The chemical enrichment of long-GRB nurseries up to z=2

0.27 MB·

by  S. D. Vergani

Astronomy&Astrophysicsmanuscriptno.VerganiHGz2LE1c (cid:13)cESO2017
January11,2017
=
The chemical enrichment of long-GRB nurseries up to z 2
S.D.Vergani1,2,3,J.Palmerio2,R.Salvaterra4,J.Japelj5,F.Mannucci6,D.A.Perley7,P.D’Avanzo3,T.Krühler8,M.
Puech1,S.Boissier9,S.Campana3,S.Covino3,L.K.Hunt6,P.Petitjean2,andG.Tagliaferri3
1 GEPI, Observatoire de Paris, PSL Research University, CNRS, Univ. Paris Diderot, Sorbonne Paris Cité, Place Jules Janssen,
F-92195Meudon,France;e-mail:[email protected]
2 Institutd’AstrophysiquedeParis,UniversitéParis6-CNRS,UMR7095,98bisBoulevardArago,F-75014Paris,France
3 INAF-OsservatorioAstronomicodiBrera,viaE.Bianchi46,I-23807Merate,Italy
4 INAF-IASFMilano,viaE.Bassini15,I-20133,Milano,Italy
5 INAF-OsservatorioAstronomicodiTrieste,viaG.B.Tiepolo11,I-34131Trieste,Italy
6 INAF-OsservatorioAstrofisicodiArcetri,LargoE.Fermi5,I-50125Firenze,Italy
7 7 DarkCosmologyCentre,NielsBohrInstitute,UniversityofCopenhagen,JulianeMariesVej30,DK-2100Copenhagen,Denmark
1 8 Max-Planck-InstitutfürextraterrestrischePhysik,Giessenbachstraße,D-85748Garching,Germany
0 9 AixMarseilleUniv,CNRS,Laboratoired’AstrophysiquedeMarseille,F-13388,Marseille,France
2
Received;accepted

n
a
J ABSTRACT

9
Aims.Weinvestigatetheexistenceofametallicitythresholdfortheproductionoflonggamma-raybursts(LGRBs).
Methods.WeusedthehostgalaxiesoftheSwift/BAT6sampleofLGRBs.Weconsideredthestellarmass,starformationrate(SFR),
E] andmetallicitydeterminedfromthehostgalaxyphotometryandspectroscopyuptoz=2andusedthemtocomparethedistribution
ofhostgalaxiestothatoffieldgalaxiesinthemass-metallicityandfundamentalmetallicityrelationplane.
H
Results. WefindthatalthoughLGRBsalsoformingalaxieswithrelativelylargestellarmasses,thelargemajorityofhostgalax-
. ies have metallicities below log(O/H) ∼ 8.6. The extension to z = 2 results in a good sampling of stellar masses also above
h
Log(M /M )∼9.5 and provides evidence that LGRB host galaxies do not follow the fundamental metallicity relation. As shown
p ∗ (cid:12)
bythecomparisonwithdedicatednumericalsimulationsofLGRBhostgalaxypopulation,theseresultsarenaturallyexplainedby
-
o theexistenceofamild(∼0.7Z(cid:12))thresholdfortheLGRBformation.Thepresentstatisticsdoesnotallowustodiscriminatebetween
r differentshapesofthemetallicitycutoff,buttherelativelyhighmetallicitythresholdfoundinthisworkissomewhatindisagreement
t tomostofthestandardsingle-starmodelsforLGRBprogenitors.
s
a
Keywords. Gamma-rayburst:general–Galaxies:abundances–Galaxies:starformation
[

