The Needle in the 100 deg2 Haystack: Uncovering Afterglows of Fermi GRBs with the Palomar Transient Factory

3.5 MB·English

by  Leo P. Singer

THEASTROPHYSICALJOURNAL,806:52
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
THENEEDLEINTHE100deg2HAYSTACK:UNCOVERINGAFTERGLOWSOFFERMIGRBsWITHTHEPALOMAR
TRANSIENTFACTORY
LEOP.SINGER1,32,2, MANSIM.KASLIWAL3,S.BRADLEYCENKO2,4,DANIELA.PERLEY33,5,GEMMAE.ANDERSON6,7,G.C.
ANUPAMA8,IAIRARCAVI9,10,VARUNBHALERAO11,BRIAND.BUE12,YICAO5,VALERIECONNAUGHTON13,ALESSANDRACORSI14,
ANTONINOCUCCHIARA32,2,ROBP.FENDER6,7,DEREKB.FOX15,NEILGEHRELS2,ADAMGOLDSTEIN32,16,J.GOROSABEL17,18,19,
ASSAFHORESH20,KEVINHURLEY21,JOELJOHANSSON22,D.A.KANN23,24,CHRYSSAKOUVELIOTOU16,KUIYUNHUANG25,S.R.
KULKARNI5,FRANKMASCI26,PETERNUGENT27,28,ARNERAU24,UMAAD.REBBAPRAGADA12,TIMD.STALEY6,7,DMITRY
SVINKIN29,C.C.THÖNE17,A.DEUGARTEPOSTIGO17,30,YUJIURATA31,ANDALANWEINSTEIN1
(Received2015January2;Accepted2015February28;Published2015June8)
TheAstrophysicalJournal,806:52
5
1
ABSTRACT
0
TheFermiGamma-raySpaceTelescopehasgreatlyexpandedthenumberandenergywindowofobservations
2
ofgamma-raybursts(GRBs). However,thecoarselocalizationsoftenstoahundredsquaredegreesprovided
n bytheFermiGRBMonitorinstrumenthaveposedaformidableobstacletolocatingthebursts’hostgalaxies,
u
measuring their redshifts, and tracking their panchromatic afterglows. We have built a target-of-opportunity
J
modefortheintermediatePalomarTransientFactoryinordertoperformtargetedsearchesforFermiafterglows.

9 Here, wepresenttheresultsofoneyearofthisprogram: 8afterglowdiscoveriesoutof35searches. Twoof
theburstswithdetectedafterglows(GRBs130702Aand140606B)wereatlowredshift(z=0.145and0.384
] respectively)andhadspectroscopicallyconfirmedbroad-lineTypeIcsupernovae. Wepresentourbroadband
E
follow-upincludingspectroscopyaswellasX-ray,UV,optical,millimeter,andradioobservations. Westudy
H
possible selection effects in the context of the total Fermi and Swift GRB samples. We identify one new
. outlierontheAmatirelation. Wefindthattwoburstsareconsistentwithamildlyrelativisticshockbreaking
h
out from the progenitor star, rather than the ultra-relativistic internal shock mechanism that powers standard
p
- cosmologicalbursts. Finally,inthecontextoftheZwickyTransientFacility,wediscusshowwewillcontinue
o toexpandthisefforttofindopticalcounterpartsofbinaryneutronstarmergersthatmaysoonbedetectedby
r AdvancedLIGOandVirgo.
t
s Subject headings: gamma-ray burst: individual (GRB 130702A, GRB 140606B) — supernovae: general —
a
methods: observational—surveys—gravitationalwaves
[

4
v
5 [email protected]
9 1LIGOLaboratory, CaliforniaInstituteofTechnology,Pasadena, CA
4 91125,USA
0 2AstrophysicsScienceDivision,NASAGoddardSpaceFlightCenter, 18UnidadAsociadaGrupoCienciaPlanetariasUPV/EHU-IAA/CSIC,
Code661,Greenbelt,MD20771,USA Departamento de Física Aplicada I, E.T.S. Ingeniería, Universidad del
0 3ObservatoriesoftheCarnegieInstitutionforScience,813SantaBar- País-VascoUPV/EHU,AlamedadeUrquijos/n,E-48013Bilbao,Spain
1. baraStreet,PasadenaCA91101,USA 19Ikerbasque,BasqueFoundationforScience,AlamedadeUrquijo36-
4JointSpace-ScienceInstitute,UniversityofMaryland,CollegePark, 5,E-48008Bilbao,Spain
0
MD20742,USA 20Benoziyo Center for Astrophysics, Weizmann Institute of Science,
5 5Cahill Center for Astrophysics, California Institute of Technology, 76100Rehovot,Israel
1 Pasadena,CA91125,USA 21SpaceSciencesLaboratory,UniversityofCalifornia-Berkeley,Berke-
: 6Astrophysics, Department of Physics, University of Oxford, Keble ley,CA94720,USA
v Road,OxfordOX13RH,UK 22TheOskarKleinCentre,DepartmentofPhysics,StockholmUniver-
i 7Physics & Astronomy, University of Southampton, Southampton sity,SE-10691Stockholm,Sweden
X SO171BJ,UK 23Thüringer Landessternwarte Tautenburg, Sternwarte 5, D-07778
8IndianInstituteofAstrophysics,Koramangala,Bangalore560034,In-
r Tautenburg,Germany
a dia9LasCumbresObservatoryGlobalTelescopeNetwork,6740Cortona 24Max-Planck Institut für Extraterrestrische Physik, Giessenbach-
strasse1,D-85748Garching,Germany
Dr.,Suite102,Goleta,CA93117,USA
10Kavli Institute for Theoretical Physics, University of California, 25DepartmentofMathematicsandScience, NationalTaiwanNormal
SantaBarbara,CA93106,USA University,Lin-kouDistrict,NewTaipeiCity24449,Taiwan
11Inter-UniversityCentreforAstronomyandAstrophysics(IUCAA), 26InfraredProcessingandAnalysisCenter,CaliforniaInstituteofTech-
PostBag4,Ganeshkhind,Pune411007,India nology,Pasadena,CA91125,USA
12Jet Propulsion Laboratory, California Institute of Technology, 27Department of Astronomy, University of California, Berkeley, CA
Pasadena,CA91109,USA 94720-3411,USA
13CSPAR and Physics Department, University of Alabama in 28PhysicsDivision,LawrenceBerkeleyNationalLaboratory,Berkeley,
Huntsville,320SparkmanDrive,Huntsville,AL35899,USA CA94720,USA
14Texas Tech University, Physics Department, Lubbock, TX 79409- 29IoffePhysical-TechnicalInstitute,Politekhnicheskaya26,StPeters-
1051,USA burg194021,Russia
15DepartmentofAstronomyandAstrophysics,PennsylvaniaStateUni- 30DarkCosmologyCentre,NielsBohrInstitute,JulianeMariesVej30,
versity,UniversityPark,PA16802,USA CopenhagenØ,DK-2100,Denmark
16Astrophysics Office, ZP12, NASA Marshall Space Flight Center, 31InstituteofAstronomy,NationalCentralUniversity,Chung-Li32054,
Huntsville,AL35812,USA Taiwan
17InstitutodeAstrofísicadeAndalucía(IAA-CSIC),GlorietadelaAs- 32NASAPostdoctoralFellow
tronomías/n,E-18008,Granada,Spain 33HubbleFellow
2 Singeretal.
1. INTRODUCTION the instrumental pipeline composed of the Palomar 48 inch
Oschintelescope(P48;Rahmeretal.2008),theroboticPalo-
Deepsynopticopticalsurveys,includingthePalomarTran-
mar 60 inch telescope (P60; Cenko et al. 2006), and associ-
sient Factory (PTF; Law et al. 2009; Rau et al. 2009) and
atedspectroscopicresourcesincludingthePalomar200inch
Pan-STARRS(Kaiseretal.2010), haverevealedawealthof
Haletelescope(P200).
new transient and variable phenomena across a wide range
In Singer et al. (2013b), we announced the first discov-
ofcharacteristicluminositiesandtimescales(Kasliwal2011).
