Peculiar Transients in the Era of LSST: Observed Populations, … · 2017-06-07 · Chapter 6: Transients Characterization, ... Luminous red novae t Prospects GW Wave Follow-up Oﬀ-axis

PeculiarTransientsintheEraofLSST:ObservedPopulations,Nature,andIntrinsicRatesSupernovae:TheLSSTRevolu1onJune1st,2017MariaR.DroutHubble,Carnegie-DunlapFellow

LSSTTransientsandVariableStarsWorkingGroup:FastTransientsSub-groupmaincontact:MariaR.DroutSub-groupMAFExpert:EricBellm

1.  Transientswithoverallshort1mescales

2.  Short1mescaleemission

componentsassociatedwithlongerlivedtransients

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PeculiarTransientsPeculiardoesnot(necessarily)meanintrinsicallyrare.Whatwediscoverisacombina1onofwhattheuniverseproducesandhowwearelooking.

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Type Ia Type Ib/c Type II

WD disruption Massive Star Core-Collapse

Supernovae

PeculiarTransientsEverythingelse…

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

TransientSearches12 M. R. Drout et al.

High-zsearches(SNLS,SDSS,ESSENCE)

PTF,PS1Calan/Tololosurvey LOSS

(AdaptedfromMarkSullivan)

SNPhaseSpace2 M. R. Drout et al.

Fig. 1.—

2 M. R. Drout et al.

Fig. 1.—

SNPhaseSpace

2 M. R. Drout et al.

Fig. 1.—

SNPhaseSpace

PS1 Type IIn SN 3

Fig. 2.—

SNPhaseSpace

SNPhaseSpace

Chapter 6: Transients

Characterization, on the other hand, implies that basic properties of the event may be determinedfrom the LSST observations, including but not limited to the classification of the event. It isparticular relevant for science cases requiring analysis of large samples of completed lightcurves.

Characterization and classification of transient events benefits from substantial temporal samplingover the finite duration of the event along with color information (perhaps contemporaneous).Transient events slower than ⇠ weeks may be adequately sampled by a uniform LSST cadence.Obtaining adequate sampling for faster-evolving events may require special scheduling strategies.For some event types, LSST can only be expected to provide a discovery service, and followupwill necessarily be performed elsewhere—so long as the cadence is su�cient to identify the eventtype. For some events, such as detecting electromagnetic counterparts to gravitational wave events(GWs), serendipitous discoveries are unlikely, but enabling a ToO program would provide theopportunity for LSST to contribute significantly to this science.

We consider a non-exhaustive set of “astronomical transients” in the paragraphs that follow. Fora few of these transients, we quantify the ability of LSST cadences to produce data useful forvarious science goals. These case studies include SNe, GRBs, and GWs. For other transientfamilies (Novae, LBVs, TDEs) we provide more general information in Section 6.6, and we invitethe community to further develop the ideas proposed here, as well as further other related goals,into quantified science cases.

6.1.1 Targets and Measurements

Table 6.1 is a non-exhaustive list of phenomena to which we are referring as Eruptive and Explosivetransients in this document.

10−2 10−1 100 101 102

−30

−25

−20

−15

−10

−5

Characteristic Timescale [day]

Peak

Lum

inos

ity [M

V]

PTF11agg

iPTF13dsw

iPTF13bxl

iPTF14ybiPTF13ekl

iPTF14aue

iPTF14bfu

iPTF14cvaiPTF14cyb

Thermonuclear Supernovae

Classical Novae

Luminous Red Novae

Intermediate Luminosity Red Transients

Core−Collapse Supernovae

Luminous Supernovae Relativistic Explosions

.Ia Explosions

Calcium−rich Gap Transients

1038

1040

1042

1044

1046

Peak

Lum

inos

ity [e

rg s−1

]

Figure 6.1: Peak luminosity-characteristic decay timescale plot for explosive transients (adapted from Kasliwal2011).

