notestim

ated

astheASK

APflu

xisincludingflu

xfrom

aneighbor

background

source

thatcannot

beproperly

accountedas

inATC

Amap)

cAlternativenameJ1651-4246

dAlternativenameJ1717-4054

TableE4

Listof

HASH

PNeassociated

toASK

APScorpiosourcesColum

n1indicatesthesource

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

here

rsquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

n2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

n4and5indicatetheob

jectnameas

repo

rtedin

theHASH

catalogu

eandifthe

objectisconfi

rmed

(1)o

racand

idate(0)Th

eob

jectpo

sitio

nfrom

HASH

isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthepo

sitio

noff

setb

etweentheHASH

PNandtheASK

APsource

inarcsecC

olum

n9isthePN

major

diam

eter

inarcsec

asrepo

rtedin

thedatabase

(whenavailable)C

olum

ns10

-11and16

-17arethemeasuredsource

fluxdensity

andtotalerror

at91

2MHz(A

SKAP)

and21GHz

(ATC

A)respectivelyColum

ns12

-13arethemeasuredsource

fluxdensity

anderrora

t843

MHzrepo

rtedin

theHASH

catalogu

eandtakenwith

theMOST

(Murphyetal2

007)C

olum

ns14

-15arethemeasured

source

fluxdensity

anderrora

t14

GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theVLA

(Con

donetal1

998

Con

donandKaplan19

98)Colum

ns18

-19arethemeasuredsource

fluxdensity

anderror

at48GHzrepo

rtedin

theHASH

catalogu

eandob

tained

with

theATC

A(van

deSteene

andPo

ttasch19

93)Colum

n20

and21

aretheestim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

(whenavailable)

Source

name(1)

Island

(2)Com

p(3)Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

d ma

j(9)S 9

12(10)

∆S 9

12(11)

S 843

(12)

∆S 8

43(13)

S 140

0(14)

∆S 1

400(15)

S 210

0(16)

∆S 2

100(17)

S 480

0(18)

∆S 4

800(19)

α(20)

∆α(21)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

(mJy)