Withawide(7deg2)instantaneousfieldofview(FOV),mod- ery of an optical afterglow based solely on a Fermi GBM
localization.34 That explosion, GRB130702A/iPTF13bxl,
eratelydeepsensitivity(reachingR=20.6magin60s),acon-
was noteworthy for several reasons. First, it was detected
sortiumoffollow-uptelescopes,sophisticatedimagesubtrac-
by Fermi LAT. Second, it was at moderately low redshift,
tionandmachinelearningpipelines,andaninternationalteam
z=0.145,yethadpromptenergeticsthatbridgedthegapbe-
of human-in-the-loop observers, PTF has been a wellspring
tween“standard,”brightcosmicallydistantburstsandnearby
of new or rare kinds of explosive transients (for instance,
sub-luminous bursts and X-ray flashes. Third, due to its
Quimbyetal.2011;Kasliwaletal.2012)andearly-timeob-
lowredshift,anaccompanyingSNwasspectroscopicallyde-
servations of supernova (SNe) or their progenitors (see, for
tectable.
example, Nugent et al. 2011; Corsi et al. 2012; Ofek et al.
Inthiswork,webeginwithadetaileddescriptionoftheop-
2013; Gal-Yam et al. 2014). PTF has even blindly detected
eration of the iPTF GRB afterglow search. We then present
the optical emission (Cenko et al. 2014; S. B. Cenko et al.,
seven more GBM–iPTF afterglows from the first 13 months
in preparation) from the rarest, brightest, and briefest of all
of this project. In each of the eight cases, the association
known cosmic explosions, GRBs, hitherto only discoverable
between the optical transient and the GRB was proven by
with the aid of precise localizations from space-based gam-
the presence of high-redshift absorption lines in the opti-
ma-ray observatories. PTF has also detected explosions that
cal spectra and the coincident detection of a rapidly fading
optically resemble GRB afterglows but may entirely lack
X-ray source with Swift XRT. In two cases, the positions
gamma-rayemission(Cenkoetal.2013b).
were further corroborated by accurate Fermi LAT error cir-
GRBsandtheirbroadbandafterglowsarenotoriouslychal-
cles, and in four cases by accurate InterPlanetary Network
lengingtocapture. Theynaturallyevolvefrombrighttofaint,
(IPN)triangulationsinvolvingdistantspacecraft. Inonecase
and from high (gamma- and hard X-ray) to low (optical and
(GRB140508A),theIPNtriangulationwasperformedrapidly
radio) photon energies, with information encoded on energy
andwasinstrumentalinselectingwhichopticaltransientcan-
scalesfrom1to1016GHz(Perleyetal.2014d)andtimescales
didates to follow up. In six cases, radio afterglows were de-
from 10−3 to 107 s. Only with a rapid sequence of handoffs
tected. Ourdiscoveryrateof8outof35eventsisconsistent
between facilities graded by energy passband, FOV, and po-
with the ages and searched areas of the GBM bursts, com-
sitionaccuracyhavewebeenabletofindthem,pinpointtheir
binedwiththeluminosityfunctionofopticalafterglows.Con-
hostgalaxies,andconstraintheirphysics. TheSwift mission
sequently,bytilinglargerareasand/orstackingexposures,the
(Gehrelsetal.2004),withits1.4sr-wide(50%coded)Burst
iPTFafterglowsearchshouldbeabletoscaletocoarserlocal-
Alert Telescope (BAT; Barthelmy et al. 2005) and its ability
izations and fainter optical signals, such as those associated
toslewandtrainitsonboardX-rayTelescope(XRT;Burrows
withshortGRBs.
etal.2005)andUV/OpticalTelescope(UVOT;Romingetal.
Next,wepresentextensivefollow-upobservations,includ-
2005) on the location of a new burst within 100 s, has tri-
ingR-bandphotometryfromtheP48,multicolorphotometry
umphedhere: innineyearsofoperation,ithastrackeddown
from the P60, spectroscopy (acquired with the P200, Keck,
700X-rayafterglowsandenabledextensivepanchromatic
≈ Gemini, APO, Magellan, Very Large Telescope (VLT), and
observations by a worldwide collaboration of ground-based
GTC), and radio observations with the Karl G. Jansky Very
opticalandradiofacilities.
Large Array35 (VLA), the Combined Array for Research in
Meanwhile,theFermisatellitehasopenedupanewenergy
Millimeter-wave Astronomy (CARMA; Bock et al. 2006;
regime extending up to 300 GeV, with the Large Area Tele-
Corder et al. 2010), the Australia Telescope Compact Array
scope(LAT;Atwoodetal.2009)detectinghigh-energypho-
(ATCA; Frater et al. 1992), and the Arcminute Microkelvin
tonsforaboutadozenburstsperyear. TheGamma-rayBurst
Imager (AMI; Zwart et al. 2008). We provide basic physi-
Monitor (GBM; Meegan et al. 2009), an all-sky instrument
cal interpretations of the broadband spectral energy distribu-
sensitivefrom8keVto40MeV,detectsGRBsprolificallyat
tions (SEDs) of these afterglows. We find that seven of the
arateof 250yr−1,withalargenumber(about44yr−1)be-
events are consistent with the classic model of synchrotron
≈
longingtotherarershort,hardbursts(Paciesasetal.2012a).
cooling of electrons that have been accelerated by a single
Although LAT can provide localizations that are as accurate
forward shock encountering either the constant-density cir-
as 10(cid:48), Fermi GBM produces error circles that are several
cumburstinterstellarmedium(ISM;broadbandbehaviorpre-
∼
degreesacross. SincemostburstsseemtolackGeVemission
dictedinSarietal.1998)orastellar(i.e.,Wolf–Rayet)wind
detectable by LAT, most Fermi GBM bursts do not receive
environment (Chevalier & Li 1999). The possible excep-
deep,broadbandfollow-up. Consequently,theirredshiftsand
tion, GRB140620A/iPTF14cva, can probably be explained
the properties of their afterglows have remained largely un-
by standard extensions of this model, a reverse shock or an
known.
inverseComptoncomponent.
