Mon. Not. R. Astron. Soc. 000, 000–000 (0000) Printed August 23, 2017 (MN LATEX style file v2.2)
Determining the torus covering factors for a sample oftype 1 AGN in the local Universe
Savithri H. Ezhikode,1? Poshak Gandhi,2 Chris Done,3 Martin Ward,3
Gulab C. Dewangan,4 Ranjeev Misra,4 Ninan Sajeeth Philip11 Department Of Physics, St. Thomas College, Kozhencherry, Kerala 689641, India2 School of Physics & Astronomy, University of Southampton, Highfield, Southampton SO17 1BJ, UK3 Centre for Extragalactic Astronomy, Department of Physics, Durham University, South Road, Durham DH1 3LE, UK4 Inter-University Centre for Astronomy & Astrophysics, Post Bag 4, Ganeshkhind, Pune, India
August 23, 2017
ABSTRACTIn the unified scheme of active galactic nuclei, a dusty torus absorbs and then repro-cesses a fraction of the intrinsic luminosity which is emitted at longer wavelengths.Thus, subject to radiative transfer corrections, the fraction of the sky covered by thetorus as seen from the central source (known as the covering factor fc) can be esti-mated from the ratio of the infrared to the bolometric luminosities of the source asfc=Ltorus/LBol. However, the uncertainty in determining LBol has made the estima-tion of covering factors by this technique difficult, especially for AGN in the localUniverse where the peak of the observed SEDs lies in the UV (ultraviolet). Here, wedetermine the covering factors of an X-ray/optically selected sample of 51 type 1 AGN.The bolometric luminosities of these sources are derived using a self-consistent, energy-conserving model that estimates the contribution in the unobservable far-UV region,using multi-frequency data obtained from SDSS, XMM-Newton, WISE, 2MASS andUKIDSS. We derive a mean value of fc ∼ 0.30 with a dispersion of 0.17. Sample corre-lations, combined with simulations, show that fc is more strongly anti-correlated withλEdd than with LBol. This points to large-scale torus geometry changes associated withthe Eddington-dependent accretion flow, rather than a receding torus, with its innersublimation radius determined solely by heating from the central source. Furthermore,we do not see any significant change in the distribution of fc for sub-samples of radio-loud sources or Narrow Line Seyfert 1 galaxies (NLS1s), though these sub-samples aresmall.
Key words: galaxies:active − galaxies:Seyfert − X-rays:galaxies − ultraviolet: galax-ies − infrared: galaxies
1 INTRODUCTION
Studying broadband spectral energy distributions (SEDs) ofactive galactic nuclei (AGN) can shed light on the emissionmechanisms operating in the distinct physical componentsof the AGN. For example, the big blue bump (BBB) (Rich-stone & Schmidt 1980; Sanders et al. 1989; Elvis et al. 1994)in the optical/UV band is associated with modified blackbody emission from the accretion disc, whereas the Comp-tonised emission from the corona (Haardt & Maraschi 1993)produces an X-ray power law continuum above ∼ 2 keV.Even though the spectral features of various classes of AGNare distinct, the so-called unification scheme (for reviews
? Contact e-mail: [email protected]
see Antonucci 1993; Urry & Padovani 1995; Netzer 2015)postulates that the different types of AGN are intrinsicallysimilar. The model suggests the presence of a dusty, molec-ular torus shaped structure, surrounding the central source,which gives rise to anisotropic emission in polar directions.The observed characteristics of AGN are governed by theorientation of this obscuring torus with respect to our lineof sight.
This torus is optically-thick, with a size of 0.1−10 pc(Suganuma et al. 2006; Kishimoto et al. 2007; Burtscheret al. 2013). The region has a gas density in the rangeof 104−107 cm−3 while the column density ranges from∼ 1022 to ∼ 1025 cm−2 (Netzer 2013). In the most simpleorientation-based unification scheme, the broad-line regionmay or may not be obscured by the torus material depend-
c© 0000 RAS
arX
iv:1
610.
0042
9v2
[as
tro-
ph.H
E]
22
Aug
201
7
2 Ezhikode et. al.
ing on the inclination angle of the system with respect to ourline of sight. In this picture, the classification as a type 2 or atype 1 AGN is determined by the orientation alone. The ob-scuration is parameterised by the opening angle of the toruswhich in turn determines the covering factor fc. Hence thecovering factor is defined as the fraction of the sky that thetorus blocks/absorbs the emission from the central source.
The torus is directly exposed to the emission from thecentral engine and the photons illuminating this region areabsorbed by the dust grains. The heated dust then re-radiates these absorbed optical/UV photons in the infrared(IR) band. Hence the IR continuum is attributed to thethermal emission from the silicate and graphite grains witha broad temperature distribution extending up to the subli-mation temperatures of about 1500 K. Cooler dust on largerscales emits at longer wavelengths and also shows a silicatefeature around ∼ 10 µm.
Dust grains can no longer survive if heated above theirsublimation temperature. As a result, the inner radius of thedust is determined by the distance from the centre, at whichthe dust is sublimated by the primary continuum. A moreluminous AGN heats the dust more strongly and hence thesublimation radius will be larger, with R ∝ L1/2
Bol (Barvainis1987). If the dust distribution has a fixed scale height (asopposed to scaling with mass and/or mass accretion rate),then this means that the covering factor of the dusty torusdecreases as the luminosity of the source increases. This anti-correlation between fc and the luminosity was suggested byLawrence (1991) and is called the receding torus model. Thisdoes not depend on the detailed dust distribution. For exam-ple a clumpy torus (Nenkova et al. 2008; Honig & Kishimoto2010) would exhibit much the same behaviour as the cov-ering factor is determined mostly by the total solid anglecovered by the dust. Similarly, if most of the mid-infrared(MIR) emission arises from scales beyond the classical torus(Honig et al. 2013; Lopez-Gonzaga et al. 2016; Asmus et al.2016), then fc would be a measure of the efficiency of thisextended emission component and how it scales with AGNpower.
There are various methods for determining fc. One is astatistical approach, based on optical demographic studiesof AGN which compared the fraction of type 1 and 2 AGN,with L[OIII] used as a proxy for LBol. Alternatively, this canbe done in X−rays, using the fraction of X−ray unobscuredto obscured AGN, with LX−ray tracing LBol. Both types ofstudies generally find a decrease in the fraction of obscuredAGN with increasing luminosity, consistent with the reced-ing torus model (e.g. Simpson 2005; Hao et al. 2005; Uedaet al. 2003; Steffen et al. 2003; La Franca et al. 2005; Treister& Urry 2006; Hasinger 2008; Toba et al. 2013; Mateos et al.2016) but with a few exceptions (Dwelly & Page 2006; Eckartet al. 2006) or additional correlations (Hasinger 2008). How-ever, neither the X−rays nor the optical emission lines, usedas proxies, give a reliable estimate of the bolometric lumi-nosity, LBol (Vasudevan & Fabian 2007; Lusso et al. 2013;Jin et al. 2012a,b, hereafter J12a, J12b), which is the keydriving parameter in the receding torus model, and there arealso multiple selection effects (Lawrence & Elvis 2010).
Alternatively, the SED can be fitted over as wide abandpass as possible to directly constrain LBol from obser-vations, and then the ratio of the IR luminosity to the bolo-metric luminosity LIR/LBol can be used to derive fc based
on dust (re)emission. Again these studies generally show ananti-correlation with LBol (Gallagher et al. 2007; Maiolinoet al. 2007; Hatziminaoglou et al. 2008, 2009; Gandhi et al.2009; Roseboom et al. 2013; Lusso et al. 2013). Nonethe-less, there are still uncertainties. A major difficulty in theSED-based analysis is determining the bolometric luminos-ity of the AGN. A substantial part of the AGN luminosityemerges in the UV region and is unobservable due to theinterstellar absorption with our Galaxy. Furthermore, theIR luminosity can be self-obscured, with radiation transfereffects through optically-thick dust affecting the IR luminos-ity observed for type 2 (obscured) objects (Stalevski et al.2016; Treister et al. 2008; Pier & Krolik 1992).
In this work, we use a sample of 51 unobscured AGN,so these should not be affected by radiative transfer ef-fects. They all have well sampled broadband optical-UV-X−ray data to define the SED. Additionally, they have LBol
estimated by fits using a self-consistent energy-conservingmodel to bridge across the unobservable far-UV (FUV) re-gion (J12a, J12b). In this paper, we extend these SEDs toinclude far-infrared (FIR) wavelengths to estimate the cov-ering factors of the AGN in our sample. We then investigatethe dependence of fc with different AGN properties.
This paper is organised as follows. In Section 2, wediscuss the sample selection and preparation of the multi-wavelength data. Section 3 gives a detailed description of themodelling of the broadband SED of the sample. Section 4deals with the main results obtained in the work. Section 5is dedicated to the discussion of the results and the con-clusions are given in Section 6. The details regarding thelocal models used in this work, the broadband SED plot foreach source, notes on selected sources, and other relevantinformation is given in the Appendix. Throughout this pa-per, we have adopted a cosmology with Hubble constant ofH0 = 70 km s−1 Mpc−1, ΩΛ=0.73 and ΩM=0.27.
2 SAMPLE SELECTION & DATAPREPARATION
For our study, we choose the sample of 51 type 1 AGN anal-ysed by J12a and J12b. These are selected to have goodSDSS spectra (DR7) with z < 0.4 so that Hα and Hβlines (black hole mass estimator) are included in the band-pass, and good quality XMM-Newton X-ray data are avail-able, without complex absorption features. Optical bandcontinuum points were extracted from the SDSS data byremoving line emission, Balmer continuum, and host galaxycontribution. The XMM-Newton satellite also has the Op-tical Monitor (OM) which provides simultaneous optical-UV photometry. These photometric points were extractedusing 6′′diameter apertures to minimise host galaxy con-tamination. Therefore these sources all have well sampledSEDs, which can be modelled to give a good estimate oftheir bolometric luminosity. The sample spans a broad rangeof AGN types comprising 12 Narrow Line Seyfert 1 galax-ies (NLS1s), 39 Broad Line Seyfert 1 galaxies (BLS1s), abroad absorption line (BAL) quasar which is also radio-loud (PG 1004+130, No. 13) and two more radio-loud AGNRBS 0875 (No. 14) and PG 1512+370 (No. 45). Further in-formation of these sources is given in Table 1 of J12a.
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 3
[]
Table 1. The bandwidth in µm (Col. 3), the effective wavelength
λeff in µm (Col. 4), the zero-point flux density Fν0 in Jy (Col. 5)
to convert from magnitudes and the extinction correction factor(Col. 6) for WISE, 2MASS & UKIDSS bands.
Survey BandBandwidth λeff Fν0 Aλ/AV
µm µm Jy
WISE
W 1 0.663±0.001 3.35±0.01 309.5±4.6 0.069
W 2 1.042±0.001 4.60±0.02 171.8±2.5 0.053
W 3 5.510±0.020 11.56±0.04 31.7±0.5 0.068
W 4 4.100±0.040 22.09±0.12 8.4±0.3 0.052
2MASS
J 0.162±0.001 1.235±0.006 1594.0±27.8 0.282
H 0.251±0.002 1.662±0.009 1024.0±20.0 0.190
K 0.262±0.002 2.159±0.011 666.7±12.6 0.114
UKIDSS
Y 0.102 1.031 2026.0 0.380
J 0.159 1.248 1530.0 0.282
H 0.292 1.631 1019.0 0.190
K 0.351 2.201 631.0 0.114
2.0 2.5 3.0 3.5W2−W3(mag)
0.5
1.0
1.5
W1−
W2(m
ag)
12
2028
36
42
Figure 1. WISE colour-colour plot for our sample. The solid linedescribes the AGN wedge (Mateos et al. 2012) and the colour-cut(Stern et al. 2012) is plotted with the dashed line. Blue triangles
denote objects with significant host galaxy contribution (Lhost >10% of LBol) in the model fit.