J1655006-405534

S3192

1IRAS

16515-4050

12537518minus409263

23

-582

067

--

--

1169

037

--

-a-

J1654434-404146

S3227

1MPA

J1654-4041

02536799minus406963

29

150

983

130

--

--

746

030

--

minus058

010

J1654512-414356

S3226

1PHR

J1654-4143b

02537125minus417305

68

650

986

133

--

--

385

028

--

minus103

017

J1656152-403641

S2998

1Pre

11

02540629minus406120

18

60

1297

140

--

--

1272

042

--

minus016

010

J1653000-403047

S3479

1DGPK

20

2532483minus405129

57

40

248

063

--

--

337

046

--

064

032

J1655472-414840

S3101

1MGE

3436641+009584

02539454minus418117

44

210

618

095

--

--

-c-

--

--

J1700201-415043

S2499

1MGE

3441648+002733

02550833minus418453

23

240

3785

394

2790

320

--

-c-

--

--

J1656341-434615

S3028

1PM1-119

12541416minus437708

12

47

11379

1147

8530

350

--

--

25100

400

060

002

J1707306-442250

S1651

1WRAY

16-251

12568774minus443806

09

150

3366

345

3280

240

--

--

--

--

J1706226-441311

S1811

1Pe1-8

12565940minus442194

11

230

7739

786

8030

320

--

--

--

--

J1705389-435620

S1901

1Vd1-9

12564121minus439390

05

-975

102

--

--

--

--

--

J1642187-421445

S5124

1PHR

J1642-4214

12505775minus422459

11

100

600

076

--

--

--

--

--

J1657239-413758

S2867

1H

1-5

12543489minus416327

18

52

14789

1491

11480

410

--

--

--

--

J1653555-414400

S3365

1PHR

J1653-4143

12534804minus417333

31

150

1386

146

1870

190

--

--

--

--

J1653314-423923

S3438

1H

1-3

12533806minus426562

12

190

3386

356

3240

250

--

--

--

--

J1710274-415249

S1151

1H

1-7

12576141minus418804

09

106

18777

1899

15900

560

--

--

--

--

J1707572-430905

S1552

1PHR

J1707-4309

12569874minus431514

34

160

774

083

--

--

--

--

--

J1706592-424109

S1637

1H

1-6

12567464minus426859

15

140

3462

352

3070

180

--

--

--

--

J1706197-431533

S1770

1Kn98

02565817minus432592

14

40

711

076

--

--

--

--

--

J1715470-422406

S560

1CBF

31

2589450minus424017

25

76

361

046

--

--

--

--

--

J1715160-430354

S663

1MPA

J1715-4303

12588163minus430649

18

70

1041

111

1760

220

--

--

--

--

J1714252-414433

S697

1Kn102

02586042minus417428

24

-248

042

--

--

--

--

--

J1712222-423041

S968

1PM1-131e

12580918minus425115

31

22

298

038

--

--

--

135

01

091

008

J1644241-400321

S4880

1SuWt

31

2511006minus400557

06

319

235

052

--

450

070

--

--

152

088

J1656547-412824

S2931

1Pre

38f

02542275minus414733

20

60

1257

131

--

--

--

--

--

J1711422-403527

S957

1Mo16

02579251minus405910

24

180

2135

225

--

--

--

--

--

J1659290-391500

S2535

1PreRo2

02548708minus392503

09

60

557

103

--

--

--

--

--

J1651339-400256

S3701

1Vd1-5

12528902minus400488

32

116

562

062

--

370

060

--

--

minus097

064

J1650203-403004

S3878

1MPA

J1650-4030

12525840minus405009

20

90

509

058

--

--

--

--

--

J1705107-405310

S1829

1IC4637

12562938minus408857

28

189

23776

2412

21580

730

--

--

401d

-033

004

J1714494-400608

S552

1PHR

J1714-4006

12587055minus401025

11

200

579

095

2560

680

--

--

--

--

J1713109-401556

S766

1MPA

J1713-4015

12582950minus402656

20

70

1032

118

--

--

--

--

--

J1656402-391237

S2909

1MPA

J1656-3912

12541671minus392102

10

40

416

054

--

490

080

--

--

038

069

J1655520-390023

S3020

1PHR

J1655-3900

12539663minus390058

33

120

182

048

--

--

--

--

--

J1654275-384412

S3228

1Vd1-6

12536138minus387362

30

160

1344

159

1300

210

1120

060

--

--

minus037

024

J1650256-390819

S3853

1Vd1-4

12526055minus391386

40

50

331

040

--

420

060

--

--

055

062

J1649330-392109

S3994

1Vd1-3

12523870minus393525

27

-312

038

--

500

050

--

--

110

052

J1715517-393309

S388

1RPZM

81

2589651minus395523

10

-1033

132

--

1230

180

--

--

041

064

aIRAS

16515-4050radiospectru

mcannot

bewelld

escribed

byasinglepower-la

wmodelandwas

fittedwith

athermalfree-freeem

ission

model(see

text)

bPHR

J1654-4143associated

toatwo-component

island

inboth

ASK

APandATC

AmapsTh

eflu

xconsidered

forthe

spectra

lindex

isthesum

ofboth

components

cMGE

3436641+009584andMGE

3441648+002733PN

candidates

aredetected

andresolved

inATC

Aobservations

at21GHzWedidnotreportthe

fluxdensity

measurementsin

theTable

dTh

isflu

xmeasurementw

asobtained

with

Parkes

atareferencefrequencyof

5000

MHz(M

ilneandAller1

975)N

oflu

xerrorisp

rovidedin

thereferencepaperTh

espectra

lindex

fitwas

thus

perfo

rmed

assumingno

fluxerrorsin

allm

easurementsA

nadditio

nal

fluxmeasurement(

S=117mJy)o

btainedwith

Parkes

at147GHz(M

ilneandAller1

982)

isreporte

din

theHASH

cataloguebutn

otshow

nin

theTableandnotincludedin

thefit

eTh

issource

matches

also

with

acandidateYSO

(2MASSJ17122205-4230414)

fTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16534-4123)