144

LSSTObservingStrategyWhitePaper(adaptedfromKasliwal2011)

PeculiarTransients:ObservationsObservedTransient PossibleProgenitor/

ExplosionMechanism(s)Why“peculiar”

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PeculiarTransients:ObservationsObservedTransient PossibleProgenitor/

ExplosionMechanism(s)Why“peculiar”

Super-luminousSN V.massivestar,ExtremeMassLoss,Magnetar

Intrinsicallyrare;preferlow-masshosts

Nuclear(op1cal)flares TidalDisrup1onEvents,ExtremeAGNFlares

rare,imagesubtrac1ondifficult,AGNconfusion

LuminousBlueOp1calTransients(FBOTs)

… rapid

RapidlyDecliningTypeISN

WDHeliumDetona1on,Ultra-StrippedCCSN

faint,rapid

TypeIaxSN WDDeflagra1on faint,somewhatrapid

Calcium-richTransients WD-WD,NS-WD faint,somewhatrapid

Intermediateluminosityop1caltransients(ILOTs)

LBV-likeerup1on,ElectronCaptureSN

veryfaint

Luminousrednovae Mergers,Common-EnvelopeEjec1on,EC-SN

veryfaint

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PeculiarTransients:TheoryTheoreDcalTransients PossibleObservedCases ScienceCase(s)

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PeculiarTransients:TheoryTheoreDcalTransients PossibleObservedCases ScienceCase(s)

TidalDisrup1onEvents Op1calnuclearflares;X-rayflares

ProbeofdormantSMBH;jetphysics

Kilonovae(NS-NSmerger)

GRB130603B GWlocaliza1on;r-processnucleosynthesis

Off-axisGRBaderglows … Beamingfrac1on:rates,energy;jetstructure

WDAccre1onInducedCollapse

… WDphysics

Detona1onofaheliumshellonaWD(.IaSN)

RapidTypeISN WDaccre1on,nuclearphysics

Failedsupernovae … BHforma1on;feedback

Ejec1onofastellarenvelope

Luminousrednovae;ILOTs

Massloss;common-envelopephysics

…

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PeculiarTransients•  Whataretheobservedpopula:ons?•  Whatistheirnature?•  Whataretheirintrinsicrates?

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout


•  Whataretheimplica:ons?stellarevolu1on,fateofstellarsystems,binaryinterac1ons,physicsofcompactobjectsandexplosions,chemicalenrichment

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout


Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Margueetal.(2014)

PeculiarTransientsObservedTransient

Super-luminousSN

Nuclear(op1cal)flares



TypeIaxSN

Calcium-richTransients

IntermediateLuminosityOp1calTransients(ILOTs)

Luminousrednovae

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Prospects

GWWaveFollow-up

PeculiarTransientsObservedTransients

Super-luminousSN

Nuclear(opDcal)flares



TypeIaxSN



Luminousrednovae

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Prospects

GWWaveFollow-up

Super-LuminousSN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Explosion Mechanisms and Energy Sources Powering SUPER-LUMINOUS SUPERNOVAE

1. Scientific Rationale: Transient surveys recently led to the discovery of Super-Luminous Supernovae (SLSNe, Quimby 2011, Chomiuk 2011), a new and diverse class of stellar explosions with extreme peak luminosities reaching Lpeak~1045 erg s-1 (M<-20 mag), ~10-100 times larger than common SNe (Fig. 1). These huge luminosities require extreme properties of the progenitor stars (e.g., ZAMS mass >100 M¤; Gal-Yam 2009, 2012), exotic explosion mechanisms (pair instability; Kasen 2011) and/or sources of energy that are different from those of ordinary SNe, including magnetars as central engines (Kasen & Bildsten 2010; Woosley 2010) or shock interaction with a dense H and/or He-rich medium (Smith 2007, Quimby 2011, Chevalier 2011). Associated to the death of the most massive stars, SLSNe sample the extreme of the initial mass function and allow us to explore the new physical mechanisms that lead to their explosion. Furthermore, with high luminosities for an extended period of time (~weeks, Fig 1) and exceptionally blue colors, we showed that SLSNe are useful cosmic beacons that serve as probes of the interstellar medium of distant galaxies (Berger 2012). Finally, the study of SLSNe is motivated by their potential role in the chemical evolution of the Universe and to its re-ionization.