2.1 IR Data
We extended the continuum of the SEDs discussed in J12bto include IR data obtained from the Wide-field InfraredSurvey Explorer (WISE ; Wright et al. 2010), Two MicronAll-Sky Survey (2MASS; Skrutskie et al. 2006) and UKIRTInfrared Deep Sky Survey (UKIDSS) catalogues (Hewettet al. 2006). These span a wavelength range from ∼ 1 µmto 20 µm, giving good coverage from near-infrared (NIR)to mid-infrared wavelengths. For cases where both 2MASSand UKIDSS data were available for each source, we optedto use the UKIDSS data for the NIR band due to its smalleraperture size. Our principal results, and specifically the dis-tribution of covering factors, do not depend on the choiceof 2MASS or UKIDSS data, a point which is expanded onin Appendix E. We used 2MASS only for sources for whichthere are no UKIDSS observations. One object in the sam-
ple (2XMM J100523.9+410746; No. 12) does not have eitherUKIDSS or 2MASS data. In that case, we could use onlythe WISE data for the IR analysis. The bandpasses withtheir zero magnitude flux and aperture sizes are given in Ta-ble 1, together with the extinction corrections in each band,Aλ, extracted from Cardelli et al. (1989) with RV =3.1 (seeGandhi et al. (2011) and references therein for WISE extinc-tion corrections). The resulting NIR-MIR fluxes, correctedfor Galactic reddening, are listed in Table 2. We incorporatethese into xspec using the ftool flx2xsp.
We investigated the dominant IR emission mechanismbased on WISE colour selection thresholds of Mateos et al.(2012) and Stern et al. (2012). The MIR colour-cut de-fined by Stern et al. (2012) identifies the AGN candidateswith W 1-W 2 > 0.8. In addition, the AGN wedge of Ma-teos et al. (2012) is designed to select objects with redMIR power-law SEDs in the first three bands of WISE.Fig. 1 shows that most of the sources in our sample, ex-cept 2XMM J100523.9+410746 (No. 12), RX J1140.1+0307(No. 20), RX J1233.9+0747 (No. 28), 1E 1346+26.7 (No. 36)& NGC 5683 (No. 42), are within the AGN wedge and abovethe colour-cut. This confirms that the MIR is likely domi-nated by the AGN rather than the host galaxy in most cases.The MIR fluxes for those which lie below the colour wedgeand the colour-cut are likely to be dominated by the stellarpopulation or star formation activity in the host galaxy, es-pecially in the case of W 1 filter. For example, the MIR SEDof NGC 5683 (No. 42) shows a significant contribution fromthe host galaxy.
3 THE BROADBAND SED MODEL
Multiwavelength observations are a crucial ingredient in un-derstanding the physical processes occurring in AGN andto study the structure of their inner regions. Some notablefeatures in the broadband SED model of AGN are; thehard X-ray power law, the soft X-ray excess below 2 keV,the big blue bump which peaks in optical/UV region, andthe infrared bump at ∼10 µm. The optical/UV emission inAGN is thought to arise from a multi-temperature accre-tion disc. The power-law component originates from the in-verse Compton scattering of accretion disc photons by a hot,optically-thin corona. The infrared emission results from re-processing of the absorbed optical/UV/X-ray emission fromthe AGN.
3.1 Modelling the Optical/UV & X−rays
The SED can be phenomenologically fitted by a black bodycomponent for the accretion disc and thermal Comptoni-sation from an optically-thin, high-temperature corona tomodel the hard X-ray power law above 2 keV. At lowerwavelengths, the soft X-ray excess can be modelled withan optically-thick, low-temperature thermal Comptonisa-tion model. In J12b, they modelled the Optical/UV &X−ray continua with the XSPEC model optxagnf 1 (Done
1 A description of optxagnf can be found in the XSPEC website
http://heasarc.nasa.gov/xanadu/xspec/models/optxagn.html
c© 0000 RAS, MNRAS 000, 000–000
4 Ezhikode et. al.
et al. 2012). This model is fully self-consistent (energy con-serving), and associates all components with emission fromthe accretion disc itself and energy extracted from it i.e. thesoft X-ray component and the hard X-ray power-law. optx-agnf is parameterised by black hole mass (MBH), Eddingtonratio (L/LEdd or λEdd), black hole spin (a), coronal radius(Rcor), outer radius of the accretion disc (Rout), electrontemperature (kTe) and optical depth (τ) of the corona pro-ducing soft X-ray component, hard X-ray (2−10 keV) pho-ton index (Γ), fraction of the coronal energy emitted in thehard X-ray power law (fpl) and redshift (z) i.e., ten parame-ters in total. In detail, this model introduces Rcor, the radiusdown to which the gravitational energy is released as blackbody emission in the disc. Within this radius, the energy isemitted as the soft X-ray excess and the high energy powerlaw. In this model, the mass accretion rate is constrained bythe optical/UV luminosity. If the black hole mass is known,then these parameters can be used to estimate the total lu-minosity. J12b assumed an accretion efficiency of 0.057 fora Schwarzschild black hole, and hence determined the to-tal luminosity. The study presented in J12b is a refinementto the SED fitting given in J12a. This now includes a self-consistent colour temperature correction for the standarddisc emission.
We adopted the same procedures as described in J12b,and performed the spectral analysis of the optical/UV andX-ray continua using optxagnf in XSPEC version 12.8.2.We model the Galactic absorption and reddening using thestandard routines wabs and redden respectively, and any in-trinsic absorption/reddening using zwabs/zredden. The in-trinsic column density was left free to vary during the spec-tral fitting while the Galactic column density for each sourcewas frozen to the value obtained from LAB Survey (Kalberlaet al. 2005). We fixed the black hole mass to the best-fit value obtained by J12b in their models, who fixed up-per(lower) limits to the mass from the broad(intermediate)velocity width of the Hβ line decomposition. We also followthe method described in J12a, J12b and fix the outer ra-dius of the accretion disc to be 104Rg. This is probably anupper limit to the size of the disc, as both the self-gravityradius and the best-fit to the disc emission generally indicatea somewhat smaller disc (Hao et al. 2010; Collinson et al.2016). The redshift2 z is fixed to the value given in Table 2,this leaves 9 free parameters for the X-ray fitting portion ofthe model.
3.2 Modelling the Infrared
Our extended wavelength IR coverage is modelled as acombination of dust re-radiation and host galaxy emission.For the dust, we use the Seyfert 1 (unobscured) template ofSilva et al. (2004), and for the host galaxy we use a range of13 templates spanning ellipticals, spirals, and star-forminggalaxies, from the SWIRE library (Polletta et al. 2007). Weincorporate these into xspec as local models, which we callagndust and hostpol. Further details of agndust and hostpoltemplates are given in APPENDIX A. We use only theSeyfert 1 dust template in our fits since all our objects are
2 Redshift for each source is taken from https://ned.ipac.
caltech.edu/
unobscured, but we do investigate the entire range of hostgalaxy templates, and then adopt the one which gives thebest-fit to each object.
The optical/UV, X-ray and IR continua of each source inthe sample were fitted by the final model constant(hostpol+agndust+wabs× redden× zwabs× zredden× optxagnf),where the constant only allows for small cross-calibrationdifferences in normalisation between the X-ray spectraobtained from the three independent cameras EPIC-pn,MOS1 and MOS2, aboard the XMM-Newton satellite.The broadband SED model for each source has 11 freeparameters (including the normalizations of agndust andhostpol), apart from the multiplicative factor of the modelconstant. The fit-statistic and the best-fit hostpol templatefor each source are given in Table 3. The plots showingthe data and individual model components for all thesources are given in the APPENDIX B. Mrk 0110 (No. 9)shows a clear discrepancy in the SDSS data due to extremevariability, and therefore has a very large χ2. Also, there aretwo super-Eddington sources in our sample, KUG 1031+398(No. 15) and PG 2233+134 (No. 50). These sources arediscussed individually in APPENDIX C. We note that thepotential discrepancies mentioned above do not influenceany of our resulting inferences on the fc distribution.
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 5
Table
2.
Der
edd
ened
IRfl
ux
for
the
sam
ple
inea
chb
an
dof
WIS
E,