As part of the intermediate Palomar Transient Factory
(iPTF), over the past year we have developed the ability to 34 There are two earlier related cases. The optical afterglow of
rapidlytilethese 100deg2 GBMerrorcirclesandpinpoint GRB090902BwasdetectedexpostfactointiledobservationswithRobotic
the afterglows. T∼his target-of-opportunity (TOO) capability OpticalTransientSearch(ROTSE)about80minutesaftertheburst,butthe
afterglowwasinitiallydiscoveredwiththehelpofanX-raydetectioninSwift
uses and briefly redirects the infrastructure of our ongoing
observationsoftheLATerrorcircle. GRB120716AwasidentifiedbyiPTF
synoptic survey (the operation of which is discussed in Gal- bysearchinga≈2deg2IPNerrorbox(Cenkoetal.2012).
Yametal.2011), notablythemachinelearningsoftwareand 35http://www.vla.nrao.edu
TheNeedleinthe100deg2Haystack 3
Two of the afterglows (GRB130702A/iPTF13bxl and a well-known empirical prescription of the systematic errors
GRB140606B/iPTF14bfu)fadedawaytorevealspectroscop- of the GBM localization. Paciesas et al. (2012b) state that
ically detected broad-line Type Ic SNe (SNeIc-BL). Despite thetotaleffectiveerrorradiusintheFERMI_GBM_FIN_POS
theabundantphotometricevidenceforSNeinafterglowlight localizations is well described by the quadrature sum of the
curves (see Li & Hjorth 2014 and references therein), the statistical radius and a systematic contribution, where the
distinction of SN spectroscopy has been shared by scarcely systematic is 2.◦6 for 72% of bursts and 10.◦4 for 28% of
tens36 out of 800 long Swift bursts in nine years of opera- bursts. We use the weighted rms of these two values, r =
sys
tion. ≈ (cid:112)0.72(2.◦6)2+0.28(10.◦4)2 6◦. The total error radius is
We estimate the kinetic energies of the relativistic blast (cid:112) ≈
then r = r 2+r 2. We construct a Fisher–von Mises
waves of these events from their X-ray afterglows (Freed- eff stat sys
distribution, centered on the best-estimate position, with a
man & Waxman 2001). We find that although the
concentrationparameterof
gamma-ray energetics of these eight bursts are broadly
similar to the Swift sample, two low-luminosity bursts (cid:104) (cid:16) π (cid:17)(cid:105)−1
κ= 1−cos r . (1)
(GRBs130702A and 140606B) have significantly lower ki- 180◦ eff
neticenergies.Wediscussthepossibilitythatthesetwobursts
With the FERMI_GBM_FIN_POS alert, the Fermi GBM
arisenotfromastandardultra-relativisticinternalshock,but
teamalsodistributesadetailedlocalizationmapthataccounts
fromamildlyrelativisticshockasitbreaksoutfromthepro-
for the systematic effects (Connaughton et al. 2015). The
genitorstar(see,forexample,Nakar&Sari2012).
TOO Marshal retrieves from the Fermi data archive a file
We conclude by discussing prospects for targeted optical
that describes the 1σ, 2σ, and 3σ significance contours. If
transient searches in wide areas. This is especially relevant
thelocalizationhassignificantasymmetry,wealsoretrievea
for optical counterparts of gravitational wave (GW) events.
2D FITS image whose pixel values correspond to the GBM
Weillustratethatopticalafterglowsofshortbursts,whichare
localization significance, and use this instead of the Fish-
intimatelylinkedtotheprimesourcesfortheAdvancedLaser
er–vonMisesdistribution.
Interferometer GW Observatory (LIGO) and Virgo, should
Giving preference to fields for which deep co-added ref-
bewellwithinthereachofasimilarapproachusingZwicky
erence images exist, the TOO Marshal selects 10 P48 fields
Transient Facility (ZTF; Kulkarni 2012; Bellm 2014; Smith
etal.2014). spanninganareaof 72deg2tomaximizetheprobabilityof
≈
enclosingthetrue(butasyetunknown)locationofthesource,
2. SEARCHMETHODOLOGY assumingtheabovedistribution.
The Marshal then immediately contacts a team of humans
We begin by describing our TOO observations and after-
(the authors) by SMS text message, telephone, and e-mail.
glowsearchstepbystep.
Thehumansaredirectedtoamobile-optimizedwebapplica-
2.1. AutomatedTOOMarshal: AlertsandTiling tiontotriggertheP48(seeFig.11).
A program called the iPTF TOO Marshal moni-
2.2. TriggeringtheP48
tors the stream of Gamma-ray Coordinates Network
(GCN) notices37 from the three redundant, anonymous Withintheaboveconstraints, wedecidewhethertofollow
NASA/GSFC VOEvent servers. It listens for notices of up the burst based on the following criteria. The event must
type FERMI_GBM_GND_POS, sent by GBM’s automated be(cid:46)12hroldwhenitfirstbecomesobservablefromPalomar,
on-ground localization, or FERMI_GBM_FIN_POS, sent by andwemustcoverenoughoftheerrorcircletohavea(cid:38)30%
theGBMburstadvocate’shuman-in-the-looplocalization.38 chance of enclosing the position of the source. We discard
Uponreceivingeitherkindofnotice,theTOOMarshalde- anyburststhataredetectedandaccuratelylocalizedbySwift
termines if the best-estimate sky position is observable from BAT,becausethesearemoreefficientlyfollowedupbycon-
Palomar at any time within the 24 hr after the trigger. The ventional means. We also give preference to events that are
criterionforobservabilityisthatthepositionisatanaltitude out of the Galactic plane and that are observable for at least
>23.◦5 (i.e. airmass (cid:46)2.5), at least 20◦ from the center of 3hr.
themoon,atanhouranglebetween 6.h5,andthattheSunis There are some exceptional circumstances that override
atleast12◦belowthehorizonatPal±omar. theseconsiderations. Iftheburst’spositionestimateisacces-
Ifthepositionisobservableandthe1σ statisticalerrorra- sible within an hour after the burst, we may select it even if
dius r reported in the GCN notice is less than 10◦, the theobservabilitywindowisverybrief.Iftheburstisverywell
stat
localizedorhasthepossibilityofasubstantiallyimprovedlo-
TOO Marshal selects a set of 10 P48 fields that optimally
cover the error region.39 It converts the GBM position es- calizationlaterduetoaLATorIPNdetection,wemayselect
itevenifitisintheGalacticplane.
timate and radius into a probability distribution by applying
The default observing program is three epochs of P48 im-
36 Betweenphotometric,late-timeredbumpsandunambiguousspectral ages at a 30-minute cadence. The human may shorten or
identifications,therearealsoGRB–SNethathavesomeSN-associatedspec- lengthen the cadence if the burst is very young or old (see
tralfeatures. ThenumberofGRBswithspectroscopicSNeis,therefore,ill
thediscussionofEquation(2)inSection2.4below), change
defined. SeeHjorth&Bloom(2012,p. 169,andreferencestherein)fora
the number of epochs, or add and remove P48 fields. When
morecompletecensus.