Figure 1 SLSNe outshine ordinary explosions by a factor 10-100 in luminosity and mark the death of the most massive stars. SLSNe share extreme peak luminosities but otherwise show impressive spectroscopic diversity, ranging from type Ic-like explosions with no evidence for H or He (e.g. SN2005ap, Quimby 2007), to type IIn-like SNe with strong signs of circumstellar interaction (e.g. SN2006gy; Smith 2007). Following standard practice, SLSNe can be divided into two groups: SLSN-II, with clear evidence for H in their spectra, and SLSN-I that instead lack spectroscopic signatures of H (Fig. 1). The extremely rare SLSNe-Ir have been proposed to be powered by the radioactive decay of 56Ni (Gal-Yam 2012, Fig. 1), a picture that was questioned by

Nicholl 2013. The most compelling question pertaining SLSNe is the nature of the energy source that powers their exceptional luminosity. The proposed sources are (i) shock breakout into a dense H-poor shell (relevant for type-I SLSNe, Chevalier 2011); (ii) interaction with an H-rich shell (type IIn-like explosions) or (iii) a magnetar central engine (Kasen & Bildsten 2010; Woosley 2010). Motivated thus, here we propose a focused Swift/UVOT and XRT program as part of our multi-wavelength (radio, NIR, optical) effort aimed to: (1) Pin down the energy source of SLSNe. (2) Map the diversity of their progenitor stars and pre-explosion evolution. This effort requires knowledge of the (copious) UV emission that characterizes most SLSNe. In the following we show that UV observations are crucial to (i) accurately determine the bolometric luminosity and radiated energy; (ii) measure the temporal evolution of the temperature and radius of the photosphere; (iii) search for evidence of line blanketing caused by large amounts of Nickel. At the same time, Swift/XRT observations are of primary importance to (i) reveal the presence of a magnetar central engine and (ii) constrain the presence of a Gamma-Ray Burst -like jet. This effort is timely: a number of SLSNe are being discovered every year. Complementing our approved programs on Chandra, XMM, optical/NIR facilities (MMT, MDM, Magellan, UKIRT, LBT, Bok, Kuiper), and proposed radio (VLA), Swift will allow us to study SLSNe across the spectrum. 1.1 SLSNe-Ir: early identification of pair instability explosions SLSNe-Ir have been proposed to be powered by large amounts of radioactive 56Ni synthesized during the explosion (MNi>5 M¤ for SN2007bi, Gal-Yam 2009). The extremely large 56Ni mass strongly suppresses the UV flux due to line blanketing, so that early time UV to optical colors will clearly distinguish between the very blue, central engine/circum stellar medium powered SLSNe (as suggested by Nicholl 2013) and the redder, Ni-powered SLSNe (Gal-Yam 2009). SLSNe-Ir have been suggested to be the first example of stellar explosions undergoing electron-positron pair production instability and collapse (e.g. Gal-Yam 2009) as opposed to the standard iron-core collapse. Pair instability only occurs in massive oxygen cores with M>50M

! which implies a total

initial mass M≥ 200 M!

, far above the Galactic limit of 150 M

!. UV to optical colors therefore hold the

keys to understand the energy source of SLSNe and to the early identification of pair instability explosions, which are responsible for the death of the most massive stars in our Universe.

1.2 The source of energy powering SLSN-I SLSNe-I are characterized by hot early spectra with copious UV flux extending for weeks after the explosion and no H or He feature (e.g. Quimby 2011, Lunnan 2013,

Twoclasses:hydrogen-rich,hydrogen-poor.Probesofevolu1onofhighmassstars,exo1cexplosionmechanisms,extrememass-lossProbesofdistantISMRoleinchemicalevolu1on,re-ioniza1on.