2M
AS
S,
an
dU
KID
SS
.
No.
Ob
ject
Redsh
ift
IRfl
ux
(mJy)
WIS
E2M
ASS
UK
IDSS
W1
W2
W3
W4
JH
Ks
YJ
HK
1U
M269
0.3
08
2.1
0±
0.0
62.9
0±
0.0
74.2
2±
0.2
67.9
4±
1.3
70.4
5±
0.0
50.7
0±
0.0
80.9
3±
0.0
7-
--
-
2M
rk1018
0.0
43
26.5
8±
0.6
730.5
1±
0.7
252.3
8±
1.0
785.3
6±
3.8
68.3
4±
0.5
811.7
3±
0.9
915.3
5±
0.7
5-
--
-
3N
VSS
J030639
0.1
07
5.1
5±
0.1
36.2
2±
0.1
516.5
2±
0.4
241.5
0±
2.2
8-
--
1.0
2±
0.0
31.3
4±
0.0
41.9
9±
0.0
62.9
7±
0.0
9
42X
MM
J074601.2
+280732
0.1
45
1.3
5±
0.0
41.5
3±
0.0
42.4
3±
0.1
83.8
4±
1.1
90.5
2±
0.0
60.6
6±
0.0
90.7
9±
0.0
8-
--
-
52X
MM
J080608.0
+244421
0.3
58
1.1
2±
0.0
31.5
6±
0.0
43.5
8±
0.2
28.0
0±
1.2
8-
--
0.2
5±
0.0
10.3
0±
0.0
10.4
0±
0.0
10.5
6±
0.0
2
6H
S0810+
5157
0.3
77
1.8
9±
0.0
53.1
8±
0.0
87.9
9±
0.2
521.0
7±
1.5
10.6
3±
0.0
60.8
7±
0.0
91.2
9±
0.0
9-
--
-
7R
BS
0769
0.1
60
3.3
3±
0.0
95.3
6±
0.1
315.4
0±
0.3
932.6
9±
1.6
60.9
0±
0.0
61.2
7±
0.1
11.8
2±
0.0
9-
--
-
8R
BS
0770
0.0
33
37.6
5±
0.9
745.2
6±
1.0
7105.7
6±
2.1
7211.8
7±
8.6
710.0
3±
0.4
613.4
0±
0.7
320.0
1±
0.7
5-
--
-
9M
rk0110
0.0
35
21.6
7±
0.5
629.5
7±
0.7
065.5
8±
1.3
9109.3
6±
4.8
84.5
7±
0.2
35.6
5±
0.3
28.5
8±
0.3
3-
--
-
10
PG
0947+
396
0.2
06
7.9
3±
0.2
012.7
1±
0.3
26.9
1±
0.5
755.2
7±
2.7
71.9
9±
0.0
82.7
5±
0.1
15.3
4±
0.1
7-
--
-
11
2X
MM
J100025.2
+015852
0.3
73
0.3
0±
0.0
10.5
4±
0.0
21.4
7±
0.1
44.4
6±
1.0
5-
--
0.1
1±
0.0
10.1
1±
0.0
10.1
6±
0.0
10.2
1±
0.0
1
12
2X
MM
J100523.9
+410746
0.2
06
0.3
5±
0.0
10.5
0±
0.0
20.5
4±
0.1
53.5
5±
0.1
2-
--
--
--
13
PG
1004+
130
0.2
41
8.1
0±
0.2
111.9
0±
0.2
836.2
0±
0.7
983.7
3±
4.0
43.4
1±
0.1
23.5
4±
0.1
65.0
0±
0.1
8-
--
-
14
RB
S0875
0.1
78
6.7
2±
0.1
78.8
6±
0.2
114.8
4±
0.4
129.3
3±
1.9
1.7
0±
0.1
02.2
1±
0.1
53.5
8±
0.1
5-
--
-
15
KU
G1031+
398
0.0
43
11.6
6±
0.2
919.3
2±
0.4
570.4
6±
1.4
0132.3
3±
5.6
93.5
7±
0.1
44.3
8±
0.2
15.8
6±
0.2
2-
--
-
16
PG
1048+
342
0.1
67
3.8
9±
0.1
05.5
8±
0.1
413.6
1±
0.3
427.7
8±
1.5
31.2
8±
0.0
61.5
7±
0.0
82.4
6±
0.1
1-
--
-
17
1R
XS
J111007
0.2
62
2.0
7±
0.0
53.0
8±
0.0
78.6
2±
0.2
722.1
4±
1.3
30.3
3±
0.0
50.7
1±
0.0
71.1
6±
0.0
9-
--
-
18
PG
1115+
407
0.1
55
9.0
0±
0.2
311.5
4±
0.2
724.1
8±
0.5
349.5
3±
2.3
62.2
1±
0.1
23.1
4±
0.1
85.3
6±
0.2
0-
--
-
19
2X
MM
J112328.0
+052823
0.1
01
2.6
4±
0.0
73.6
5±
0.0
99.4
3±
0.2
816.4
2±
1.3
0-
--
0.4
9±
0.0
20.6
4±
0.0
20.9
6±
0.0
31.5
5±
0.0
5
20
RX
J1140.1
+0307
0.0
81
1.3
5±
0.0
41.3
2±
0.0
44.2
8±
0.1
912.0
6±
1.3
5-
--
0.3
5±
0.0
10.4
5±
0.0
10.6
5±
0.0
2-
21
PG
1202+
281
0.1
65
7.4
1±
0.1
910.5
7±
0.2
529.0
2±
0.6
587.0
3±
3.7
42.0
3±
0.0
82.5
3±
0.1
04.8
2±
0.1
5-
--
-
22
1A
XG
J121359+
1404
0.1
54
1.0
6±
0.0
31.3
4±
0.0
42.8
9±
0.1
77.5
7±
1.0
8-
--
0.3
6±
0.0
10.4
7±
0.0
10.6
1±
0.0
20.7
9±
0.0
2
23
2E
1216+
0700
0.0
80
6.4
5±
0.1
77.7
9±
0.1
817.4
2±
0.4
624.8
5±
1.6
5-
--
1.2
8±
0.0
41.7
5±
0.0
52.7
0±
0.0
83.9
0±
0.1
2
24
1R
XS
J122019
0.2
86
2.0
8±
0.0
63.0
6±
0.0
85.6
6±
0.2
07.8
3±
1.0
90.5
2±
0.0
60.7
8±
0.1
01.3
5±
0.1
1-
--
-
25
LB
QS
1228+
1116
0.2
36
3.0
8±
0.0
84.6
8±
0.1
211.1
7±
0.2
830.9
9±
1.8
6-
--
0.9
6±
0.0
30.9
9±
0.0
31.2
9±
0.0
41.5
6±
0.0
5
26
2X
MM
J123126.4
+105111
0.3
04
0.4
9±
0.0
20.6
6±
0.0
21.7
1±
0.1
72.3
9±
0.9
1-
--
0.1
2±
0.0
00.1
4±
0.0
10.1
9±
0.0
10.3
3±
0.0
1
27
Mrk
0771
0.0
64
13.3
4±
0.3
518.4
2±
0.4
357.3
8±
1.1
7159.3
9±
6.6
84.6
3±
0.2
66.1
5±
0.3
79.4
6±
0.3
9-
--
-
28
RX
J1233.9
+0747
0.3
71
0.5
1±
0.0
20.4
8±
0.0
21.0
9±
0.1
85.3
3±
0.1
9-
--
0.1
1±
0.0
00.1
5±
0.0
10.1
9±
0.0
10.2
4±
0.0
1
29
RX
J1236.0
+2641
0.2
09
3.0
8±
0.0
84.6
1±
0.1
113.1
2±
0.3
737.9
5±
1.9
80.7
4±
0.0
51.0
4±
0.0
81.6
6±
0.0
8-
--
-
30
PG
1244+
026
0.0
48
6.8
6±
0.1
710.0
3±
0.2
441.5
3±
0.8
8114.8
3±
4.8
7-
--
1.3
4±
0.0
41.5
8±
0.0
52.1
5±
0.0
63.2
3±
0.1
0
31
2X
MM
J125553.0
+272405
0.3
16
1.1
3±
0.0
31.3
6±
0.0
42.3
9±
0.1
76.3
7±
0.9
3-
--
0.2
8±
0.0
10.3
5±
0.0
10.5
1±
0.0
20.7
8±
0.0
2
32
RB
S1201
0.0
91
3.2
0±
0.0
83.5
9±
0.0
98.2
9±
0.2
519.6
5±
1.4
3-
--
1.0
3±
0.0
31.2
7±
0.0
41.7
4±
0.0
51.9
2±
0.0
6
33
2X
MM
J132101.4
+340658
0.3
34
0.7
7±
0.0
20.9
4±
0.0
31.7
3±
0.1
13.9
3±
0.7
80.3
1±
0.0
50.2
9±
0.0
60.6
1±
0.0
8-
--
-
34
1R
XS
J132447
0.3
06
1.6
5±
0.0
52.2
3±
0.0
64.2
1±
0.1
811.4
0±
1.0
6-
--
0.8
2±
0.0
30.8
4±
0.0
30.9
6±
0.0
31.1
9±
0.0
4
35
UM
602
0.2
37
3.7
4±
0.1
05.1
9±
0.1
38.3
4±
0.2
318.2
5±
1.3
1-
--
0.5
2±
0.0
20.5
7±
0.0
20.8
0±
0.0
21.4
7±
0.0
4
36
1E
1346+
26.7
0.0
59
4.3
4±
0.1
34.8
7±
0.1
318.7
4±
0.4
234.3
0±
1.8
6-
--
1.0
9±
0.0
31.4
0±
0.0
42.1
2±
0.0
62.6
6±
0.0
8
37
PG
1352+
183
0.1
51
7.0
9±
0.1
810.0
7±
0.2
418.0
3±
0.4
135.7
8±
1.8
01.6
1±
0.0
82.1
1±
0.1
03.7
8±
0.1
4-
--
-
38
Mrk
0464
0.0
50
3.6
9±
0.1
05.0
3±
0.1
214.3
2±
0.3
428.2
4±
1.5
91.9
4±
0.0
82.3
3±
0.1
03.0
4±
0.1
2-
--
-
39
1R
XS
J135724
0.1
06
0.7
8±
0.0
21.0
0±
0.0
33.5
1±
0.1
215.3
1±
1.1
60.2
8±
0.0
50.5
3±
0.0
80.5
9±
0.0
9-
--
-
40
PG
1415+
451
0.1
14
10.3
9±
0.2
713.3
1±
0.3
129.3
2±
0.6
264.9
3±
2.9
33.8
0±
0.1
25.5
9±
0.1
98.6
9±
0.2
7-
--
-
41
PG
1427+
480
0.2
21
3.6
4±
0.0
95.5
7±
0.1
314.1
8±
0.3
346.3
9±
2.2
41.3
3±
0.0
71.6
8±
0.1
02.7
6±
0.1
2-
--
-
42
NG
C5683
0.0
37
5.3
7±
0.1
44.9
6±
0.1
219.5
5±
0.4
550.0
3±
2.3
25.4
6±
0.2
76.4
7±
0.3
77.1
0±
0.2
9-
--
-
43
RB
S1423
0.2
08
3.8
0±
0.1
05.7
4±
0.1
412.2
6±
0.3
125.9
0±
1.5
6-
--
0.5
0±
0.0
20.7
0±
0.0
20.9
9±
0.0
31.9
4±
0.0
6
44
PG
1448+
273
0.0
65
11.3
5±
0.2
915.1
0±
0.3
645.5
3±
0.9
3117.2
9±
4.9
25.6
4±
0.1
97.2
9±
0.2
410.2
9±
0.3
1-
--
-
45
PG
1512+
370
0.3
71
4.1
8±
0.1
06.6
7±
0.1
613.8
4±
0.3
334.3
6±
1.6
11.0
9±
0.0
61.3
6±
0.0
82.2
6±
0.1
2-
--
-
46
Q1529+
050
0.2
18
2.1
8±
0.0
62.9
2±
0.0
76.0
3±
0.2
019.4
8±
1.4
8-
--
0.4
0±
0.0
10.5
2±
0.0
20.6
8±
0.0
21.2
4±
0.0
4
47
1E
1556+
27.4
0.0
90
1.9
0±
0.0
52.0
8±
0.0
54.0
1±
0.1
66.9
3±
0.8
3-
--
0.8
5±
0.0
31.2
2±
0.0
41.3
4±
0.0
41.6
1±
0.0
5
48
Mrk
0493
0.0
31
19.9
3±
0.5
124.4
6±
0.5
872.0
3±
1.4
3162.8
1±
6.6
76.0
0±
0.2
98.2
1±
0.4
510.6
5±
0.4
4-
--
-
49
IIZ
w177
0.0
81
3.7
9±
0.1
04.7
2±
0.1
120.7
0±
0.4
860.9
9±
2.7
61.7
7±
0.0
92.5
6±
0.1
32.9
1±
0.1
5-
--
-
50
PG
2233+
134
0.3
26
5.2
5±
0.1
38.2
6±
0.2
020.9
7±
0.5
057.3
9±
2.9
20.9
9±
0.0
61.1
8±
0.0
72.0
3±
0.1
2-
--
-
51
Mrk
0926
0.0
47
69.6
0±
1.8
086.1
6±
2.0
3123.6
9±
2.5
3256.3
7±
10.7
58.9
7±
0.3
612.4
3±
0.4
917.0
0±
0.6
4-
--
-
c© 0000 RAS, MNRAS 000, 000–000
6 Ezhikode et. al.
4 RESULTS
4.1 Broadband SEDs & Covering factors
Our best-fit model parameters for the SED fits are givenin Table 3. We integrate the intrinsic optxagnf model (af-ter correcting for all absorption components) over the en-ergy range of 10−6−100 keV (∼10−5−1000 µm) to obtainLBol, and over 2−10 keV to get the hard X-ray luminosityLX−ray. Now, in order to estimate the covering factor wecan compare these to the IR luminosity of the torus Ltorus,which we find by integrating the agndust model componentover ∼1−1000 µm. We also give the luminosity of the hostgalaxy Lhost, obtained by integrating the host galaxy tem-plate over ∼0.1−1000 µm, and give its type in Table 3. Thedistributions of these luminosities are shown in Fig. 2. Thebolometric luminosities of most of the sources in our studyare lower than the values of J12b by a factor of ∼1.7. Thisis due to a change in energy grid handling of optxagnf inan older version of XSPEC3. Though individual LBol valuesare affected, the overall sample trends remain the same. Thehost galaxy is significantly detected in 38 sources. We notethat the host galaxy morphological types of 21 sources inour sample are known from the literature and many of these(No. 8, 13, 15, 16, 20, 21, 23, 27, 30, 37, 38, 39, 42, 44, 48,50) are different to those indicated by our best-fit. We haverefit these objects using the galaxy template fixed to the lit-erature values. Although there is a marginal increase in χ2
when using the substituted templates, this has no significantimpact on the parameters derived for either the AGN or thetorus.
The primary goal of this work is to determine the distri-bution of covering factors fc of the unobscured AGN sample.This distribution of fc obtained from the ratio of the torusluminosity to the bolometric luminosity is given in Fig. 3.The distribution, with an average value around 0.30, has ascatter of ∼0.17. The source with the minimum covering fac-tor is 1E 1556+27.4 (No. 47). The value of fc ∼0.02 suggeststhat this source could be a hot-dust-poor AGN (Hao et al.2010) characterised by weak IR emission. At the oppositeextreme, 2E 1216+0700 (No. 23) has the maximum value offc (∼ 0.88).