37http://gcn.gsfc.nasa.gov thehumanpressesthe“Go”button,theTOOMarshalsendsa
38 Usually,theFermiteamsuppressesthenoticesiftheburstisdetected machine-readablee-mailtotheP48robot. Therobotaddsthe
andlocalizedmoreaccuratelybySwiftBAT. requestedfieldstothenight’sschedulewiththehighestpos-
39WemadeoneexceptiontoourGBMerrorradiuscutoff: wefollowed siblepriority,ensuringthattheyareobservedassoonasthey
upGRB140219A,whichhadaGBMerrorcirclewitharadiusof12.8◦,but
arevisible.
hadanIPNlocalizationspanning0.6deg2 (Hurleyetal.2014a). Despite
searchingabout80%oftheIPNpolygon,wedetectednoafterglow(Singer
2.3. AutomatedCandidateSelection
etal.2014d).Thisispotentiallyadarkburstcandidate.
4 Singeretal.
As the night progresses, the TOO Marshal monitors the brighter than the 1σ limiting magnitude of the exposures.
progress of the observations and the iPTF real-time image For example, given the P48’s typical limiting magnitude of
subtractionpipeline(P.E.Nugentetal. 2015,inpreparation). R=20.6andthestandardcadenceofδt =0.5hr, ifaburstis
Thereal-timepipelinecreatesdifferenceimagesbetweenthe observed t =3hr after the trigger, its afterglow may be ex-
new P48 observations and co-added references composed of pected to have detectable photometric evolution only if it is
observations from months or years earlier. It generates can- brighterthanR=18.3. NotingthatlongGRBspreferentially
didatesbyperformingsourceextractiononthedifferenceim- occurathighredshiftsandinintrinsicallysmall,faintgalaxies
ages. Amachinelearningclassifierassignsareal/bogusscore (Svenssonetal.2010), weconsiderfaintsourcesthatdonot
(RB2; Brinketal.2013)toeachcandidatethatpredictshow displayevidenceoffadingiftheyarenotspatiallycoincident
likely the candidate is to be a genuine astrophysical source withanysourcesinSDSSorarchivaliPTFobservations.
(ratherthanaradiationhit, aghost, animperfectimagesub- If a faint source is near a spatially resolved galaxy, then
tractionresidual,oranyotherkindofartifact). we compute its distance modulus using the galaxy’s redshift
Table 1 lists the number of candidates that remain after orphotometricredshiftfromSDSS.WeknowthatlongGRB
eachstageofcandidateselection. First, requiringcandidates optical afterglows at t =1 day typically have absolute mag-
to have signal-to-noise ratio (S/N)>5 gives us a median of nitudes of −25 < M < −21mag (1σ range; see Figure 9
B
35,000 candidates. This number varies widely with galac- of Kann et al. 2011). Most SNe are significantly fainter:
tic latitude and the area searched (a median of 500 deg−2). Type Ia are typically MB −19mag whereas Ibc and II are
Second, we only select candidates that have R∼B2>0.1, re- MB −17mag, with lum∼inous varieties of both Type Ibc
∼
ducing the number of candidates to a median of 36% of the and II extending to MB −19mag (Richardson et al. 2002;
∼
originallist.40 Third,werejectcandidatesthatcoincidewith Li et al. 2011). Therefore, if the candidate’s presumed host
known stars in reference catalogs (Sloan Digital Sky Sur- galaxywouldgiveitanabsolutemagnitudeMR<−20mag,it
vey (SDSS) and the PTF reference catalog), cutting the list isconsideredpromising. Thiscriterionisonlyusefulforlong
to17%. Fourth,weeliminateasteroidscatalogedbytheMi- GRBs because short GRB afterglows are typically 6 mag
∼
norPlanetCenter,reducingthelistto16%. Fifth,wedemand fainterthanlongGRBafterglows(Kannetal.2011).
atleasttwosecureP48detectionsaftertheGBMtrigger, re- Thehumansavesallcandidatesthatareconsideredpromis-
ducingthelisttoafewpercent,or 500candidates. ingbythesemeasurestotheiPTFTransientMarshaldatabase.
Whentheimagesubtractionpipe∼linehasfinishedanalyzing This step baptizes them with an iPTF transient name, which
atleasttwosuccessiveepochsofanyonefield,theTOOMar- consistsofthelasttwodigitsoftheyearandasequentialal-
shal contacts the humans again and the surviving candidates phabeticdesignation.
arepresentedtothehumansviatheTreasuresportal.
2.5. ArchivalvettingintheTransientMarshal
2.4. VisualscanninginTreasuresPortal OncenamedintheTransientMarshal,weperformarchival
vetting of each candidate using databases including VizieR
The remaining candidate vetting steps currently involve (Ochsenbein et al. 2000), NED,42 the High Energy Astro-
human participation and are informed by the nature of the physics Science Archive Research Center (HEASARC),43
othertransientsthatiPTFcommonlydetects:foregroundSNe
andCatalinaReal-timeTransientSurvey(Drakeetal.2009),
(slowly varying and in low-z host galaxies), active galactic
inordertocheckforanypasthistoryofvariabilityatthatposi-
nuclei(AGNs),cataclysmicvariables,andM-dwarfflares.
tion(seeFigure13forascreenshotoftheTransientMarshal).
IntheTreasuresportal, wevisuallyscanthroughtheauto-
We check for associations with known quasars or AGNs
matically selected candidates one P48 field at a time, exam-
in Véron-Cetty & Véron (2010) or with AGN candidates in
ining 10objectsperfield(seeFigure12forascreenshotof
∼ Flesch(2010).
theTreasuresportal). Wevisuallyassesseachcandidate’sim-
M dwarfs can produce bright, blue, rapidly fading opti-
agesubtractionresidualcomparedtotheneighboringstarsof
cal flares than can mimic optical afterglows. To filter our
similarbrightnessinthenewimage. Iftheresidualresembles
M dwarfs, we check for quiescent infrared counterparts in
the new image’s point-spread function, then the candidate is
WISE (Cutri & et al. 2014). Stars of spectral type L9–M0
consideredlikelytobeagenuinetransientorvariablesource.
peak slightly blueward of the WISE bandpass, with typical
Next,welookatthephotometrichistoryofthecandidates.
colors(Wrightetal.2010)
Given the time, t, of the optical observation relative to the
burst and the cadence, δt, we expect that a typical optical 3(cid:46) [R−W1] (cid:46)12
afterglow that decays as a power law Fν t−α, with α=1, 0.1(cid:46)[W1−W2](cid:46)0.6
wouldfadebyδm=2.5log10(1+δt/t)mag∝overthecourseof 0.2(cid:46)[W2−W3](cid:46)1
ourobservations. Anysourcethatexhibitsstatisticallysignif-
icantfading(δm/m 1)consistentwithanafterglowdecay 0(cid:46)[W3−W4](cid:46)0.2.
(cid:29)
becomesaprimetarget.41 Therefore, a source that is detectable in WISE but that is ei-
Notethata1σdecayinbrightnessrequiressuchasourceto ther absent from or very faint in the iPTF reference images
be suggestsaquiescentdwarfstar.