Image:R.Margue

TidalDisruptionEvents

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

ProbesofquiescentSMBHs:masses,sub-pcenvironmentsTwoclasses:op1calflares,X-rayflaresProbejetphysicsLSST:coulddiscoverthousandsofTDEs/yrChallenges:dis1nguishingfromothernucleartransients


Super-luminousSN


LuminousBlueOpDcalTransients(FBOTs)


TypeIaxSN



Luminousrednovae

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Prospects

GWWaveFollow-up

Off-axisGRBaderglows

LuminousandBlueTransients

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PS1 Type IIn SN 3

Fig. 2.—


Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

PS1 Rapidly-Evolving Transients 3

Fig. 1.— PS1 absolute magnitude, rest-frame, light curves for gold sample transients. Circles represent grizP1 detections and trianglesrepresent 3σ upper limits. Vertical dashed lines indicate epochs when spectroscopic observations were acquired. The grey shaded region isthe R−band type Ibc template from Drout et al. (2011), normalized to the peak magnitude of the PS1-MDS transient.

Fig. 2.— Same as Figure 1 for silver sample objects.

Fig. 3.— PS1 apparent magnitude, observer-frame, light curves for our bronze (non-spectroscopic) sample. Symbols have the samemeaning as Figure 1.

Ia, type Ib/c and type II SN. In V-band, typical type Iaand type Ib/c SN will rise by 0.6 − 0.75 mag in the 10days immediately prior to maximum light and decline by1.0 − 1.2 mag in the 25 days post-maximum (Riess et al.1999; Drout et al. 2011; Li et al. 2011). At higher red-shifts PS1 will probe bluer wavelengths (which typicallyevolve more rapidly) but this will compete with the ef-fects of time dilation. Many type II SN would easily passthe rise time requirement, but they decline by !0.5 mag

in the 25 days post maximum (Li et al. 2011; Hamuy2003). The most rapid type Ic SN (e.g. SN1994I; Rich-mond et al. 1996) and most rapid SN1991bg-like type IaSN (Taubenberger et al. 2008) would barely pass all threeof our requirements if caught at maximum light. How-ever, any given PS1 band is only observed every threedays and we require the same band to pass both ourrise and decline cuts. We therefore expect our selected

Drout,M.R.etal.(2014)


Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout


SampleProperDes:•  Luminous•  BlueColors•  SpectraDominatedbyCon1nua•  Starforminghostgalaxies


Fig. 12.— Left Panel: Pseudo-bolometric light curves for thegold and silver transients. Right Panel: Pseudo-bolometric lightcurves for other rapidly-evolving transients from the literature: thetype Ic SN 2005ek (Drout et al. 2013) and 2010X (Kasliwal et al.2010), the type Ib SN 2002bj (Poznanski et al. 2010), the type IIbSN 1993J (Schmidt et al. 1993), and the type IIn PTF09uj (Ofeket al. 2010).

by our highest redshift event, and the approximate rangeof our spectra in Section 5.Using this formulation, our peak pseudo-bolometric lu-

minosities span a range of approximately 2×1042 ergs s−1

< L < to 3×1043 ergs s−1, and are plotted in Figure 7.If we had instead utilized a UV bolometric correctionthat integrated the entire best fit blackbody, these valueswould be approximately a factor of 2 higher. In the leftpanel of Figure 12 we plot our derived pseudo-bolometriclight curves. The number of epochs for which we can con-strain the luminosity is limited for the silver transients.The energy radiated by the six gold transients between−3 and +20 days ranged from 2 ×1048 ergs (PS1-12bb)to 2 ×1049 ergs (PS1-11qr).For comparison, we also show the pseudo-bolometric

light curves of several other rapidly-evolving events (rightpanel): SN 2002bj (Poznanski et al. 2010), PTF09uj(Ofek et al. 2010), SN2010X (Kasliwal et al. 2010) andSN2005ek (Drout et al. 2013), as well as the doublepeaked SN1993J (Schmidt et al. 1993). The curvesfor SN2005ek, SN2002bj, and SN1993J were calculatedfrom multi-band photometry in a manner similar tothat utilized here, while the curves for SN 2010X andPTF09uj were derived based on r-band data only. Wenote that SN2005ek and SN2010X, which have been pro-posed to be powered mainly by radioactive decay (Droutet al. 2013; Kasliwal et al. 2010; Tauris et al. 2013),are less luminous than the PS1-MDS events, while themore luminous PTF09uj and the first peak of SN1993Jare thought to be powered by cooling envelope emis-sion/interaction. These power sources will be discussedfurther in Section 8.