4.2 Comparison with Previous Work
Many previous studies have discussed the obscured AGNfraction. Lawrence & Elvis (2010) reviewed the studies thatdealt with the fraction of obscured AGN and concluded thatthe obscured fraction for the non-X-ray selected samples hasa mean value of about 0.58 with a dispersion of ∼ 0.05. Theyalso gave a rough estimate of 0.53 for the obscured fractionusing the updated Swift/BAT hard X-ray catalogue (Tuelleret al. 2010), after applying the correction for missing Comp-ton thick objects (Risaliti et al. 1999). The covering factorsof a sample of 5281 WISE, UKIDSS and SDSS selected highluminosity quasars (LBol >1046 erg s−1) with redshift < 1.5was determined by Roseboom et al. (2013). They found thatthe covering factors (estimated by using the ratio of IR to
3 See the XSPEC patch 12.8.2j available at https:
//heasarc.gsfc.nasa.gov/docs/xanadu/xspec/issues/
archive/issues.12.8.2q.html
UV/optical luminosity) obey a log-normal distribution witha mean observed value of ∼ 0.39 and a dispersion of ∼ 0.2.The study by Lusso et al. (2013) estimated the coveringfactor for a sample of X-ray selected type 1 AGN with aneven wider span of redshifts (0.10 6 z 6 3.75). They deter-mined the covering factor of the optically-thin torus and thiswas observed to be in the range from ∼ 0.45 to ∼ 0.75. Incomparison, the result from our sample is relatively low, es-pecially since our sample is comprised only of Seyfert type 1objects, so radiative transfer corrections, which reduce fcdue to the torus being optically-thick to its own radiation athigh inclination, should not be important (but see Stalevskiet al. 2016). At high luminosities, two other studies, Mor &Trakhtenbrot (2011) and Landt et al. (2011), concentratedon the hot (NIR-emitting) dust component and found evenlower hot dust covering factors of ∼0.13 and ∼0.07, respec-tively.
In APPENDIX D, we investigate the effect of varyingthe template SEDs on the fc measurements. In particular,we examined the impact of using two other IR SEDs fortype 1 AGN −(1) proposed by Mullaney et al. (2011)based upon observations of local sources in which the AGNdominates the IR portion of the SED over the host galaxycontribution; and (2) a theoretical clumpy torus SED byHonig & Kishimoto (2010). While the resultant mean fcvalue and the distribution of values from these template fitsdo not dramatically differ from those obtained previouslywhen we used the SEDs from Silva et al. (2004), we foundthat both these sets of templates required an additional hotdust component in the NIR regime, suggesting that theyare too restricted to account for the broadband featuresin our observations. This finding is interesting, but furtherinvestigation is beyond the scope of our present paper.However, we note that similar hot dust components arerequired in other quasar SED studies (e.g. Mor et al. (2009)).
4.3 fc in Sub-samples
From our main sample, we have constructed sub-samples ofthree radio-loud objects and 12 NLS1 galaxies. The distri-butions of covering factor for these sub-samples are shownin Fig. 3.
4.3.1 Radio-loud sources
A significant contribution from a jet may reduce the cover-ing factor in radio-loud sources. One of these objects in oursample, PG 1004+130 (No. 13), is a BAL quasar with a veryweak X-ray spectrum. J12b suggests that the origin of X-rays in this source could be a sub-parsec-scale jet. However,the covering factors of the three radio-loud objects (No. 13,No. 14 & No. 45) in our sample are 0.33, 0.65 and 0.24,respectively, indicating that the covering factors may notbe affected by the radio-loudness/quietness of the sources,although we cannot draw general conclusions from such asmall sample of radio-loud sources.
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 7
4.3.2 NLS1 galaxies
It is known that the NLS1s, in general, tend to have highervalues of Eddington ratio. Since there is an anti-correlationbetween fc and λEdd we expect the 12 NLS1s in our sampleto have low values for fc. The covering factors of NLS1s inour sample range from ∼0.06 to ∼0.38 with a mean valuearound 0.23 and a dispersion of about 0.1. The Kolmogorov-Smirnov test (K-S test) with a probability of ∼75.5% revealsthat the distribution of fc in NLS1s does not differ signifi-cantly from that of the overall sample. But we caution thatno strong statement can be made, based on such a smallsub-sample of NLS1s. We need further investigations to getbetter constraints on this result.
4.4 Correlations obtained
J12b carried out a systematic study of the correlation be-tween the different AGN parameters in this sample. We fol-low J12b, but also include the new torus parameters, Ltorus
and covering fraction fc and compute the correlations us-ing the Spearman’s rank-order method (Press et al. 1992).The rank coefficient ρs and probability ds (also known asthe p-value) for the Spearman’s correlation between differ-ent parameters are listed in Table 4. We recover most ofthe correlations, obtained by J12b, between LBol, λEdd, Γand the hard X-ray bolometric correction κ2−10 defined asLBol/LX−ray (Vasudevan & Fabian 2007). Additionally, wefind a marginal correlation of fc with LBol (ds=0.05) and astronger correlation between fc and λEdd (ds=0.002). Butfc shows no significant correlation with LX−ray or Ltorus, asshown in Fig. 4.
J12b suggest that the combination of λEdd and MBH
drives the correlations seen in the AGN parameters, withλEdd changing the accretion flow geometry (resulting in thecorrelations in Γ, Rcor and fpl which drive the correlationwith κ2−10) while MBH and λEdd together set the overallluminosity and the peak temperature of the disc. As fc cor-relates with λEdd and LBol ∝ MBHλEdd then it is clear thatthe correlations of fc with λEdd and LBol, shown in Fig. 4,will also give rise to correlations of fc with the other AGNparameters. However, the major statistical correlation withthe new torus parameters is that fc correlates with λEdd andLBol. This is in accord with the physical expectations thatλEdd and LBol are the two key parameters which determinethe properties of the accretion flow.
There is also a weak correlation between Lhost and MBH.This is expected from the black hole mass and bulge lumi-nosity (LBulge) relation of Magorrian et al. (1998). Since theSED of a normal galaxy stellar population peaks in the H-band, we calculated the corresponding luminosity for oursample in terms of Lhost and the relative H-band flux fromPolletta et al. (2007) SED templates. Here, we are not con-sidering Mrk 0110 (No. 9) since Lhost is not well constrainedfor this source. We find that the sources in which our SEDfits include a significant host galaxy contribution approxi-mately follow the Magorrian relationship, though there issignificant scatter (∼0.5 dex) between MBH and LBulge (seeFig. 5). This shows that our host galaxy fitted componentsare largely reasonable. Here, we note that among the 13sources for which there is no significant host galaxy contribu-tion, most of them are fitted by starburst galaxy templates
43 44 45 46log10(LBol; erg s−1)
0
5
10
15
Cou
nt
43 44 45 46log10(LX− ray; erg s−1)
43 44 45 46log10(Ltorus; erg s−1)
0
5
10
15
Cou
nt
43 44 45 46log10(Lhost; erg s−1)
Figure 2. The distribution of LBol (upper left), LX−ray
(2−10 keV) (upper right), Ltorus (lower left) and Lhost (lower
right)
.
(e.g. No. 50, No. 16). These starburst templates peak in theFIR and show a minimum around NIR wavelengths. Sinceour SEDs do not cover the FIR wavelengths and the star-bursts tend to be extended we may be underestimating theirFIR emission, and also their bulge luminosities. In these fewcases, we cannot rule out that we are then over-estimatingthe torus contribution and hence also their fc values. Whenexcluding the 13 sources for which Lhost are not constrained,neither the scatter (∼ 0.17) nor the mean value (∼ 0.33) forthe remaining 38 sources are changed significantly from thecovering factors of the overall sample. This indicates thatthe uncertainties in Lhost do not strongly bias our overallresults.
c© 0000 RAS, MNRAS 000, 000–000
8 Ezhikode et. al.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9fc =Ltorus/LBol
0
5
10
15
20
Cou
nt
Overall sampleNLS1sRadio-Loud
Figure 3. Histogram of covering factors for the sample of 51
type 1 AGN (thick black line) and the sub-samples of 12 NLS1s(dashed blue line) and 3 radio-loud sources (thin red line).
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 9
Table
3.
Bro
adband
SE
Dfi
ttin
gre
sult
s.hostpol
model:
best
-fithostpol
tem
pla
te(∗
starb
urs
tgala
xie
s);χ2 red:
reducedχ2
for
the
best
-fit
model;
NG
al
H&N
Int
H:
Gala
cti
cand
intr
insi
ccolu
mn
densi
ties,
resp
ecti
vely
,
in1020cm
−2;
MBH
:bla
ck
hole
mass
in107M
(fi
xed
toth
eb
est
-fit
valu
es
of
J12b);λEdd
:E
ddin
gto
nra
tio;
Rcor:
coro
nal
radiu
sin
Rg;
kTe:
ele
ctr
on
tem
pera
ture
for
the
soft
Com
pto
nis
ati
on
com
ponent
inkeV
;
τ:
opti
cal
depth
of
the
soft
Com
pto
nis
ati
on
com
ponent;
Γ:
hard
X-r
ay
photo
nin
dex;
f pl:
fracti
on
of
the
pow
er
belo
wR
cor
em
itte
din
the
hard
Com
pto
nis
ati
on
com
ponent;L
X−
ray:
unabso
rbed
X-r
ay
lum
inosi
ty
inth
eband
of
2−
10
keV
;L
host:
host
gala
xy
lum
inosi
tyin
the∼
0.1−
1000µ
mband;L
torus:
infr
are
d(∼
1−
1000µ
m)
lum
inosi
tyof
the
toru
sem
issi
on;L
Bol:
bolo
metr
iclu
min
osi
tyin
the
range
of
10−
6−
100
keV
;
κ2−
10:
hard
X−
ray
bolo
metr
iccorr
ecti
on
(LBol/L
X−
ray);fc:
Coveri
ng
facto
r,th
era
tio
ofL
torus
andL
Bol.
All
lum
inosi
ties
are
expre
ssed
in1044
erg
s−1.
No.