(cid:18) (cid:19)
δt
−2.5log10 t√2 (2) 2.6. Photometric,Spectroscopic,andBroad-band
Follow-up
40ThisRB2thresholdissomewhatdeeperthanthatwhichisusedinthe
The above stages usually result in 10 promising optical
iPTFsurvey. Animprovedclassifier,RB4(Bueetal.2014),enteredevalua- ∼
transient candidates that merit further follow-up. If, by this
tionin2014AugustshortlybeforeGRB140808A.
41 A source that exhibits a statistically significant rise is generally also
followedup,butaspartofthemainiPTFtransientsurvey,ratherthanasa 42http://ned.ipac.caltech.edu
potentialopticalafterglow. 43http://heasarc.gsfc.nasa.gov
TheNeedleinthe100deg2Haystack 5
Table1
NumberofOpticalTransientCandidatesSurvivingEachVettingStage
S/N RB2 Not Notin Detected Savedfor
GRB >5 >0.1 Stellar MPCa Twice Follow-up RB2b
130702A 14629 2388 1346 1323 417 11 0.843
131011A 21308 8652 4344 4197 434 23 0.198
131231A 9843 2503 1776 1543 1265 10 0.137
140508A 48747 22673 9970 9969 619 42 0.730
140606B 68628 26070 11063 11063 1449 28 0.804
140620A 152224 50930 17872 17872 1904 34 0.826
140623A 71219 29434 26279 26279 442 23 0.873
140808A 19853 4804 2349 2349 79 12 0.318
Medianreduction 36% 17% 16% 1.7% 0.068%
aNotinMinorPlanetCenterdatabase
bRB2scoreofopticalafterglowinearliestP48detection
point, data from Fermi LAT or from IPN satellites are avail- For some bursts (GRB 140606B), we also obtain pho-
able, we can use the improved localization to select an even tometry with the Large Monolithic Imager (LMI) mounted
smallernumberoffollow-uptargets. on the 4.3m Discovery Channel Telescope (DCT) in Happy
For sources whose photometric evolution is not clear, we Jack, Arizona. Standard CCD reduction techniques (e.g.,
performphotometricfollow-up. Wemayscheduleadditional bias subtraction, flat-fielding) are applied using a custom
observations of some of the P48 fields if a significant num- IRAF pipeline. Individual exposures are aligned with re-
ber of candidates are in the same field. We may also use spect to astrometry from the Two Micron All Sky Survey
the P48 to gather more photometry for sources that are su- (2MASS;Skrutskieetal.2006)usingSCAMP(Bertin2006)
perimposedonaquiescentsourceorgalaxy,inordertomake andstackedwithSWarp(Bertinetal.2002).
useoftheimagesubtractionpipelinetoautomaticallyobtain Where GROND (Greiner et al. 2008) and RATIR (Butler
host-subtracted magnitudes. For isolated sources, we sched- etal.2012)havereportedmulticolorphotometryinGCNcir-
uleoneormoreepochsofr-bandphotometrywiththeP60.If, culars, we include their published data in Table 5 and our
bythispoint,anycandidatesshowstrongevidenceoffading, light-curveplots.
webeginmulticolorphotometricmonitoringwiththeP60. We monitor GBM–iPTF afterglows with CARMA, a mil-
Next, we acquire spectra for one to three candidates per limeter-wave interferometer located at Cedar Flat near Big
burstusingtheP200,Gemini,Keck,Magellan,orHimalayan Pine, California. All observations are conducted at 93 GHz
Chandra Telescope (HCT). A spectrum that has a relatively insingle-polarizationmodeinthearray’sC,D,orEconfigu-
featureless continuum and high-redshift absorption lines se- ration. Targetsaretypicallyobservedoncefor1–3hrwithina
cures the classification of the candidate as an optical after- fewdaysaftertheGRB,establishingthephasecalibrationus-
glow. ingperiodicobservationsofanearbyphasecalibratorandthe
Once any single candidate becomes strongly favored over bandpass and the flux calibration by observations of a stan-
the others based on photometry or spectroscopy, we trigger dardsourceatthestartofthetrack.Ifdetected,weacquiread-
X-rayandUVobservationswithSwiftandradioobservations ditionalobservationsinapproximatelylogarithmicallyspaced
withVLA,CARMA,andAMI. DetectionofaradioorX-ray time intervals until the afterglow flux falls below detection
afterglowtypicallyconfirmsthenatureoftheopticaltransient, limits. All observations are reduced using MIRIAD using
evenwithoutspectroscopy. standardinterferometricflaggingandcleaningprocedures.
Finally, we promptly release our candidates, upper limits, We look for radio afterglows at 6.1 and/or 22 GHz with
and/orconfirmedafterglowdiscoveryinGCNcirculars. VLA. VLAobservationsarereducedusingtheCommonAs-
tronomy Software Applications (CASA) package. The cali-
brationisperformedusingtheVLAcalibrationpipeline. Af-
2.7. Long-termMonitoringandDataReduction
ter running the pipeline, we inspect the data (calibrators and
The reported P48 magnitudes are all in the Mould R band target source) and apply further flagging when needed. The
and in the AB system (Oke & Gunn 1983), calibrated with VLA measurement errors are a combination of the rms map
respecttoeitherr(cid:48)pointsourcesfromSDSSorfornon-SDSS error, which measures the contribution of small unresolved
fieldsusingthemethodsdescribedinOfeketal.(2012). fluctuationsinthebackgroundemissionandrandommapfluc-
Tomonitortheopticalevolutionofafterglowsidentifiedby tuations due to receiver noise, and a basic fractional error
our program, we typically request nightly observations in ri (here estimated to be 5%) which accounts for inaccura-
(and occasionally gz) filters for as long as the afterglow re- cies of the flux density≈calibration. These errors are added
mained detectable. Bias subtraction, flat-fielding, and other inquadrature,andtotalerrorsarereportedinTable6.
basic reductions are performed automatically at Palomar by Starting in 2014 August, we also look for radio emission
the P60 automated pipeline using standard techniques. Im- withAMI.AMIiscomposedofeight12.8mdishesoperating
ages are then downloaded and stacked as necessary to im- inthe13.9–17.5GHzrange(centralfrequencyof15.7GHz)
prove the S/N. Photometry of the optical afterglow is then whenusingfrequencychannels3–7(channels1,2,and8are
performed in IDL using a custom aperture-photometry rou- disregarded due to their currently susceptibility to radio in-
tine, calibrated relative to SDSS secondary standards in the terference). Forfurther detailson the reductionand analysis
field(whenavailable)orusingourownsolutionforsecondary performedontheAMIobservationspleaseseeAndersonetal.
field standards constructed during a photometric night (for (2014b).
fieldsoutsidetheSDSSfootprint).