4000 6000 8000 10000

PS1-12bv (+2d)

PS1-12bb (+5d)

PTF 09uj (+2d)

SN2002bj (+7d)

PS1-11bbq (+2d)

Rest Wavelength (!)

Norm

aliz

ed F

lux

(fλ)

+ C

onst

ant

HαHβ[OII]MgII

PS1-12bb (+33d)

PS1-12brf (+4d)

SN1993J (-18d)

PS1-13duy (+2d)

SN2005ek (-1d)

SN2010X (+23 d)

SN2007gr (-4d)

SN2011fe (-15d)

Fig. 13.— Explosion spectra for five PS1-MDS transients (col-ored) in comparison to events from the literature (black). With theexception of PS1-12bb our events are dominated by a blue contin-uum, with a lack of strong P Cygni features. Some contributionsfrom the host galaxy (e.g. nebular emission lines) are still presentin these events. PS1-12bb shows a redder continuum and a no-table lack of broad nebular features at +33 days. The top panelshows literature objects thought to be powered by cooling enve-lope emission/interaction while the bottom panel shows objectspowered by radioactive decay (type Ib SN 2002bj Poznanski et al.2010; type IIb SN 1993J Barbon et al. 1995; type IIn PTF09ujOfek et al. 2010; type Ia SN 2011fe Pereira et al. 2013; type IcSN2007gr Valenti et al. 2008; type Ic SN 2005ek Drout et al. 2013;type Ic SN2010X Kasliwal et al. 2010).

5. SPECTROSCOPIC PROPERTIES

We obtained spectra during outburst for five tran-sients: PS1-11bbq, PS1-12bb, PS1-12bv, PS1-12brf, andPS1-13duy. These spectra are plotted in Figure 13. Onespectrum was obtained between two and four days postmaximum for each event and, in the case of PS1-12bb, asecond spectrum was obtained at +33 days. The spectrashown for PS1-11bbq, PS1-12bv, PS1-12brf, and PS1-13duy still contain some contribution from their hostgalaxies, as evidenced from presence of nebular emissionlines. However, both the lack of a 4000 A break and thefaint apparent magnitude of all four hosts (23 − 25 mag)give us confidence that a majority of the continuum isdue to the transient itself.

5.1. Basic Properties and Comparison to other Events

From Figure 13 we see that the spectra are dominatedby continua as opposed to strong P Cygni features. Fourof the events, PS1-11bbq, PS1-12bv, PS1-12brf and PS1-

ImplicaDons/Progenitors:•  Shockbreakout/coolingfromextendedstellarenvelopeordensewind

•  Winds/ouhlowsfromcompactobjects


Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Maria R. Drout – Bok Postdoctoral Fellowship – Research Interests 2

Figure 1: Left: Timescale versus luminosity phase space for SN explosions. Red stars (highlighted region) representthe newly discovered, but prevalent, rapidly-evolving transients. Right: A template R-band light curve of one classof rapidly-evolving transient located at 100 Mpc. Due to the distance of all previously discovered events (z∼0.2)only the optical/UV emission in the ∼30 days post explosion are known (green shaded region). However, late timephotometric/spectroscopic and multi-wavlength (radio/X-ray) observations for objects at smaller distances can beused to distinguish between proposed explosion scenarios. Dashed colored lines represent a range of models for thelate-time behavior.

date were either recognized after-the-fact or exploded at distances too large (z∼0.2) to obtain meaningfullate-time followup (See Figure 1).