hostpol
χ2 red
NG
al
HN
Int
HM
BH
λEdd
Rcor
kTe
τΓ
f pl
LX−
ray
Lhost
Ltorus
LBol
κ2−
10
fc
model
1020cm
−2
1020cm
−2
107
M
Rg
keV
1044
erg
s−1
1044
erg
s−1
1044
erg
s−1
1044
erg
s−1
1E
51.3
41.7
93.4
341.0
00.1
197.0
0.1
128.1
1.8
50.4
93.0
<0.5
14.9
41.6
13.7
40.3
6
2S0
1.9
92.4
31.2
96.9
20.0
894.2
0.1
916.9
1.8
20.3
60.4
0.6
1.9
8.3
18.8
00.2
3
3S0
1.3
46.3
10.6
64.0
70.2
012.6
0.1
815.2
1.8
70.8
20.3
0.9
3.9
10.1
37.9
40.3
8
4IR
AS
22491-1
808∗
1.2
33.4
92.6
960.0
00.0
184.3
0.3
414.3
1.6
70.4
50.5
<0.4
0.9
8.0
17.2
80.1
2
5S0
1.2
93.5
38.0
88.7
10.9
734.8
0.2
014.3
2.1
60.1
51.0
1.6
13.1
72.0
71.6
00.1
8
6IR
AS
22491-1
808∗
1.3
94.2
40.0
031.4
00.2
921.0
0.4
211.0
1.8
50.4
52.4
17.1
28.8
73.8
30.9
00.4
0
7IR
AS
22491-1
808∗
1.2
01.3
30.0
03.8
00.9
910.5
0.1
532.7
2.2
00.3
10.2
3.7
7.7
42.7
178.1
00.1
8
8S0
1.8
93.1
23.8
03.4
70.2
921.7
0.1
814.4
1.8
00.5
10.5
0.2
2.2
14.1
26.4
20.1
6
9Sa
19.6
21.3
00.0
02.5
10.6
517.4
0.2
615.5
1.7
30.8
31.0
−1.4
22.5
22.7
00.0
6
10
S0
1.7
01.7
41.0
121.9
00.3
839.5
0.4
99.6
1.7
60.2
82.7
<0.2
27.9
87.0
32.3
30.3
2
11
IRA
S20551-4
250∗
1.0
01.7
20.0
08.3
20.3
823.1
0.4
512.3
1.5
40.8
41.0
8.1
4.1
22.8
23.0
50.1
8
12
S0
1.1
71.2
00.0
06.6
10.0
519.3
0.1
054.1
1.9
40.8
80.2
0.7
1.1
3.5
17.8
10.3
0
13
IRA
S22491-1
808∗
1.9
23.5
60.0
0158.4
90.0
79.9
0.1
915.0
1.4
50.9
30.5
62.1
34.5
104.5
193.8
60.3
3
14
S0
1.4
71.7
60.0
017.3
80.1
1100.0
0.2
320.3
1.7
70.8
12.4
2.5
12.6
19.3
7.9
50.6
5
15
Sd
1.6
71.3
12.8
10.2
02.6
4100.0
0.3
08.7
2.2
00.0
30.0
40.8
1.8
7.8
190.6
40.2
3
16
IRA
S22491-1
808∗
1.5
01.7
02.2
119.9
50.1
828.9
0.3
013.0
1.7
40.3
41.2
<1.3
8.1
39.6
34.3
30.2
0
17
Arp
220∗
1.0
60.6
51.1
36.1
70.5
193.8
0.4
89.0
1.6
10.1
20.4
18.0
13.6
30.9
72.2
60.4
4
18
E5
1.8
21.4
50.0
013.1
80.6
111.6
0.2
514.0
2.2
00.4
71.0
<0.1
13.4
91.9
94.8
30.1
5
19
S0
1.0
03.7
00.0
05.1
00.0
469.1
0.1
620.6
1.9
40.2
10.1
0.3
2.0
2.5
36.3
90.7
8
20
Sd
1.2
41.9
10.0
00.6
30.1
918.0
0.2
022.5
2.2
00.5
70.1
0.5
0.5
1.6
30.1
80.3
0
21
M82∗
3.8
21.7
70.0
09.6
00.4
623.7
0.2
118.6
1.7
70.7
83.0
<0.1
17.6
48.6
16.2
70.3
6
22
S0
1.2
12.7
54.3
76.9
00.0
877.7
0.1
223.1
1.8
20.3
30.3
0.6
1.7
6.6
21.6
80.2
5
23
S0
2.1
01.5
90.0
010.0
00.0
2100.0
0.3
614.1
1.5
20.4
90.1
0.7
2.3
2.6
20.2
30.8
8
24
S0
2.9
01.6
30.0
018.0
00.1
539.8
0.2
225.1
1.9
10.9
52.6
2.1
13.7
23.9
9.3
20.5
7
25
IRA
S22491-1
808∗
2.0
02.3
40.0
126.9
20.3
316.2
0.3
413.0
1.7
90.5
22.3
<4.6
12.9
87.2
38.7
90.1
5
26
S0
1.2
02.3
16.5
05.0
10.7
616.5
0.3
59.2
2.0
10.1
0.2
0.3
4.1
34.5
171.7
50.1
2
27
IRA
S22491-1
808∗
1.6
82.7
51.5
97.2
40.1
628.3
0.1
120.1
2.2
00.2
90.3
4.5
3.9
15.8
47.1
50.2
4
28
S0
1.6
41.4
55.5
59.1
20.6
775.3
0.3
710.6
1.6
90.2
21.4
0.6
4.9
50.0
36.1
70.1
0
29
Arp
220∗
1.2
31.1
81.3
17.4
10.4
612.9
0.1
515.0
2.1
80.4
50.5
21.2
11.9
35.6
73.9
30.3
4
30
IRA
S22491-1
808∗
1.2
41.8
70.0
01.8
60.2
019.0
0.2
319.3
2.2
00.4
60.1
4.2
1.1
5.3
36.9
60.2
1
31
Sb
1.5
50.8
40.0
050.0
00.0
6100.0
0.4
014.2
1.4
50.7
41.4
<0.7
8.1
19.9
14.5
90.4
1
32
S0
1.6
40.9
00.0
04.1
70.1
1100.0
0.3
413.3
1.8
50.4
90.4
0.6
1.5
5.7
13.2
30.2
6
33
S0
2.0
51.0
70.0
08.3
20.3
313.6
0.2
020.2
2.1
90.7
90.6
0.8
6.9
23.3
39.4
50.2
9
34
S0
1.4
81.8
31.9
651.0
00.0
654.7
0.2
316.4
1.8
70.4
01.4
5.3
10.9
28.6
20.3
60.3
8
35
S0
2.6
71.7
60.0
04.6
40.9
077.5
0.3
118.4
1.9
00.9
76.2
1.4
13.6
41.0
6.6
70.3
3
36
Sb
1.4
31.1
82.4
81.0
00.2
025.6
0.4
47.9
2.1
90.3
41.0
0.4
1.0
2.8
36.2
50.3
5
37
Sa
1.8
41.8
20.9
817.0
00.1
893.1
0.1
620.1
2.1
20.3
91.4
0.3
9.8
37.2
26.1
00.2
6
38
S0
1.5
61.4
20.6
56.1
70.0
1100.0
0.2
314.1
1.7
20.9
00.2
0.2
0.6
1.1
7.3
10.5
5
39
Arp
220∗
1.0
11.3
61.4
52.1
90.0
891.9
0.2
217.5
2.0
80.3
50.1
4.2
0.6
2.2
25.0
60.2
7
40
S0
2.1
50.7
70.0
07.5
90.2
015.5
0.3
511.7
1.9
30.4
50.4
2.5
8.4
19.2
45.3
50.4
4
41
IRA
S22491-1
808∗
1.4
81.8
10.8
213.8
00.5
914.0
0.3
911.0
1.8
60.5
11.8
17.3
15.1
83.8
47.9
10.1
8
42
Sc
2.5
62.8
60.0
05.5
00.0
247.9
0.2
415.0
1.8
90.7
90.1
0.6
0.3
1.3
10.4
20.2
2
43
IRA
S22491-1
808∗
1.5
42.6
90.2
312.0
00.3
6100.0
0.2
913.3
1.8
70.4
63.2
<0.4
12.9
46.2
14.2
80.2
8
44
E2
1.4
42.7
84.1
13.6
30.7
112.7
0.3
310.0
2.1
40.2
0.2
1.7
3.2
35.0
150.6
70.0
9
45
IRA
S20551-4
250∗
2.5
41.4
62.2
660.2
60.4
249.7
0.2
016.9
1.9
30.3
79.3
<1.5
51.9
212.0
22.8
50.2
4
46
IRA
S22491-1
808∗
1.5
94.0
211.3
636.0
00.0
6100.0
0.1
131.5
1.9
30.4
01.4
<2.6
7.2
24.4
18.0
70.3
0
47
E2
1.4
13.7
817.6
39.0
50.1
9100.0
0.1
129.7
1.8
70.2
00.7
<0.1
0.6
23.2
33.6
60.0
2
48
S0
1.4
22.1
10.0
02.9
50.0
935.1
0.3
711.6
1.9
10.2
40.1
0.2
1.2
3.7
38.0
50.3
2
49
M82∗
1.5
14.9
00.0
05.3
00.0
7100.0
0.1
924.4
2.2
00.5
00.2
3.7
1.4
4.7
25.6
50.3
0
50
IRA
S22491-1
808∗
5.1
74.5
10.0
023.9
92.0
79.3
0.6
27.3
2.2
00.4
62.0
<20.5
49.4
430.3
216.6
00.1
1
51
S0
2.2
52.9
10.8
73.9
80.1
9100.0
0.1
339.6
1.7
80.9
61.5
0.4
5.9
9.8
6.5
10.6
0
c© 0000 RAS, MNRAS 000, 000–000
10 Ezhikode et. al.
10-2 10-1 100
λEdd
10-2
10-1
100
fc
100 101 102 103
LBol (1044 erg s−1)
10-1 100
LX− ray (1044 erg s−1)
10-2
10-1
100
fc
100 101 102
Ltorus (1044 erg s−1)
Figure 4. The variation of fc with λEdd (upper left), LBol (upper
right), LX−ray (lower left) and Ltorus (lower right).
5 DISCUSSION
The effect of the well-known receding torus model (Lawrence1991) predicts that fc decreases with increasing LBol. Whilstour data marginally supports this prediction we find astronger anti-correlation between fc and λEdd. So it isnot clear which parameter (or both) is the fundamentaldriver of the trends seen. This is made more complex asfc = Ltorus/LBol, so there is an implicit bias where fc willanti-correlate with LBol.
6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5log10 (MBH; M¯)
7.0
7.5
8.0
8.5
9.0
9.5
10.0
10.5
log 1
0(L
Bulg
e;L¯)
Magorrian
Figure 5. Plot showing the relationship between the bulge lumi-
nosity and the black hole mass. The circles are the data points andthe solid black line describes the Margorrian relationship (Magor-rian et al. 1998) for our data. The open circles with lower arrow
denote the sources for which we have considered the upper limits
of Lhost.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9fc
0
1
2
3
4
5
Nor
malise
dC
ount
simulationdata
Figure 6. Histogram of fc for the original sample (blue dashed
line) and the simulation (red solid line). Blue (dashed) and red(solid) vertical lines show the mean values of the respective dis-
tributions of fc.
0.6 0.4 0.2 0.0 0.2 0.4ρs
0
1000
2000
3000
Cou
nt
fc v/sLBol
simulationdata
0.0 0.2 0.4 0.6 0.8 1.0ds
2000
4000
6000
Cou
nt
Figure 7. Histogram of ρs (upper panel) and ds (lower panel) ofSpearman’s correlation between fc and LBol for the simulation.
The blue dashed line denotes ρs of the original data (upper panel),and ∼36% of the realisations lie below this showing stronger anti-
correlation than the original sample.
In order to investigate this, we carried out a simulationto predict the correlation between these parameters givenour sample. We generated 10000 realisations of random datawhere pairs of LBol and λEdd are drawn from the data us-ing bootstrap sampling with replacement and then assigneda redshift from the sample, z, again with replacement. Asample of 51 fc values is produced from a uniform distri-bution within the range of 0.0− 0.9, roughly correspondingto the observational range. This allows us to calculate theIR flux which would result from assuming a random dis-tribution of covering fraction. We applied an effective IRdetection threshold to the simulations, similar to what is
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 11
0.6 0.4 0.2 0.0 0.2 0.4 0.6ρs
0
1000
2000
3000
Count
fc v/sλEdd
simulationdata
0.0 0.2 0.4 0.6 0.8 1.0ds
1000
2000
3000
Count
Figure 8. Histogram of ρs (upper panel) and ds (lower panel)
of Spearman’s correlation between fc and λEdd for 10000 real-isations. Here, only ∼1% of the realisations show stronger anti-
correlation than the original data.
expected for the data. We use the standard 12 µm limitingflux of 1 mJy for the WISE all-sky survey as our threshold(Wright et al. 2010). If the simulated IR flux is more than1 mJy and the corresponding IR luminosity is within therange of the original data, the source is retained; otherwise,it is discarded and new values of LBol, λEdd, z and fc aredrawn. We checked that this gives rise to realisations withthe same distribution of Ltorus as found in the actual data.There is a difference at the level of 0.02 according to the K-Stest, which is only marginally significant. Further tests car-ried out with randomised log(fc), make the difference evenless significant. However, Fig. 6 shows that the distributionof fc in the data is very different from the simulation, wherean assumed initial random distribution is modified by theIR selection.
In each realisation, we calculate the Spearman’s corre-lation between LBol and fc. The rank coefficient of the orig-inal data was −0.27, very similar to the value of ρs ∼ −0.29seen in the simulation, and ∼36% of the realisations showan anti-correlation stronger than that found in the actualdata (see Fig. 7). So, in our sample, there is no significantanti-correlation between LBol and fc.
We repeat this procedure for λEdd and fc, but here theresults are very different. The rank coefficient of ∼ −0.43seen in the data is very different to the rank coefficient ofthe simulation, with only ∼1% of the realisations showingan anti-correlation stronger than that based on the originaldata (see Fig. 8), so this is significant at close to 3σ.