6 Singeretal.
3. THEGBM–iPTFBURSTS P60 imaging and P200 spectroscopy for three significantly
varying sources. Of the three, iPTF13bxl showed the clear-
Todate,wehavesuccessfullyfollowedup35FermiGBM
est evidence of fading in the P48 images. Its spectrum at
bursts and detected eight optical afterglows. The detections
∆t =1.2daysconsistedofafeaturelessbluecontinuum. We
arelistedinTable2,andalloftheP48tilingsarelistedinTa-
triggeredSwift,whichfoundabrightX-raysourceattheposi-
ble3. Figure5showstheGBMlocalizationsandP48tilings
tionofiPTF13bxl(Singeretal.2013c;D’Avanzoetal.2013).
forthedetectedbursts.InFigure1,thelightcurvesareshown
ShortlyafterweissuedourGCNcircular(Singeretal.2013c),
in the context of a comprehensive sample of long GRB af-
Cheungetal.(2013)announcedthatthebursthadenteredthe
terglowscompiledbyD.A.Kann(2015,privatecommunica-
FOVofLATat∆t=250s. TheLATerrorcirclehadaradius
tion).
of0.◦5,anditscenterwas0.◦8fromiPTF13bxl. AnIPNtrian-
Theoutcomeofanindividualafterglowsearchislargelyde-
gulationwithMESSENGER(GRNS),INTEGRAL(SPI-ACS),
termined by two factors: how much probability is contained
Fermi-GBM, andKonus-Wind (Hurleyetal.2013)yieldeda
within the P48 footprints, and how bright the afterglow is at
0.◦46-wideannulusthatwasalsoconsistentwiththeOT.
thetimeoftheobservations(seeFigure2). Wecalculatethe
The afterglow’s position is 0.(cid:48)(cid:48)6 from an R = 23.01mag
expectedsuccessrateasfollows. Foreachburst,wefindthe
sourcethatisjustbarelydiscernibleintheP48referenceim-
priorprobabilitythatthepositioniscontainedwithintheP48
ages.AspectrumfromNOT+ALFOSC(Leloudasetal.2013)
fieldsthatweobserved. Wethencomputethefractionofaf-
determined a redshift of z = 0.145 for a galaxy 7.(cid:48)(cid:48)6 to the
terglowsfromKann’ssample(whichhasameanandstandard
southofiPTF13bxl. At∆t=2.0days,weobtainedaMagel-
deviation of 22 2 mag at t =1 day) that are brighter than
R=20.6mag at±the same age as when the P48 observations lan+IMACSspectrum(Mulchaeyetal.2013)andfoundweak
emission lines at the location of the afterglow that we inter-
started. Theproductofthesetwonumbersisthepriorproba-
preted as Hα and [OIII] at the same redshift. Kelly et al.
bilityofdetectionforthatburst. Bysummingoverallofthe
(2013)characterizedtheburst’shostenvironmentindetailand
iPTF/GBM bursts, we obtain the expected number of detec-
concludedthatitexplodedinadwarfsatellitegalaxy.
tions. Within95%confidencebootstraperrorbars,wefindan
Joining the two P48 observations at ∆t < 1 day to the
expected 5.5–8.5 detections, or a success rate of 16%–24%.
late-time P60 light curve requires a break at ∆t = 1.17
Thisisconsistentwiththeactualsuccessrateof23%. 0.09 days, with slopes α =0.57 0.03 and α =1.05±
This suggests that the success rate is currently limited by 0.03 before and after thOe,1break, ±respectively.O,2The XR±T
the survey area and the response time (dictated by sky posi-
light-curvebeginsjustpriortothisapparentbreakandseems
tionandweather). Wecouldincreasethesuccessratebyde-
to follow the late-time optical decay (until the SN begins to
creasing the maximum time since trigger at which we begin
dominateat∆t =5days),althoughtheautomatedSwiftlight
follow-up. We could increase the success rate without ad-
curve analysis (Evans et al. 2009) also suggests a possible
verselyaffectingthenumberofdetectionsbysimplysearch-
X-raybreakwithaboutthesametimeandslopes. Thishints
ingagreaterareaforcoarselylocalizedevents.
at an achromatic break, normally a signature of a jet. How-
Fig. Set3. LightcurvesandSEDs
ever,thelateslopeandthechangeinslopearebothunusually
Overthenextfewsections,wesummarizetheobservations
shallow for a jet break. Furthermore, the radio light curve
and general physical interpretation of all of the GBM–iPTF
does not exhibit a break. The change in slope is also a little
afterglows detected to date. Figure 3 shows the light curves
toolargeforcoolingfrequencycrossingtheband(forwhich
andSEDsspanningX-ray,UV,optical,IR,andradiofrequen-
onewouldexpect∆α=1/4). Anenergyinjectionorastruc-
cies.Table4containsalogofourspectroscopicobservations.
turedjetmodelmayprovideabetterfit(Panaitescu2005).
Table 5 lists a selection of ultraviolet, optical, and infrared
Late-time∆t>1dayobservationsincludeseveralP60gri
observations, including all of our P48 and P60 observations.
observations, three RATIR r(cid:48)i(cid:48)ZYJH epochs, an extensive
Table6listsallofourradiodetections.
SwiftXRTandUVOTlightcurve,andradioobservationswith
VLAandCARMA(althoughoftheVLAdata,weonlyhave
3.1. GRB130702A/iPTF13bxl
access to the first observation). The optical and X-ray spec-
ThisisthefirstGBMburstwhoseafterglowwediscovered tralslopesaresimilar,β =0.7 0.1andβ =0.8 0.1. An
O X
with iPTF (Singer et al. 2013b), indeed the first afterglow SED at 2<∆t <2.3 days is w±ell explained by the±standard
ever to be pinpointed based solely on a Fermi GBM local- external shock model (Sari et al. 1998) in the slow cooling
ization. It is also the lowest-redshift GRB in our sample, so regime, with ν lying between the VLA and CARMA fre-
m
ithastherichestandmostdenselysampledbroadbandafter- quencies and ν in the optical. This fit requires a relatively
c
glowdata. Ithastwoothermajordistinctions: itsassociated flat electron spectrum, dn /dγ γ −p with p 1.6, cut off
e e e
SN(SN2013dx,Schulzeetal.2013;Pozanenkoetal.2013; athighenergies. Applyingther∝elevantclosure≈relations(for
Cenkoetal.2013a;D’Eliaetal.2013)wasdetectedspectro- thecaseof1< p<2,seeDai&Cheng2001)toα andβ
X X
scopically,anditspromptenergeticsareintermediatebetween permitseitheranISMorwindenvironment.
low-luminosity GRBs (llGRBs) and standard cosmic bursts Our late-time spectroscopy and analysis of the SN will be
(seebelow). publishedseparately(S.B.Cenkoetal.2015,inpreparation).
Based on the Fermi GBM ground localization with an er-
ror radius of 4◦, we imaged 10 fields twice with the P48
3.2. GRB131011A/iPTF13dsw
at ∆t = t−t = 4.2 hr after the burst.44 We scheduled
GBM
44 Atthetime,ourtilingalgorithmselectedfieldsbasedonanempirical questionandthenconstructeda2Dangularprobabilitydistributionfromit.
calibrationofFermiGBM’ssystematicerrors. Wehadselectedburststhat Forsufficientlylargeerrorradii,thisprescriptionproducedprobabilitydis-
weredetectedbybothSwiftandFermiandconstructedafittoacumulative tributionsthathadaholeinthemiddle. Forthisreason,thetilingalgorithm
histogramofthenumberofburstswhoseBATorXRTpositionswerewithin pickedoutP48fieldsthatformedanannulusaroundtheGBM1σerrorcir-
agivennumberofnominal1σstatisticalradiiofthecenteroftheFermierror cle(not,aswestatedinSingeretal.2013b,becauseofalackofreference
circle. Ourtilingalgorithmscaledthisfitbythe1σ radiusoftheburstin images).