TableE

5Listof

WISEHiiregion

sfou

ndassociated

toASK

APScorpiosourcesColum

n1indicatesthe

source

namefollo

wingEM

UIAU-app

rovedconventio

n(EMU

ESJHHMMSSs+DDMMSS)w

herersquoErsquostands

forE

arly

Sciencesurvey

andrsquoSrsquofor

sourceC

olum

ns2and3indicatetheisland

nameandcompo

nent

idassign

edby

thecaesar

finderColum

ns4and5indicatetheob

jectnameas

repo

rtedin

theWISEcatalogu

eandiftheob

ject

isconfi

rmed

(1)o

racand

idate(0)Th

eob

ject

positio

nfrom

And

ersonet

al2

014isrepo

rtedin

columns

6-7in

J200

0coordinatesColum

n8representsthedistance

betweensource

centroids

Colum

n9istheradius

oftheHiiregion

asrepo

rtedin

theWISEcatalogu

eColum

ns10

-13arethemeasuredsource

fluxdensity

andtotale

rror

at91

2MHzand21GHzrespectiv

elyColum

ns14

and15

arethe

estim

ated

spectra

lind

exandits

errorfrom

ASK

APATC

Aandliteraturedata

Source

name(1)

Island

(2)

Com

p(3)

Objnam

e(4)

C(5)

RA(6)

Dec

(7)

d(8)

Radius(9)

S 912

(10)

∆S 9

12(11)

S 210

0(12)

∆S 2

100(13)

α(14)

∆α(15)