Thus fc in our sample is not significantly correlatedwith LBol, but it is with λEdd at ∼99% significance. This in-dicates that changes in the covering factor are driven moreby changes in the Eddington ratio, rather than by changesin the bolometric luminosity. This adds to a growing bodyof evidence that there are large-scale changes in the SEDwith λEdd (Vasudevan & Fabian 2007; Vasudevan & Fabian2009; Lusso et al. 2013, J12a, b). Therefore we find that themost basic of the unification models in which it is proposedthat the observed AGN properties only depend on inclina-
tion are too simple, and there are changes in the shape ofthe SED which depend on λEdd, as well as MBH which setsthe overall luminosity scale. However, the anti-correlationof the dust covering fraction with λEdd rather than LBol in-dicates a change in the larger scale geometry of the AGNrather than just the expected response of the dust to in-creasing illumination. Such a large-scale change may alsobe required to produce the observed anti-correlation of theforbidden series of the narrow emission lines with λEdd, asNarrow Line Seyfert 1s and other high λEdd AGN are knownto have weak [OIII] (e.g. Boroson & Meyers 1992; Done &Jin 2016). Furthermore, Leighly (2004) speculate that thisis due to the very inner regions of the accretion flow be-ing progressively shielded by a wind, with increasing λEdd.Thereby even if there is copious dust present, the irradiatedfraction decreases as the ionising radiation becomes morecollimated, and hence the reprocessed fraction drops. Fabianet al. (2008) have discussed the fact that efficient couplingof dust to gas boosts the effect of radiation pressure feed-back. The result is that absorbed AGN are mostly found atlow Eddington ratios. Here, we are seeing a decrease of the(illuminated) dust fraction in type 1 AGN. The effect couldbe related to that noted by Fabian et al. (2008) in absorbedAGN, with the feedback in our sample occurring out of ourdirect line of sight. Conversely, if the bulk of the MIR isemitted by dust located in the polar directions, then thisresult relates to the relative efficiency of illuminated dustemission in the line of sight.
6 CONCLUSIONS
We present a detailed study of the dust covering factorsfor an X-ray/optically selected sample of unobscured type 1AGN in the local Universe using the data available fromXMM-Newton, SDSS, WISE, 2MASS, and UKIDSS. Weused the method of SED modelling analysis to determinethe covering factor of each source. Two important aspectsof this work are that we have broadband spectra to deter-mine LBol, and a self-consistent model to estimate the con-tribution of the unobservable FUV region. We find a meancovering factor of fc = 0.30, with 0.02 < fc < 0.88 and adispersion of the individual values of σf = 0.17. The distri-bution shows a trend of anti-correlation with λEdd and LBol,but further analysis based on simulations shows that onlythe trend with λEdd is significant at ∼99%. This implies alarge-scale change in the geometry of the illuminated dust,rather than simply to a response from increasing LBol. Divi-sion into sub-samples of radio-loud AGN and NLS1, do notreveal any significant differences in the distribution of cover-ing factors from the whole sample. This argues against thepresence of a strong additional driving parameter for fc inthese sub-samples. However, the number of objects is small,and further studies are needed.
Our study is based on 51 sources, for which we havecomprehensive multi-wavelength coverage. It would be valu-able to extend this type of study to a larger sample of AGN,with a wider range of redshifts to test the correlations wefind and the conclusions that we have drawn. It is relevant toselect sources with different combinations of black hole massand accretion rate since the behaviour of the accretion discspectrum depends on these parameters. Further extension
c© 0000 RAS, MNRAS 000, 000–000
12 Ezhikode et. al.
Table 4. Spearman’s correlation between different parameters
Parameter 1 Parameter 2 ρs ds
λEdd MBH -0.17 0.24
λEdd LBol 0.61 1.64×10−06
λEdd LX−ray 0.28 0.04
λEdd Ltorus 0.439 0.001
λEdd κ2−10 0.54 4.49×10−05
λEdd Rcor -0.47 5.41×10−05
λEdd fpl -0.234 0.099
λEdd Γ 0.26 0.06λEdd fc -0.428 0.002
LBol LX−ray 0.76 7.8×10−11
LBol Ltorus 0.87 1.15×10−16
LBol κ2−10 0.34 0.02
LBol Rcor -0.368 0.008
LBol fpl -0.08 0.56LBol Γ -0.08 0.57
LBol fc -0.27 0.05MBH LBol 0.61 2.16×10−06
MBH LX−ray 0.69 1.71×10−08
MBH Ltorus 0.65 2.66×10−07
MBH κ2−10 -0.13 0.35
MBH Rcor 0.06 0.65
MBH fpl 0.11 0.46MBH Γ -0.383 0.005
MBH fc 0.076 0.596
LX−ray Ltorus 0.79 8.66×10−12
LX−ray κ2−10 -0.29 0.04
LX−ray Rcor 0.05 0.71
LX−ray fpl 0.27 0.06LX−ray Γ -0.34 0.01
LX−ray fc 0.007 0.96Ltorus κ2−10 0.16 0.26
Ltorus Rcor -0.234 0.098
Ltorus fpl 0.09 0.55Ltorus Γ -0.14 0.33
Ltorus fc 0.17 0.22
κ2−10 Rcor -0.65 2.31×10−07
κ2−10 fpl -0.52 7.89×10−05
κ2−10 Γ 0.434 0.001
κ2−10 fc -0.378 0.007Rcor fpl -0.07 0.61
Rcor Γ -0.31 0.03Rcor fc 0.32 0.02fpl Γ -0.29 0.04
fpl fc 0.3 0.03
Γ fc -0.18 0.21
of this study can be expected from AstroSat (Singh et al.2014) which can obtain simultaneous observations in X-rayand UV bands.
ACKNOWLEDGEMENTS
We thank the referee Konrad Tristram for his thoroughreview and useful comments. We acknowledge the UGC-UKIERI Thematic Partnership 2015 (UGC 2014-15/02) forthe support of the grant for this work. The first authoris grateful to the Department of Science and Technology(No.SR/S2/HEP-07/2012) for the financial support. P.G.acknowledges the support of STFC (No. ST/J003697/2). CDacknowledges support under STFC grant ST/L00075X/1.
This work is based on observations obtained withXMM-Newton, an ESA science mission with instrumentsand contributions directly funded by ESA Member Statesand NASA. This research has made use of the NASA/IPAC Infrared Science Archive, which is operated by the JetPropulsion Laboratory, California Institute of Technology,under contract with the National Aeronautics and SpaceAdministration. This publication makes use of data prod-ucts from the Wide-field Infrared Survey Explorer, which isa joint project of the University of California, Los Angeles,and the Jet Propulsion Laboratory/California Institute ofTechnology, funded by the National Aeronautics and SpaceAdministration. We have used the UKIDSS data from DataRelease 10. The UKIDSS project is defined in (Lawrenceet al. 2007). UKIDSS uses the UKIRT Wide Field Camera(WFCAM; Casali et al. (2007)). The photometric systemis described in Hewett et al. (2006), and the calibration isdescribed in Hodgkin et al. (2009). The pipeline process-ing and science archive are described in Irwin et al (2009,in prep) and Hambly et al. (2008). This publication makesuse of data products from the Two Micron All Sky Survey,which is a joint project of the University of Massachusettsand the Infrared Processing and Analysis Center/CaliforniaInstitute of Technology, funded by the National Aeronauticsand Space Administration and the National Science Foun-dation.
Funding for the SDSS and SDSS-II has been providedby the Alfred P. Sloan Foundation, the Participating In-stitutions, the National Science Foundation, the U.S. De-partment of Energy, the National Aeronautics and SpaceAdministration, the Japanese Monbukagakusho, the MaxPlanck Society, and the Higher Education Funding Councilfor England. The SDSS Web Site is http://www.sdss.org/.
The SDSS is managed by the Astrophysical ResearchConsortium for the Participating Institutions. The Partic-ipating Institutions are the American Museum of Natu-ral History, Astrophysical Institute Potsdam, University ofBasel, University of Cambridge, Case Western Reserve Uni-versity, University of Chicago, Drexel University, Fermilab,the Institute for Advanced Study, the Japan ParticipationGroup, Johns Hopkins University, the Joint Institute forNuclear Astrophysics, the Kavli Institute for Particle As-trophysics and Cosmology, the Korean Scientist Group, theChinese Academy of Sciences (LAMOST), Los Alamos Na-tional Laboratory, the Max-Planck-Institute for Astronomy(MPIA), the Max-Planck-Institute for Astrophysics (MPA),New Mexico State University, Ohio State University, Uni-versity of Pittsburgh, University of Portsmouth, PrincetonUniversity, the United States Naval Observatory, and theUniversity of Washington.
This research has made use of the NASA/IPAC Ex-tragalactic Database (NED) which is operated by the JetPropulsion Laboratory, California Institute of Technology,under contract with the National Aeronautics and SpaceAdministration.
References
Antonucci R., 1993, ARA&A, 31, 473Appenzeller, I., Thiering, I., Zickgraf, F.-J., et al. 1998,ApJS, 117, 319
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 13
Asmus D., Gandhi P., Honig S. F., Smette A., Duschl W. J.,2015, MNRAS, 454, 766
Asmus, D., Honig, S. F., & Gandhi, P. 2016, ApJ, 822, 109
Assef R. J. et al., 2013, ApJ, 772, 26
Barvainis, R. 1987, ApJ, 320, 537
Bessell M. S., 1991, A&A, 242, L17
Boroson, T. A., & Green, R. F. 1992, ApJS, 80, 109
Boroson, T. A., & Meyers, K. A. 1992, ApJ, 397, 442
Burtscher, L., Meisenheimer, K., Tristram, K. R. W., et al.2013, A&A, 558, A149
Cao X., 2005, ApJ, 619, 86
Cardelli J. A., Clayton G. C., Mathis J. S., 1989, ApJ, 345,245
Casali M. et al., 2007, A&A, 467, 777
Collinson, J. S., Ward, M. J., Landt, H., et al. 2016,arXiv:1610.04221
Cutri R. M. et al., 2003, 2MASS All Sky Catalog of pointsources.