This highlights the need for an optimized survey with real-time detection to increase the number ofevents with multi-wavelength follow-up. I propose to develop such a program, with a trial version to becarried out during my tenure as a Bok Fellow using the Dark Energy Camera (DECam) at the NationalOptical Astronomy Observatory in Chile. Using the software I created to measure rates/efficiencies withinPanSTARRs I am carrying out a detailed study of the optimal observing strategy to both detect andcharacterize rapidly-evolving transients at small distances. I am also part of a collaboration performing acustom survey for electromagnetic followup for gravitational wave triggers using DECam—for which a realtime reduction and alert pipeline was already created. I am developing follow-up programs on the VeryLarge Array radio observatory, the Chandra X-ray observatory, and ground based optical telescopes—allof which myself and my collaborators are extensively using at the present. I will also investigate the use ofArizona’s moderate aperture telescopes for this survey and will work with collaborators at on-going surveys(ASAS-SN, Z/PTF, BlackGEM) to understand their efficiency detecting rapid transients and follow-upand study the sub-galactic environments of any nearby examples discovered. This comprehensive approachwill allow me to constrain the nature of these new explosions.

Notably, these transients are also predicted to be one of the main contaminants when searching forelectro-magnetic counterparts to gravitational wave sources [8]. I am part of a collaboration using DECamfor follow-up of aLIGO sources, and my work containing the properties of these events has been animportant component in the development of our candidate selection criteria.

Rateofblue,luminouseventsareapprox.5500-8000events/Gpc3/year(4-7%ofthecore-collapserateatz=0.2;BoPcellaetal.2008)


Fig. 19.— PS1-MDS rapid-transient detection efficiencies for ourrate calculation. Solid lines indicate detection efficiencies for in-dividual PS1-MDS fields as a function of redshift (bottom axis)or luminosity distance (top axis). A black solid line indicates thedetection efficiencies for the survey as a whole, which peak around11% at z=0.1. The dashed black line indicates the distance atwhich we expect our detections to be located. This peaks at z=0.2.The actual distances of the PS1-MDS transients are shown as ar-rows on the bottom axis.

short-duration GRBs (Fong & Berger 2013, dashed black;SGRBs), and LGRBs (Bloom et al. 2002, solid black).In the right panel of Figure 18 we normalize these off-

sets by the g−band half light radius of the host. Alsoshown are type Ia SN (Galbany et al. 2012, blue), typeIbc and type II SN (Kelly & Kirshner 2012, green andcyan, respectively) and the same LGRB and SGRB sam-ples as described above. For objects with high signal tonoise SDSS detections, we use the Petrosian half-lightradius from SDSS DR9. For others (PS1-11bbq, PS1-13duy, and PS1-13ess), we estimate the half light radiusbased on the light within a 3′′ radius of the centroid lo-cation within the PS1 deep stacks (we caution that thesevalues are more uncertain). Both the physical and nor-malized offsets are listed in Table 5. Our sample mostclosely resembles the normalized offsets of type Ibc andtype II SN, indicating that, at least in this context, ourexplosion sites are consistent with the environments inwhich one can expect to find massive stars.

7. VOLUMETRIC AND RELATIVE RATES

In this section we use information on the true cadenceand sensitivity obtained by the PS1-MDS between Dec.2009 and March 2014 to calculate the volumetric rate ofrapidly-evolving transients similar to those presented inthis work. Because these objects were identified based ontheir light curve morphology, rather than spectroscopicclassification, we avoid the observer-dependent bias thatresults because resources limit our spectroscopic follow-up to roughly 10% of the transients discovered in thePS1-MDS.Because their timescale is rapid, even short observing

gaps due to poor weather or maintenance can have asignificant effect on the efficiency with which the surveydetects these transients. We therefore use a Monte Carloapproach similar to that utilized in Quimby et al. (2012,