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1704262-413932

S1977

1G344781-00217

12561087

minus416577

45

100

237

086

178

018

minus056

040

J1701359-421727

S2365

1G343961-00183

12553994

minus422866

150

54176

062

115

a070

minus051

114

J1700582-423000

S2454

1G343721-00223

02552426

minus425005

23

282697

320

2067

069

minus014

008

J1656582-401456

S2892

1G345029+01764

12542423

minus402424

236

100

503

205

299

a053

minus062

070

J1707154-405952

S1547

1G345631-00234

12568101

minus409903

306

57431

099

487

027

minus008

b016

J1707126-413038

S1566

1G345211-00538

02567978

minus415087

172

503435

360

2418a

102

-c-

J1705048-405806

S1843

1G345402+00103

02562710

minus409698

64

23950

182

289

029

minus154

025

J1705113-412905

S1846

1G345004-00224

02562968

minus414845

17

652644

308

6139

193

104

007

J1658421-424732

S2721_N1

1G343234-00078

12546776

minus427934

86

218965

974

4023

156

minus012

d007

J1705220-413142

S1831

2G344989-00277

12563410

minus415286

28

27658

125

458

021

minus038

011

J1705220-413142

S1831

1G344993-00265

e1

2563304

minus415175

79

411373

181

-e-

--

J1657372-430328

S2874

1G342903-00083

12544031

minus430565

87

506832

695

-f-

--

J1655382-415104

S3114

1G343615+00955

12539079

minus418520

57

882840

351

-g-

--

J1650543-440704

S3841

1G341314+00190

h1

2527273

minus441177

32

222296

244

--

--

J1654571-431956

S3250

1G342384+00122

02537382

minus433330

31

201919

223

--

--

J1655482-440348

S3152

1G341912-00456

12539505

minus440633

07

221388

168

--

--

J1652526-435422

S3559_N2

1G341701+00050

12532187

minus439075

53

5926862

2715

--

--

J1652424-442641

S3599

1G341271-00265

02531793

minus444404

185

733153

527

--

--

J1652263-435840

S3627

1G341598+00068

02531086

minus439757

79

78956

167

--

--

J1650562-440930

S3840

1G341286+00159

12527356

minus441591

50

86302

082

--

--

J1650328-435407

S3889

1G341438+00383

02526299

minus438984

279

84579

142

--

--

J1650262-441730

S3914

1G341121+00141

12526067

minus442968

205

44412

140

--

--

J1650011-440502

S3967

1G341238+00335

12525043

minus440834

15

72561

142

--

--

J1700010-420837

S2525_N1

1G343912+00116

12550404

minus421413

90

9129154

3048

--

--

J1659498-424710

S2587

1G343367-00236

02549599

minus427873

102

611865

348

--

--

J1659210-423239

S2622_N1

1G343503-00016

02548390

minus425444

60

3238056

4450

--

--

J1659041-424139

S2675_N3

1G343353-00068

02547673

minus426938

18

2920769

2121

--

--

J1657517-423517

S2813

1G343297+00173

02544632

minus425881

81

303125

338

--

--

J1657531-423543

S2813

2G343293+00162

02544715

minus425981

107

172222

288

--

--

J1657485-430427

S2843

1G342916-00120

12544538

minus430693

178

82668

102

--

--

J1657063-430519

S2949

1G342819-00029

02542736

minus430885

93

26299

063

--

--

J1656030-430443

S3091

1G342705+00126

02540104

minus430806

104

492289

242

--

--

J1656007-430418

S3091

2G342701+00134

12539973

minus430781

304

30509

100

--

--

J1655098-432135

S3214_N1

1G342388+00074

12537920

minus433594

36

687820

835

--

--

J1704135-421953

S2030

1G344224-00594

12560567

minus423291

78

9075809

7640

--

--

J1711413-411858

S993

1G345864-01103

02579216

minus413206

167

531335

200

--

--

J1710286-395312

S1063

1G346884-00069

02576188

minus398865

10

47699

179

--

--

J1709416-400727

S1168

1G346603-00088

12574208

minus401239

99

47397

131

--

--

J1709451-404741

S1194

1G346071-00498

02574369

minus407948

44

99228

093

--

--

J1709319-403326

S1210

1G346233-00322

02573775

minus405599

213

541441

272

--

--

J1709094-400817

S1238

1G346529-00013

12572843

minus401386

168

798647

1068

--

--

J1709378-413211

S1247

1G345462-00920

02574073

minus415359

19

343419

1950

--

--

J1708361-400345

S1318

1G346530+00115

12571509

minus400614

38

413499

615

--

--

J1707553-401820

S1419

1G346255+00073

12569785

minus403063

82

26306

141

--

--

J1707543-403135

S1426

1G346078-00057

12569770

minus405265

31

6021896

2301

--

--

J1707524-402226

S1427

1G346196+00040

12569679

minus403739

21

30985

123

--

--

J1707340-400302

S1450

1G346423+00279

02568956

minus400490

146

322129

246

--

--

J1707417-403125

S1455_N1

1G346056-00020

i1

2569207

minus405219

118

6836530

3839

--

--

J1707302-395651

S1457

1G346496+00352

02568754

minus399462

51

352353

251

--

--

J1707463-404635

S1458

1G345863-00188

12569455

minus407771

101

27678

240

--

--

Contin

uedon

nextpage

TableE5

ndashcontinuedfro

mprevious

page

Source

name

Island

Com

pObjnam

eC

RA

Dec

dRadius

S 912

∆S 9

12S 2

100

∆S 2

100

α∆

α

(prefix

=EMUES

)name

id(deg)

(deg)

(arcsec)

(arcsec)

(mJy)

(mJy)

(mJy)

(mJy)