TheNeedleinthe100deg2Haystack 7
5
P LongGRBs
4
8
5 Fastopticaltransients
m
in iPTF/GBMdetections
r
e
s
p
o
n
s
10 e
tim
e
15
)
R
(
g
a
m
20
25
30
10−4 10−3 10−2 10−1 100 101 102 103
t(days,observerframe)
Figure1. OpticallightcurvesofFermi–iPTFafterglowstodate.ThelightcurvesoftheeightiPTF/GBMburstsareshowninred.Forcomparison,thegraylines
showacomprehensivesampleoflongGRBopticallightcurvesfromCenkoetal.(2009),Kannetal.(2010),Perleyetal.(2014d),andD.A.Kann(2015,private
communication). Thewhiteareaoutsideofthelight-grayshadingillustratestherangeofGRBafterglowsthatareaccessiblegivenahalf-hourcadenceandthe
P48’s60slimitingmagnitudeofR=20.6.ThetwolightcurvesshowninblueareotherrelatediPTFtransients.ThefirstisPTF11agg,anafterglow-liketransient
withnodetectedgamma-rayemission(Cenkoetal.2013b).ThesecondisGRB140226A/iPTF14yb,reportedinitiallybyiPTFfromitsopticalafterglow(Cenko
etal.2014;Cenkoetal.,inpreparation),andlaterbyIPNfromitsgamma-rayemission(Hurleyetal.2014b).
We started P48 observations of Fermi trigger 403206457 At 15.1hr after the burst, we obtained a spectrum
(Jenke 2013) about 11.6hr after the burst. The optical tran- of iPTF13dsw with the Gemini Multi-object Spectrograph
sient iPTF13dsw (Kasliwal et al. 2013) faded from R = (GMOS) on the Gemini–South telescope. GMOS was con-
19.7mag to R = 20.2mag from 11.6 to 14.3hr. The latest figured with the R400 grating with a central wavelength of
pre-triggerimageon2013September25hadnosourceatthis 7200Å and the 1(cid:48)(cid:48)slit, providing coverage over the wave-
locationtoalimitofR>20.6mag. Theopticaltransientcon-
lengthrangeof5100–9300Åwitharesolutionof 3Å.No
tinued to fade as it was monitored by several facilities (Xu ≈
prominent features were detected over this bandpass, while
etal.2013a,c;Perleyetal.2013;Sudilovskyetal.2013;Vol-
thespectrumhadatypicalSNRof 3–4per1.4Åpixel.Rau
novaetal.2013). ≈
8 Singeretal.
Table2
GBM–iPTFDetections
R.A. Decl. Gal. Epeak Eγ,iso
GRB OT (J2000) (J2000) Lat.a z (keV,rest) (1052erg,rest)b,c T90(s) mR(tP48)d
GRB130702A iPTF13bxl 14h29m15s +15◦46(cid:48)26(cid:48)(cid:48) 65◦ 0.145 18± 3 <0.065± 0.001 58.9± 6.2 17.38
GRB131011A iPTF13dsw 02h10m06s -4◦24(cid:48)40(cid:48)(cid:48) -61◦ 1.874 625± 92 14.606± 1.256 77.1± 3 19.83
GRB131231A iPTF13ekl 00h42m22s -1◦39(cid:48)11(cid:48)(cid:48) -64◦ 0.6419 291± 6 23.015± 0.278 31.2± 0.6 15.85
GRB140508A iPTF14aue 17h01m52s +46◦46(cid:48)50(cid:48)(cid:48) 38◦ 1.03 534± 28 24.529± 0.86 44.3± 0.2 17.89
GRB140606B iPTF14bfu 21h52m30s +32◦00(cid:48)51(cid:48)(cid:48) -17◦ 0.384 801± 182 0.468± 0.04 22.8± 2.1 19.89
GRB140620A iPTF14cva 18h47m29s +49◦43(cid:48)52(cid:48)(cid:48) 21◦ 2.04 387± 34 7.28 ± 0.372 45.8± 12.1 17.60
GRB140623A iPTF14cyb 15h01m53s +81◦11(cid:48)29(cid:48)(cid:48) 34◦ 1.92 834± 317 3.58 ± 0.398 114.7± 9.2 18.04
GRB140808A iPTF14eag 14h44m53s +49◦12(cid:48)51(cid:48)(cid:48) 59◦ 3.29 503± 35 8.714± 0.596 4.5± 0.4 19.01
aGalacticlatitudeofopticalafterglow.ThisisoneofthemainfactorsthatinfluencesthenumberofopticaltransientcandidatesinTable1.
bEγ,isoisgivenfora1keV–10MeVrest-framebandpass.
c Therest-framespectralproperties,Epeak andEγ,iso,forGRB130702AarereproducedfromAmatietal.(2013). Forallotherbursts,wecalculatedthese
quantitiesfromthespectralfits(thescatfiles)intheFermiGBMcatalog(Goldsteinetal.2012)usingthek-correctionproceduredescribedbyBloometal.
(2001).
dR-bandapparentmagnitudeininitialP48detection.
1.0
Table3
LogofP48TilingsforFermiGBMBursts
0.8
GRBTimea FlGueBnMceb −ttbPu4r8stc APr4e8ad Prob.e bility
a
2013Jun2820:37:57 10 ±0.1 10.02 73 32% ob 0.6
→2013Jul0200:05:20 57 ±1.2 4.20 74 38% pr
2013Aug2807:19:56 372 ±0.6 20.28 74 64% nt
2013Sep2406:06:45 37 ±0.6 23.24 74 28% me 0.4
2013Oct0620:09:48 18 ±0.6 15.26 74 18% n
→22001133ONcotv10181070::4374::3309 8298 ±±00..65 114..5669 7733 5347%% ontai
2013Nov1008:56:58 33 ±0.3 17.47 73 44% c 0.2
2013Nov2516:32:47 5.5±0.3 11.72 95 26%
2013Nov2603:54:06 17 ±0.3 6.94 109 59%
2013Nov2714:12:14 385 ±1.4 13.46 60 50% 0.0
2013Dec3019:24:06 41 ±0.4 7.22 80 38% 0.0 0.2 0.4 0.6 0.8 1.0
→2013Dec3104:45:12 1519 ±1.2 1.37 30 32%
2014Jan0417:32:00 333 ±0.6 18.57 15 11% age at first obervation (days)
2014Jan0501:32:57 6.4±0.1 7.63 74 22%
2014Jan2214:19:44 9.1±0.5 11.97 75 34% Figure2. Priorprobabilityofcontainingtheburst’slocationwithintheP48
2014Feb1102:10:41 7.4±0.3 1.77 44 19% fieldsvs. ageoftheburstatthebeginningofP48observations. Afterglow
2014Feb1919:46:32 28 ±0.5 7.01 71 14% detectionsareshowninorange,andnon-detectionsareshowningray.