2013) to calculate the efficiency with which each PS1-MDS field can recover these transients as a function ofdistance.We begin by constructing light curve and temperature

evolution templates based on the gold transients (Fig-ure 1). We then construct a luminosity function whichconsists of an intrinsic gaussian distribution modified byan exponential function to account for host galaxy ab-sorption. In following with Quimby et al. (2013) weadopt P(AV) ∝ eAV/τ with τ = 0.6 (Hatano et al. 1998)for the host galaxy absorption. As a starting point for theintrinsic distribution we adopt a gaussian with the samemean and variance as the (rest-frame) peak magnitudesof our gold sample after performing a V/Vmax correctionfor the Malmquist bias. As a check we performed severalMonte Carlo simulations distributing 500 objects drawnfrom this distribution evenly in space. The objects withapparent magnitudes > 1.5 mag above the nominal PS1detection limit (a rough proxy for detectability) agreewell with our observed transients in both luminosity anddistance. The nominal values for the intrinsic (i-band)gaussian distribution we adopt below are µ = −17.2 magand σ = 1.0 mag.For each PS1-MDS field we then determine the detec-

tion efficiency within 140 distance bins between z=0.005and z=0.7. We performed 10,000 iterations of the fol-lowing process within each distance bin: (i) randomlyselect one of the six templates; (ii) choose a peak ab-solute magnitude from the aforementioned luminositydistribution; (iii) choose a random distance within thedistance bin; (iv) k-correct the template light curves,assuming the spectra can be approximated as a black-bodies described by the temperature evolution template;(v) choose a random explosion epoch between Dec. 2009and March. 2014; (vi) map the resulting explosion ontothe actual dates PS1 observed each filter for that PS1-MDS field; (vii) assign random noise to each epoch that ispulled from the actual distribution of noises obtained bytransients found in the PS1-MDS on that epoch in thatfilter; and finally (viii) run the resulting synthetic datathrough the same light curve selection criteria describedin Section 2. The efficiency within each distance bin isthen calculated as the fraction of the 10,000 iterationsthat were recovered.In Figure 19 we plot the resultant efficiencies for each

PS1-MDS field (colored lines) as well as the survey as awhole (solid black line). The survey detection efficienciesrange from ∼11% near z ∼ 0.1 to !0.1% at z ∼0.7. Thequoted efficiencies are for the entire duration of the PS1survey (Dec. 2009 to March 2014), while any given PS1-MDS field is only observed for roughly half of this timeperiod. The in-season efficiency at z∼ 0.1 for the surveywas approximately 25%. This demonstrates how muchinfluence small, in-season, observing gaps can have onthe detection of such rapid transients.Using these efficiencies we calculate the volumetric rate

of these rapid and luminous transients as:

R =N

!ϵiViti

(1)

where ϵi, Vi, and ti are the efficiency, co-moving volume,and proper time within each distance bin and N is thenumber of transients actually detected in the PS1-MDS.


DetectionEfIiciency&IntrinsicRatesLuminousandBlueTransients

IntheEraofLSSTLuminousandBlueTransients

Whatdoweneed:Moreevents:ideallynearbyLuminosityfunc1on:implica1onsforratesHowdowegettherewithLSST:Triggeredfollow-uporstand-alongobserva1ons?


Super-luminousSN




TypeIaxSN



Luminousrednovae

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Prospects

GWWaveFollow-up


Fastand/orFaintTypeISN

FastandFaintTypeISN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

The SN Ia varietyThe Extremes of Thermonuclear Supernovae 3

Fig. 1 Phase space of potentially thermonuclear transients. The absolute B-band magnitude at peakis plotted against the light-curve decline rate, expressed by the decline within 15 d from peak inthe B band, Dm15(B) (Phillips, 1993). The different classes of objects discussed in this chapterare highlighted by different colours. Most of them are well separated from normal SNe Ia in thisspace, which shows that they are already peculiar based on light-curve properties alone. The onlyexception are 91T-like SNe, which overlap with the slow end of the distribution of normal SNe Ia,and whose peculiarities are almost exclusively of spectroscopic nature. References to individualSNe are provided in the respective sections.