J1707443-404120

S1462

1G345927-00129

12569338

minus406905

64

361344

190

--

--

J1707296-403432

S1483

1G345997-00024

02568773

minus405710

213

552200

319

--

--

J1707203-402938

S1505

1G346039+00048

12568358

minus404947

51

30629

145

--

--

J1707017-401029

S1525

1G346259+00288

12567564

minus401741

35

382873

331

--

--

J1707045-403711

S1549

1G345909+00013

02567688

minus406191

22

301821

282

--

--

J1706235-405015

S1633

1G345656-00016

12565995

minus408392

86

488601

1350

--

--

J1706081-405742

S1674_N1

1G345529-00049

02565320

minus409599

86

4755079

6282

--

--

J1705119-405058

S1826

1G345510+00160

02562975

minus408493

70

321110

196

--

--

J1704281-404620

S1914

1G345486+00315

12561143

minus407739

116

4066147

6682

--

--

J1704319-404635

S1914

2G345493+00303

12561325

minus407756

34

5713627

1522

--

--

J1711244-392953

S945

1G347305+00016

02578527

minus394964

69

463192

544

--

--

J1709168-395013

S1205

1G346791+00143

02573247

minus398353

174

571213

329

--

--

J1709117-393901

S1211

1G346932+00270

02573003

minus396458

173

811441

263

--

--

J1716324-392718

S324

1G347921-00763

j1

2591392

minus394536

154

6914869

1527

--

minus074

k010

J1714327-391514

S532

1G347854-00335

02586375

minus392584

169

924045

554

--

--

J1714311-391400

S536

1G347872-00314

02586301

minus392321

45

922661

332

--

--

J1712058-385547

S841

1G347838+00250

02580162

minus389286

287

594102

538

--

--

aATC

Asource

hastwocomponentsTh

eirfl

uxes

weresummed

upto

computethespectra

lindex

bSp

ectra

lindex

valueobtained

with

noisyATC

Asubbandmeasurements

cSp

ectra

lindex

valuenotreportedas

notreliable(eghighfit

χ2dueto

noisyATC

Asubbandmeasurements)

dASK

APflu

xpotentially

unreliableSp

ectra

lindex

estim

ated

usingATC

Adataonly

eTh

isASK

APsource

includes

fluxfrom

anearby

morecompactHiiregion

(G344989-00269)thatcannotb

edeblendedBothsourcesa

reobserved

inATC

AN

oATC

Aflu

xesa

rereporte

dforthiss

ource

fTh

isASK

APsource

includes

fluxfrom

anearby

point-source(visiblein

ATC

A)thatcannotb

edeblendedMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

gMatchingsource

inATC

Adatahasa

relativ

elyextended

morphologythatcannot

beproperly

fittedwith

agaussian

model

hTh

issource

matches

also

with

aconfi

rmed

star

(IRAS

16472-4401)

iTh

issource

matches

also

with

aconfi

rmed

star

(2MASSJ17074166-4031240)

jTh

issource

matches

also

with

aconfi

rmed

star

(Cl

NGC

6318

PCA

7229)

kSp

ectra

lindex

obtained

usingMGPS

andNVSS

data

  • 1 Introduction
  • 2 Observational data of the Scorpio field
    • 21 ASKAP 912 MHz observations and data reduction
    • 22 ATCA 21 GHz observations and data reduction
    • 23 MOST 843 MHz observations
    • 24 NVSS 14 GHz observations
    • 25 TGSS 150 MHz observations
    • 26 GLEAM 200 MHz observations
    • 27 Supplementary surveys
      • 3 Compact sources in the Scorpio field
        • 31 Source finding
        • 32 Source cross-matching
        • 33 Catalogue selection effects and uncertainties
          • 4 Analysis
            • 41 Estimation of the fraction of resolved sources
            • 42 Spectral indices
            • 43 Source counts
              • 5 Source classification
                • 51 Search for known Galactic objects
                • 52 Extragalactic sources
                • 53 Unclassified sources
                  • 6 Summary
                  • A Source finder parameters
                  • B Catalogue format
                  • C Scorpio ATCA map
                  • D Source catalogue map
                  • E Source tables
Page 19: EvolutionaryMapoftheUniverse(EMU
Page 20: EvolutionaryMapoftheUniverse(EMU
Page 21: EvolutionaryMapoftheUniverse(EMU
Page 22: EvolutionaryMapoftheUniverse(EMU
Page 23: EvolutionaryMapoftheUniverse(EMU
Page 24: EvolutionaryMapoftheUniverse(EMU
Page 25: EvolutionaryMapoftheUniverse(EMU
Page 26: EvolutionaryMapoftheUniverse(EMU
Page 27: EvolutionaryMapoftheUniverse(EMU
Page 28: EvolutionaryMapoftheUniverse(EMU
Page 29: EvolutionaryMapoftheUniverse(EMU
Page 30: EvolutionaryMapoftheUniverse(EMU
Page 31: EvolutionaryMapoftheUniverse(EMU