de Grijp M. H. K., Keel W. C., Miley G. K., GoudfrooijP., Lub J., 1992, A&AS, 96, 389
Done C., Davis S. W., Jin C., Blaes O., Ward M., 2012,MNRAS, 420, 1848
Done, C., & Jin, C. 2016, MNRAS, 460, 1716
Dwelly T., Page M. J., 2006, MNRAS, 372, 1755
Eckart M. E., Stern D., Helfand D. J., Harrison F. A., MaoP. H., Yost S. A., 2006, ApJS, 165, 19
Elvis M. et al., 1994, ApJS, 95, 1
Fabian, A. C., Vasudevan, R. V., & Gandhi, P. 2008, MN-RAS, 385, L43
Gallagher S. C., Richards G. T., Lacy M., Hines D. C.,Elitzur M., Storrie-Lombardi L. J., 2007, ApJ, 661, 30
Gandhi P., Horst H., Smette A., Honig S., Comastri A.,Gilli R., Vignali C., Duschl W., 2009, A&A, 502, 457
Gandhi P. et al., 2011, ApJ, 740, L13
Greene, J. E., & Ho, L. C. 2004, ApJ, 610, 722
Grupe, D., Wills, B. J., Leighly, K. M., & Meusinger, H.2004, AJ, 127, 156
Gu M., 2013, ApJ, 773, 176
Haardt F., Maraschi L., 1993, ApJ, 413, 507
Hambly N. C. et al., 2008, MNRAS, 384, 637
Hao, L., Strauss, M. A., Fan, X., et al. 2005, AJ, 129, 1795
Hao H. et al., 2010, ApJ, 724, L59
Hasinger G., 2008, A&A, 490, 905
Hatziminaoglou E. et al., 2008, MNRAS, 386, 1252
Hatziminaoglou E., Fritz J., Jarrett T. H., 2009, MNRAS,399, 1206
Hewett P. C., Warren S. J., Leggett S. K., Hodgkin S. T.,2006, MNRAS, 367, 454
Ho L. C., Filippenko A. V., Sargent W. L. W., 1997, ApJ,487, 568
Hodgkin S. T., Irwin M. J., Hewett P. C., Warren S. J.,2009, MNRAS, 394, 675
Honig S. F., Kishimoto M., 2010, A&A, 523, A27
Honig, S. F., Kishimoto, M., Tristram, K. R. W., et al.2013, ApJ, 771, 87
Jin C., Done C., Ward M., 2016, MNRAS, 455, 691
Jin C., Ward M., Done C., Gelbord J., 2012a, MNRAS,420, 1825
Jin C., Ward M., Done C., 2012b, MNRAS, 425, 907
Kalberla, P. M. W., Burton, W. B., Hartmann, D., et al.2005, A&A, 440, 775
Khorunzhev, G. A., Sazonov, S. Y., Burenin, R. A., &Tkachenko, A. Y. 2012, Astronomy Letters, 38, 475
Kishimoto M., Honig S. F., Beckert T., Weigelt G., 2007,A&A, 476, 713
La Franca F. et al., 2005, ApJ, 635, 864
Lacy M., Petric A. O., Sajina A., Canalizo G., Storrie-Lombardi L. J., Armus L., Fadda D., Marleau F. R., 2007,AJ, 133, 186
Landt, H., Elvis, M., Ward, M. J., et al. 2011, MNRAS,414, 218
Lawrence A., 1991, MNRAS, 252, 586
Lawrence A., Elvis M., 1982, ApJ, 256, 410
Lawrence A., Elvis M., 2010, ApJ, 714, 561
Lawrence A. et al., 2007, MNRAS, 379, 1599
Leighly, K. M. 2004, ApJ, 611, 125
Lopez-Gonzaga, N., Burtscher, L., Tristram, K. R. W.,Meisenheimer, K., & Schartmann, M. 2016, A&A, 591,A47
Lusso E. et al., 2013, ApJ, 777, 86
Ma X.-C., Wang T.-G., 2013, MNRAS, 430, 3445
Magorrian J. et al., 1998, AJ, 115, 2285
Maiolino R., Shemmer O., Imanishi M., Netzer H., OlivaE., Lutz D., Sturm E., 2007, A&A, 468, 979
Mateos S. et al., 2012, MNRAS, 426, 3271
Mateos, S., Carrera, F. J., Alonso-Herrero, A., et al. 2016,ApJ, 819, 166
Mor, R., Netzer, H., & Elitzur, M. 2009, ApJ, 705, 298
Mor, R., & Trakhtenbrot, B. 2011, ApJ, 737, L36
Mullaney, J. R., Alexander, D. M., Goulding, A. D., &Hickox, R. C. 2011, MNRAS, 414, 1082
Nenkova M., Sirocky M. M., Nikutta R., Ivezic Z., ElitzurM., 2008, ApJ, 685, 160
Netzer H., 2013, The Physics and Evolution of ActiveGalactic Nuclei
Netzer H., 2015, ARA&A, 53, 365
Ogle P., Whysong D., Antonucci R., 2006, ApJ, 647, 161
Pier, E. A., & Krolik, J. H. 1992, ApJ, 401, 99
Polletta M. et al., 2007, ApJ, 663, 81
Press W. H., Teukolsky S. A., Vetterling W. T., FlanneryB. P., 1992, Numerical recipes in FORTRAN. The art ofscientific computing
Puchnarewicz, E. M., Mason, K. O., Siemiginowska, A., &Pounds, K. A. 1995, MNRAS, 276, 20
Richstone D. O., Schmidt M., 1980, ApJ, 235, 361
Risaliti G., Maiolino R., Salvati M., 1999, ApJ, 522, 157
Roseboom I. G., Lawrence A., Elvis M., Petty S., Shen Y.,Hao H., 2013, MNRAS, 429, 1494
Rush B., Malkan M. A., Spinoglio L., 1993, ApJS, 89, 1
Sanders D. B., Phinney E. S., Neugebauer G., Soifer B. T.,Matthews K., 1989, ApJ, 347, 29
Schlafly E. F., Finkbeiner D. P., 2011, ApJ, 737, 103
Silva L., Granato G. L., Bressan A., Danese L., 1998, ApJ,509, 103
Silva L., Maiolino R., Granato G. L., 2004, MNRAS, 355,973
Simpson C., 1998, MNRAS, 297, L39
Simpson, C. 2005, MNRAS, 360, 565
Singh K. P. et al., 2014, in Proc. SPIED, Vol. 9144,Space Telescopes and Instrumentation 2014: Ultravioletto Gamma Ray, p. 91441S
Skrutskie M. F. et al., 2006, AJ, 131, 1163
c© 0000 RAS, MNRAS 000, 000–000
14 Ezhikode et. al.
Stalevski M., Ricci C., Ueda Y., Lira P., Fritz J., Baes M.,2016, MNRAS, 458, 2288
Steffen A. T., Barger A. J., Cowie L. L., Mushotzky R. F.,Yang Y., 2003, ApJ, 596, L23
Stern D. et al., 2012, ApJ, 753, 30Suganuma M. et al., 2006, ApJ, 639, 46Toba Y. et al., 2013, PASJ, 65Treister E., Krolik J. H., Dullemond C., 2008, ApJ, 679,140
Treister E., Urry C. M., 2006, ApJ, 652, L79Tueller J. et al., 2010, ApJS, 186, 378Ueda Y., Akiyama M., Ohta K., Miyaji T., 2003, ApJ, 598,886
Urry C. M., Padovani P., 1995, PASP, 107, 803Vasudevan R. V., Fabian A. C., 2007, MNRAS, 381, 1235Vasudevan, R. V., & Fabian, A. C. 2009, MNRAS, 392,1124
Veron-Cetty, M.-P., Veron, P., & Goncalves, A. C. 2001,A&A, 372, 730
Veron-Cetty, M.-P., & Veron, P. 2006, A&A, 455, 773Williams, R. J., Pogge, R. W., & Mathur, S. 2002, AJ, 124,3042
Wright E. L. et al., 2010, AJ, 140, 1868Xu, D., Komossa, S., Zhou, H., et al. 2012, AJ, 143, 83
APPENDIX A: IR SED TEMPLATES
A1 agndust templates
The model agndust makes use of the templates from Silvaet al. (2004) who derived the nuclear infrared spectral en-ergy distributions for a sample of obscured and unobscuredSeyfert galaxies. They divided the observed IR SEDs intointervals of intrinsic absorbing column density NH. In or-der to exclude the objects with NH > 1025 cm−2 theseSEDs were already normalised by the unabsorbed hardX−ray (2−10 keV band) flux. They obtained four differentSEDs averaged within bins of absorbing NH. One amongthese nuclear IR SEDs corresponds to Seyfert 1 objectswith NH < 1022 cm−2 and the other three SEDs arefor Seyfert 2 galaxies with 1022 < NH < 1023 cm−2,1023 < NH < 1024 cm−2 and 1024 < NH < 1025 cm−2.The agndust model can make use of the four SEDs for mod-elling the IR data. In view of our sample selection, we makeuse of only the Seyfert 1 template by excluding the part ofthe SED at shorter wavelengths, plotted in blue in Fig. A1.
A2 hostpol templates
The hostpol component uses the IR SED templates fromSWIRE template library (Polletta et al. 2007). The SWIREtemplate library has 25 IR SED templates which cover thewavelength range between 1000A and 1000µm. The libraryconsists of 3 ellipticals, 7 spirals, 6 starbursts, 7 AGN and2 composite (starburst+AGN) templates. The AGN tem-plates comprise 3 type 1 and 4 type 2 AGN SEDs. The 13SWIRE templates we used for modelling the host galaxy IRemission using hostpol model (See Table A1) are plotted inFig. A2.
10-7 10-6 10-5 10-4 10-3 10-2 10-1
Energy (keV)
10-10
10-8
10-6
10-4
10-2
100
102
EF
E(keV
cm−
2s−
1)
Seyfert 1Seyfert 2 (22-23)Seyfert 2 (23-24)Seyfert 2 (24-25)
10-210-1100101102103104λ (µm)
Figure A1. The rest-frame Silva SED templates (Silva et al.2004) for Seyfert 1 and Seyfert 2 galaxies, normalised at 12 µm.
Seyfert 2 (22-23), Seyfert 2 (23-24) and Seyfert 2 (24-25) denoteSEDs with 1022 < NH <1023 cm−2, 1023 < NH <1024 cm−2
and 1024 < NH <1025 cm−2, respectively. We isolate the in-
frared hump of the Seyfert 1 SED to concentrate on the dustreprocessing component of the torus.
10-7 10-6 10-5 10-4 10-3 10-2 10-1
Energy (keV)
10-10
10-8
10-6
10-4
10-2
100
102
EF
E(keV
cm−
2s−
1)
S0SaSbScSd
E2E56062
82Ap2220
10-210-1100101102103104λ (µm)
Figure A2. The host galaxy SEDs from Polletta et al. 2007(flux density normalised at 5500 A) for spirals (S0−Sd), ellipticals
(E2 & E5) and starburst galaxies. The starburst templates corre-spond to the SEDs of NGC 6090 (60), NGC 6240 (62), M 82 (82),
Arp 220 (Ap), IRAS 22491-1808 (22), and IRAS 20551-4250 (20).
APPENDIX B: BROADBAND SED FITS
The broadband SED fits for the complete sample are shownin Fig. B1.
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 15
Table A1. hostpol model components and corresponding SED
templates of host galaxies used in our analysis (∗starburst galax-ies).
hostpol component SED template
host01 S0host02 Sa
host03 Sb
host04 Schost05 Sd
host06 E2host07 E5
host08 NGC 6090∗
host09 NGC 6240∗
host10 M 82∗
host11 Arp 220∗
host12 IRAS 22491-1808∗
host13 IRAS 20551-4250∗
c© 0000 RAS, MNRAS 000, 000–000
16 Ezhikode et. al.
Figure B1. The broadband SED fitting plots for the 51 sources. We fit the absorbed SEDs to the observed data and the resultant best-fit
models are shown in dashed grey line. However, we are mainly concerned with the measurements of LBol, for which we illustrate theintrinsic model (solid red) together with the deabsorption corrections applied to the data. The individual model components optxagnf,
agndust and hostpol are plotted in dashed blue, solid green and solid yellow, respectively. Data from XMM-Newton EPIC, Optical
Monitor, SDSS, UKIDSS/2MASS, and WISE are respectively represented by black dots, circles, diamonds, triangles, and squares. X-raydata have been rebinned for plotting purpose.
10-4
10-3
10-2No. 1, Object: UM 269 λEdd = 0.11
MBH = 41.0×107M¯LBol = 41.6×1044erg s−1
10-410-310-210-1100101102103
λ (µm)
10-4
10-3
10-2
No. 2, Object: Mrk 1018
λEdd = 0.08MBH = 6.92×107M¯LBol = 8.3×1044erg s−1
10-410-310-210-1100101102103
10-4
10-3
10-2
EF
E(keV
cm−
2s−
1) No. 3, Object: NVSS J030639
λEdd = 0.2MBH = 4.07×107M¯LBol = 10.1×1044erg s−1
10-4
10-3
10-2No. 4, Object: J074601.2+280732
λEdd = 0.01MBH = 60.0×107M¯LBol = 8.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
Energy (keV)
10-5
10-4
10-3
No. 5, Object: J080608.0+244421
λEdd = 0.97MBH = 8.71×107M¯LBol = 72.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
No. 6, Object: HS 0810+5157
λEdd = 0.29MBH = 31.4×107M¯LBol = 73.8×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 17
10-4
10-3
10-2
No. 7, Object: RBS 0769 λEdd = 0.990000MBH = 3.8×107M¯LBol = 42.7×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
λ (µm)
10-4
10-3
10-2
10-1 No. 8, Object: RBS 0770
λEdd = 0.29MBH = 3.47×107M¯
LBol = 14.1×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
10-3
10-2
10-1
No. 9, Object: Mrk 0110 λEdd = 0.65MBH = 2.51×107M¯LBol = 22.5×1044erg s−1
10-3
10-2
10-1No. 10, Object: PG 0947+396
λEdd = 0.38MBH = 21.9×107M¯LBol = 87.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
E
FE
(keV
cm−
2s−
1)
No. 11, Object: 2XMM J100025.2+015852
λEdd = 0.38MBH = 8.32×107M¯LBol = 22.8×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3No. 12, Object: 2XMM J100523.9+410746
λEdd = 0.05MBH = 6.61×107M¯LBol = 3.5×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
Energy (keV)
10-4
10-3
10-2
No. 13, Object: PG 1004+130
λEdd = 0.07MBH = 158.49×107M¯LBol = 104.5×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
No. 14, Object: RBS 0875
λEdd = 0.11MBH = 17.38×107M¯LBol = 19.2×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