The Extremes of Thermonuclear Supernovae 3


Taubenberger 2017MB,max

Δm15(B)

FastandFaintTypeISN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

TableCourtesyofKenShen(UCB)

FastandFaintTypeISN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Calcium-richSN:

Kasliwal+(2012)Lyman+(2016)

Proper1es:FaintHelium-richNon-standardabundancesUnusualloca1onsImplica1ons:WDbinaryphysicsChemicalcomposi1onofclustersSignificantcontribu1onofTi.

FastandFaintTypeISN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

TypeIaxSN:

Proper1es:FaintCO-burningproductsinspectraRela1velylowejectaveloci1esProgenitordetec1onconsistentwithaHe-starcompanion.Possibleprogenitor:Deflagra1onofaWDaccre1ngfromcompanion.

Foley+(2013)McCully+(2014)

FastandFaintTypeISN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Rapidly-decliningTypeISNThe SN Ia varietyThe Extremes of Thermonuclear Supernovae 3


The Extremes of Thermonuclear Supernovae 3


Taubenberger 2017MB,max

Δm15(B)

FastandFaintTypeISNRapidly-decliningTypeISNIntroduction � Galactic binary transients

Many stages where transients occur….

SN explosions

Intermittent pulsars, RRATS, magnetars, FRBs?

SFXTs Be/X-ray ULXs

Common envelope contact binary mergers

Ultra-stripped SNe

X-ray transients

LIGO sources

Geodetic precession

This talk

ImagecourtesyofT.Tauris

FastandFaintTypeISN

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Type KnownMembers IntrinsicRate

Calcium-rich ~10 ~5%ofTypeIa(Perets+2010)

TypeIax ~30 ~20-40%ofTypeIa(Foley+2013)

Rapidly-decliningTypeI

~2-4 ~1%ofTypeIa(Drout+2013)

IntheEraofLSST

Issuesandques1ons:•  Howmuchvolumeactuallyexistsatsmallenoughdistance?•  Doweneedarollingcadenceinthemainsurvey?Isthateven

sufficient?•  Thesetransientsneedspectroscopicclassifica1on.


Super-luminousSN




TypeIaxSN


IntermediateLuminosityOpDcalTransients(ILOTs)

Luminousrednovae

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Prospects

GWWaveFollow-up


“SNImposters”

SNImposters:

•  Drama1cmasslossepisodes(erup1ons,commonenvelopeejec1ons).

•  Faint(electroncapture)orweak(failed)SN

•  LSSTwilldras1callyincreasethedistancetowhichwecanseeoutburstsatthislevel.

•  Progenitorvariabilityfornearbyexamples.

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Erup1ons,mergers,andmore

Smithetal.(2009)


Super-luminousSN




TypeIaxSN



Luminousrednovae

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Prospects

GWWaveFollow-up

ProspectsI:GravitationalWaveFollow-up

•  PossibleE&McounterpartstoGWsources:kilonovae:probablyfaintandred(maybefastandblue)

•  LSSTcanbeverypowerfulifthereisToOcapability.

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Bergeretal.(2013)

PeculiarTransientsandLSST•  Realaqen1onshouldbepaidtocadences,inter-nightpairfilters,especiallyinthecontextoffasttransients.

•  RealpowerforLSSTwillbeinnumbers:sta1s1calproper1esofsub-popula1ons,beqerconstraintsonintrinsicrates.•  Willrequireunderstandingofselec1onbiases/follow-upbiases.•  Willdobeqerfortransientclassesthatdonotneedspectroscopicfollow-uptoclassify.

•  Realeffortstoobtainspectroscopicredshids/classifica1onofquiescentsources.

•  Forfaint,butnumerousandlonger-1mescaleexplosionswillprovideexcellentinsightintopre-erup1vevariability/colors.

Supe

rnovae:The

LSSTRe

volu1o

nMariaR.D

rout

Documents

Peculiar Transients in the Era of LSST: Observed Populations, … · 2017-06-07 · Chapter 6: Transients Characterization, ... Luminous red novae t Prospects GW Wave Follow-up Oﬀ-axis