2014Feb2418:55:20 24 ±0.6 7.90 72 30%
2014Mar1114:49:13 40 ±1.2 12.18 73 54%
etal.(2013)observedtheopticaltransientwiththeX-Shooter
2014Mar1923:08:30 71 ±0.3 3.88 74 48%
2014Apr0404:06:48 82 ±0.2 0.11 109 69% instrument on the ESO 8.2-m VLT. In their spectrum ex-
2014Apr2923:24:42 6.2±0.2 10.99 74 15% tending from 3000 to 24000Å, they identified several
→2014May0803:03:55 614 ±1.2 6.68 73 67% weak absorptio∼n lines fro∼m which they derived a redshift of
2014May1719:31:18 45 ±0.4 8.60 95 69%
2014May1901:01:45 39 ±0.5 4.42 73 41% z=1.874. BothspectraareshowninFigure4.
→2014Jun0603:11:52 76 ±0.4 4.08 74 56% The source was detected by Swift XRT (Page 2013), but
2014Jun0817:07:11 19 ±0.6 11.20 73 49% with insufficient photons for spectral analysis. The source
→2014Jun2005:15:28 61 ±0.6 0.17 147 59%
wasobservedwithATCA,butnoradioemissionwasdetected.
→2014Jun2305:22:07 61 ±0.6 0.18 74 4%
2014Jun2816:53:19 18 ±1.0 16.16 76 20% Largely because in our sample this is the oldest afterglow at
2014Jul1607:20:13 2.4±0.3 0.17 74 28% thetimeofdiscovery,therearenotenoughbroadbanddatato
2014Jul2900:36:54 81 ±0.7 3.43 73 65% constraintheblastwavephysics.
2014Aug0711:59:33 13 ±0.1 15.88 73 54%
→2014Aug0800:54:01 32 ±0.3 3.25 95 69%
3.3. GRB131231A/iPTF13ekl
a TimeofFermiGBMtrigger. →Afterglowdetectionsaremarkedwithan
arrow. ThecorrespondingentriesinTable2canbefoundbymatchingthe GRB131231A was detected by Fermi LAT (Sonbas et al.
datetotheGRBname(GRBYYMMDDA). 2013)andGBM(Jenke&Xiong2014), withphotonsofen-
b Observed Fermi GBM fluence in the 10–1000 keV band, in units
ergies up to 9.7 GeV. Xu et al. (2013b) observed the LAT
of10−7ergcm−2.ThisquantityistakenfromthebcatfilesfromtheFermi
error circle with the 1-m telescope at Mt. Nanshan, Xin-
GRBcatalogatHEASARC.
cAgeinhoursoftheburstatthebeginningoftheP48observations. jiang, China. At 7.9hr after the burst, they detected a single
dAreaindeg2spannedbytheP48fields. R= 17.6magsourcethatwasnotpresentinSDSSimages.
eProbability,giventheFermiGBMlocalization,thatthesourceiscontained At1∼7.3hraftertheburst,Malesanietal.(2013)observedthe
withintheP48fields.
afterglowcandidatewiththeMOSaicCAmera(MOSCA)on
the2.56-mNordicOpticalTelescope(NOT).Thesourcehad
fadedtoR=18.6.
TheNeedleinthe100deg2Haystack 9
A A A AA AA AA A
M M M MM MM MM M
RAA R R RR RR RR R
GRB 130702A / ALL A A AA AA AA A
CVV C C CC CC CC C
iPTF13bxl
12
14 100
16
λ)r 10-1
/
λ
5log(10 18 10-2 mJy)
−
B) 20 f (ν
m(A 10-3 dio,
cal/NIR, 2242 10-4 X-ray/ra
pti
o
V/
U 26
10-5
28
10-6
30
10-1 100 101 102 103
t t (days)
− burst
101
100
10-1
y)
mJ 10-2
f (ν
10-3
1.26d
10-4 2.14d
4.16d
10-5
100 101 102 103 104 105 106 107 108 109 1010
ν (GHz)
Figure3. LightcurvesandSEDsofGBM–iPTFafterglows;GRB130702A/iPTF13bxlisshownhereintheprintversion.
(Anextendedversionofthisfiguresetisavailableintheonlinejournal.)
10 Singeretal.
Table4
LogofSpectroscopicObservations
Date Telescope Instrument Wavelengths(Å) Lines References
GRB131011A/iPTF13dsw
2013Oct1208:56 GeminiSouth GMOS 5100–9300 none Kasliwaletal.(2013)
2013Oct1303:59 ESO/VLTUT3 X-shooter 3100–5560 Lyα,SiII,CII,CIV,AlII Rauetal.(2013)
··· ··· ··· 5550–10050 FeII,MgII ···
GRB131231A/iPTF13ekl
2014Jan0102:15 GeminiSouth GMOS 6000–10000 [OII],[OIII],CaIIH+K Cucchiara(2014)
GRB140508A/iPTF14aue
2014May0818:55 HCT HFOSC 3800–8400 FeII,MgII Bhalerao&Sahu(2014)
2014May0906:33 APO DIS 3200–9800 none none
GRB140606B/iPTF14bfu
2014Jun0719:16 KeckII DEIMOS 4500–9600 [OII],[OIII],Hα,CaIIH+K Perleyetal.(2014a)
GRB140620A/iPTF14cva
2014Jun2014:00 GeminiNorth GMOS 5090–9300 MgI,MgII,FeII,AlII,SiII,SiII∗ Kasliwaletal.(2014)
··· ··· ··· 4000–6600 ··· ···
GRB140623A/iPTF14cyb
2014Jun2308:10 GeminiNorth GMOS 4000–6600 MgII,FeII,AlII,SiII,AlIII,CI,CIV Bhaleraoetal.(2014)
GRB140808A/iPTF14eag
2014Aug0821:43 GTC OSIRIS 3630–7500 DLA,SII,SiII,OI,CII,SiIV,FeII,AlII,CIV Gorosabeletal.(2014a)
[O αH
]III
CC
aa
⊕ KIIHII
L SC C A FF FF MM SS SSAMM F MMM
αy iIIII IV lII eII eII eIIeII gIIgII iII∗iII iII∗iIIlIIgIIgII eII gIIgIIgI
⊕ ⊕ ⊕
[O]II CaKIICaHII [O]III CI SiII AlII FeII AlIIIMgIIMgII FeII MgIIMgIIMgI
⊕ ⊕ ⊕ ⊕
F M L SLSOCS SCFA
eII gII βy iII αyIIIII iIV iII IV eII lII
⊕ ⊕
4000 5000 6000 7000 8000 9000 4000 5000 6000 7000 8000 9000
observedwavelength(A˚) observedwavelength(A˚)
Figure4. Afterglowspectra.Thehorizontalaxisshowswavelengthinvacuumintheobserverframe,andtheverticalaxisshowsscaledflux.Linesattheredshift
oftheputativehostarelabeledinblack;linescorrespondingtoanyinterveningabsorbingsystemsarelabeledinred.Notethatincaseswhereoneorfewerlines
arediscernibleinourspectra,theredshiftshavebeenreportedinGCNsbyothergroups.