18 Ezhikode et. al.
10-4
10-3
10-2
10-1No. 15, Object: KUG 1031+398
λEdd = 2.64MBH = 0.2×107M¯
LBol = 7.7×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
λ (µm)
10-4
10-3
10-2No. 16, Object: PG 1048+342
λEdd = 0.18MBH = 19.95×107M¯
LBol =39.6×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
10-5
10-4
10-3
No. 17, Object: 1RXS J111007
λEdd = 0.51MBH = 6.17×107M¯LBol = 30.9×1044erg s−1
10-3
10-2
10-1No. 18, Object: PG 1115+407
λEdd = 0.61MBH = 13.18×107M¯
LBol = 91.9×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
E
FE
(keV
cm−
2s−
1)
No. 19, Object: J112328.0+052823
λEdd = 0.04MBH = 5.1×107M¯LBol = 2.5×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2No. 20, Object: RX J1140.1+0307
λEdd = 0.19MBH = 0.63×107M¯LBol = 1.6×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
Energy (keV)
10-3
10-2
10-1No. 21, Object: PG 1202+281
λEdd = 0.46MBH = 9.6×107M¯
LBol = 48.6×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
No. 22, Object: 1AXG J121359+140
λEdd = 0.08MBH = 6.9×107M¯LBol = 6.5×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 19
10-5
10-4
10-3
No. 23, Object: 2E 1216+0700
λEdd = 0.02MBH = 10.0×107M¯
LBol = 2.6×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
λ (µm)
10-5
10-4
10-3
No. 24, Object: 1RXS J122019 λEdd = 0.15MBH = 18.0×107M¯
LBol =23.9×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
10-4
10-3
10-2
No. 25, Object: LBQS 1228+1116
λEdd = 0.33MBH = 26.92×107M¯
LBol = 87.2×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-6
10-5
10-4
10-3 No. 26, Object: J123126.4+105111
λEdd = 0.76MBH = 5.01×107M¯LBol = 34.5×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
E
FE
(keV
cm−
2s−
1)
No. 27, Object: Mrk 0771
λEdd = 0.16MBH = 7.24×107M¯LBol = 15.8×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-7
10-6
10-5
10-4
10-3
No. 28, Object: RX J1233.9+0747
λEdd = 0.67MBH = 9.12×107M¯
LBol =50.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
Energy (keV)
10-5
10-4
10-3
10-2No. 29, Object: RX J1236.0+2641
λEdd = 0.46MBH = 7.41×107M¯LBol = 35.6×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
No. 30, Object: PG 1244+026
λEdd = 0.2MBH = 1.86×107M¯LBol = 5.2×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
20 Ezhikode et. al.
10-6
10-5
10-4
10-3No. 31, Object: J125553.0+272405
λEdd = 0.06MBH = 50.0×107M¯LBol = 19.9×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
λ (µm)
10-5
10-4
10-3
No. 32, Object: RBS 1201
λEdd = 0.11MBH = 4.17×107M¯LBol = 5.7×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
10-6
10-5
10-4
10-3 No. 33, Object: J132101.4+340658
λEdd = 0.33MBH = 8.32×107M¯LBol = 23.3×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
No. 34, Object: 1RXS J132447
λEdd = 0.06MBH = 51.0×107M¯LBol = 28.6×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
E
FE
(keV
cm−
2s−
1)
No. 35, Object: UM 602
λEdd = 0.9MBH = 4.64×107M¯LBol = 41.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
No. 36, Object: 1E 1346+26.7 λEdd = 0.2MBH = 1.0×107M¯LBol =2.8×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
Energy (keV)
10-5
10-4
10-3
10-2 No. 37, Object: PG 1352+183
λEdd = 0.18MBH = 17.0×107M¯LBol = 37.2×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
10-2 No. 38, Object: Mrk 0464 λEdd = 0.01MBH = 6.17×107M¯LBol = 1.1×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 21
10-5
10-4
10-3
10-2 No. 39, Object: 1RX J135724
λEdd = 0.08MBH = 2.19×107M¯LBol = 2.2×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
λ (µm)
10-4
10-3
10-2
No. 40, Object: PG 1415+451
λEdd = 0.2MBH = 7.59×107M¯LBol = 19.2×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
10-4
10-3
10-2
No. 41, Object: PG 1427+480
λEdd = 0.59MBH = 13.8×107M¯LBol = 83.8×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2No. 42, Object: NGC 5683 λEdd = 0.02
MBH = 5.5×107M¯LBol = 1.3×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
E
FE
(keV
cm−
2s−
1)
No. 43, Object: RBS 1423
λEdd = 0.36MBH = 12.0×107M¯LBol = 46.2×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
No. 44, Object: PG 1448+273
λEdd = 0.71MBH = 3.63×107M¯LBol = 35.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
Energy (keV)
10-4
10-3
10-2
No. 45, Object: PG 1512+370
λEdd = 0.42MBH = 60.26×107M¯LBol = 212.0×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
No. 46, Object: Q 1529+050 λEdd = 0.06MBH = 36.0×107M¯LBol = 24.4×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
22 Ezhikode et. al.
10-4
10-3
10-2
No. 47, Object: 1E 1556+27.4 λEdd = 0.19MBH = 9.05×107M¯LBol = 23.2×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
λ (µm)
10-4
10-3
10-2
No. 48, Object: Mrk 0493 λEdd = 0.09MBH = 2.95×107M¯LBol = 3.7×1044erg s−1
102 101 100 10-1 10-2 10-3 10-4
10-4
10-3
10-2
No. 49, Object: IIZw 177 λEdd = 0.07MBH = 5.3×107M¯LBol = 4.7×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-5
10-4
10-3
10-2
No. 50, Object: PG 2233+134λEdd = 2.07MBH = 23.99×107M¯LBol = 430.3×1044erg s−1
10-5 10-4 10-3 10-2 10-1 100 101
10-4
10-3
10-2
No. 51, Object: Mrk 0926
λEdd = 0.19MBH = 3.98×107M¯
LBol = 9.8×1044erg s−1
c© 0000 RAS, MNRAS 000, 000–000
Determining the torus covering factors for a sample of type 1 AGN in the local Universe 23
APPENDIX C: NOTES ON INDIVIDUALSOURCES
C1 Discrepancies in SDSS data
The SDSS data points in the SEDs of the objects Mrk 0110(No. 9), PG 0947+396 (No. 10), PG 1202+281 (No. 21),LBQS 1228+1116 (No. 25), PG 1244+026 (No. 30),RBS 1201 (No. 32), PG 1415+451 (No. 40), NGC 5683(No. 42), PG 1512+370 (No. 45) and PG 2233+134 (No. 50)show some clear deviations from the broadband continuummodel. This may be attributed to the intrinsic variability ofthe sources. Since the observations of SDSS and OM are notsimultaneous, there may be some discrepancy between thetwo data sets. The SED of Mrk 0110 shows large offset inthe SDSS data points. J12a discussed this source and men-tioned that it is an extreme example of this behaviour. Inorder to check the influence of the discrepancy, we fit thesesources with the same model without using the SDSS data.We observed that fc remain unchanged in 8 sources, andin only two cases (PG 0947+396 (No. 10) & PG 1202+281(No. 21)), there is a drop by a factor of ∼2.
C2 Sources with multiple hostpol templates
In the case of Mrk 0110 (No. 9) and PG 1115+407 (No. 18),the normalisations of all hostpol templates are nearly zeroand are equally fitted by multiple host galaxy templates.For Mrk 0110, all the hostpol templates provide the samefit-statistic and even the same spectral parameters. How-ever, Mrk 0110 is identified as an Sa galaxy (Khorunzhevet al. 2012). Hence, in the main text, we mentioned theSED of spiral galaxy type-a (host02) as the best-fit hostpolmodel for this object. In the case of PG 1115+407, morethan one hostpol template gave the same χ2 and slightlydifferent parameter values. However, the morphological typeof the source is unknown. In this case, we have adopted S0template as the best-fit component.
C3 Super-Eddington sources
There are two sources in our sample, KUG 1031+398(No. 15) and PG 2233+134 (No. 50), which have super-Eddington accretion rates. Among these KUG 1031+398 hasthe highest value of λEdd (∼2.64) and lowest black hole mass,whereas PG 2233+134 has the highest value of bolometricluminosity. In these sources, we attempt to fit the data byfixing λEdd to 1 and letting MBH be a free parameter. ForKUG 1031+398 this resulted in an improvement in the fit(∆χ2 = -58.5) for a change in MBH from 1.7×106M to3.8×106M. Here, the covering factor increased from 0.23to 0.34. As discussed by Jin et al. (2016) this may be a super-Eddington source and the black hole mass obtained from theSED-fitting may not be correct. In such sources, the super-Eddington flow may not be well fit by a model that con-serves energy. In the case of PG 2233+134, MBH remainedunchanged while χ2(/dof) increased from 893.6(/173) to2641.2(/173). In this case also, fc shows an increase from0.11 to 0.22. Although the inclination effects are not takeninto account, it seems likely that this AGN is indeed a super-Eddington source. It is perhaps surprising that we are able tofit it as a super-Eddington source since such high Eddington
sources probably power strong winds. So energy conserva-tion is not appropriate due to loss of radiative power to thewind.
APPENDIX D: OTHER TORUS TEMPLATES
Mullaney et al. (2011) (hereafter M11) have constructed arange of intrinsic MIR to FIR (6-100 µm) SEDs of a sampleof X-ray selected local AGN with moderate luminosities. Wehave fitted our sample with M11 SED by extending it downto about 0.6 µm to match the range of our agndust (Silvaet al. 2004) template for Seyfert 1. The covering factors ob-tained with this template have a very similar range, meanand scatter as that for agndust. Although the SED fits andspectral parameters are comparable to that of agndust, formost of the sources the fit resulted in comparatively poorχ2. This is probably due to fact that the extended M11 SEDis narrow and so fails to cover the peak in emission aroundthe NIR region.
We have also attempted fitting the data with ClumpySED for type 1 AGN (Honig & Kishimoto 2010) with incli-nation 30 degrees, by modifying the template to match thewavelength range of agndust. But again we note that thetemplate is narrow compared to agndust SED, and peaksaround 10µm. Also, the torus luminosity for this compo-nent over 1−1000 µm is a factor of ∼1.3 lower than thatfor agndust SED. The fits provide poor statistic for ∼70%of the sources in the sample and a lower range of coveringfactors ∼ 0.01− 0.5 with a mean around 0.18.
The data appear to require an extra hot dust componentin the NIR, not covered by the above templates. We testedthis for UM 269 (No. 1) by adding a black body component(XSPEC model bbody) with temperature ∼1000−1500 K.For this source, a black body with ∼1000 K temperature inthe NIR region, in addition to the Clumpy SED provided abetter χ2 and resulted in a similar covering factor (∼0.31) aswe obtained when using agndust. Hence the overall resultsare consistent with those obtained before.
The histograms of covering factors obtained for agndust,extended M11 and Clumpy SEDs are shown in Fig. D1.
APPENDIX E: ANALYSIS WITH 2MASS DATA
As mentioned in Section 2.1 we opted to use NIR data fromUKIDSS instead of 2MASS because of the smaller aper-ture size of the UKIDSS camera (2′′diameter for UKIDSS &4′′radius for 2MASS), which therefore reduces the contribu-tion from the host galaxy. Also, three sources in our samplelack 2MASS data and in those cases we require UKIDSSdata. However, for comparison, we have analysed the entiresample by fitting the broadband SED with 2MASS data (ifavailable) for the NIR band. We find that the distribution ofcovering factors is similar to that obtained when using theUKIDSS data. Both distributions of fc have the same meanand standard deviation. The histograms of fc are shown inFigure E1. We have also checked the correlation of fc withLBol and λEdd obtained for 2MASS data. The trend of fc be-tween these parameters is essentially the same as we foundwhen using the UKIDSS data.
c© 0000 RAS, MNRAS 000, 000–000
24 Ezhikode et. al.
0.0 0.2 0.4 0.6 0.8 1.0fc =Ltorus/LBol
0
5
10
15
20
25
30
Cou
nt
agndust
Clumpy
Mullaney
Figure D1. Histograms of covering factors for agndust SED (thick black line), extended M11 SED (dashed red line) and Clumpy SED
for type 1 AGN (thin green line). The vertical lines represent the mean values of fc for different distributions.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9fc =Ltorus/LBol
0
5
10
15
20
Cou
nt
2MASSUKIDSS
Figure E1. Distributions of covering factors obtained using 2MASS data (dashed line) and UKIDSS data (solid line) for the NIR band.
c© 0000 RAS, MNRAS 000, 000–000