1
v
1. Introduction PortegiesZwart2013;Kellyetal.2014;Perleyetal.2016b),it
2
hasbeenoneofthemoststudiedinthepastasthemetalcontent
1
Ithasbeenestablishedthatlonggamma-raybursts(LGRBs)are of the progenitor star is considered to play a major role in the
3
2linkedtotheexplosionsofmassivestars,bothfromthestudiesof formation of a LGRB explosion. Single-star evolution models
0theirhostgalaxyformationsites(Fruchteretal.2006;Svensson predictthatthemetallicityofLGRBprogenitorsshouldbevery
.et al. 2010) as well as from detections of accompanying super- low(e.g.,Hirschietal.2005;Yoon&Langer2005;Woosley&
1
novaemission(GRB-SN;seeCanoetal.2016forareview).Itis Heger2006):inthiswaytheprogenitorstarcanexpeltheouter
0
stillnotclearwhichconditionsgiverisetoLGRBsorwhatisthe envelope (hydrogen and helium are not observed spectroscopi-
7
1relation between the progenitors of LGRBs and those of other cally)withoutremovingtoomuchangularmomentumfromthe
:explosionsresultingfromdeathsofmassivestars(e.g.,Metzger rapidly rotating core. Higher metallicity values are allowed in
vetal.2015). the case of the models presented by Georgy et al. (2012), also
Xi The progenitors of nearby core-collapse supernovae can be dependingonthedifferentprescriptionsbetweenthecouplingof
directly identified as resolved stars in archived high-resolution surfaceandcoreangularmomentuminthestar.Alternatively,the
r
aimages of their birth places (Smartt 2015). However, LGRBs LGRBprogenitorscouldbecloseinteractingbinaries,inwhich
have a lower occurrence rate (e.g., Berger et al. 2003; Guetta casethemetallicityisalessconstrainingfactor(e.g.,Fryeretal.
&DellaValle2007)andareusuallyobservableatcosmological 2007;vandenHeuvel&Yoon2007).Strongobservationalcon-
distances, for which their birth places cannot be resolved. Our straintsareclearlyneededtounderstandwhichoftheevolution-
understanding ofLGRB progenitors thereforedepends onlink- arychannelscouldproduceaLGRB.
ingthepredictionsofdifferentstellarevolutionmodelswiththe DifferentobservationalworksonLGRBhostgalaxiesinthe
observed properties of LGRB multiwavelength emission (e.g., literaturehaveindeedrevealedthattheirmetallicitiesaremostly
Schulzeetal.2011;Canoetal.2016)andtheirhostgalaxyenvi- subsolar (Modjaz et al. 2008; Levesque et al. 2010a; Graham
ronment(seePerleyetal.2016aforareview).Inthiswork,we & Fruchter 2013; Vergani et al. 2015; Krühler et al. 2015; Per-
focusonthelatter. ley et al. 2016b; Japelj et al. 2016). The evidence is corrobo-
While metallicity is not the only factor that might affect rated by numerical simulations (e.g., Nuza et al. 2007; Camp-
theefficiencyoftheLGRBproduction(e.g.,vandenHeuvel& isi et al. 2011; Trenti et al. 2015). In particular, Campisi et al.
Articlenumber,page1of4
A&Aproofs:manuscriptno.VerganiHGz2LE1c
(2011) studied LGRB host galaxies in the context of the mass Table1.Swift/BAT6sampleofLGRBhostgalaxiesat1 < z < 2with
metallicitydetermination,visiblefromthesouthernhemisphere.
metallicity(e.g.Tremontietal.2004)andfundamentalmetallic-
ity (Mannucci et al. 2010, 2011) relations of field star-forming
galaxiesbycombiningahigh-resolutionN-bodysimulationwith Hostgalaxy redshift Log(M(cid:63)/M(cid:12)) SFR Metallicity
[M(cid:12)yr−1] 12+log(O/H)
asemi-analyticmodelofgalaxyformation.Campisietal.(2011) GRB080413B 1.1012 9.3 2.1+3.1 8.4+0.2
findthataverylowmetallicitycutisnotnecessarytoreproduce GRB090926B 1.2427 10.28 26+−119.2 8.44−+00..218
the observed relations. However, previous observational works GGRRBB006611102017∗∗ 11..32166203 190..2321 454.8.2−+−+14411..889 88.1.56+−+−00.000...2911083
present one or more of the following issues: (i) they are based GRB071117∗ 1.3293 <10.12 >2−.810 8.4−+00..0165
onincompletebiasedsamples(e.g.,Levesqueetal.2010a);(ii) GRB100615A 1.3979 9.27 8.6+13.9 8.14−+00.0.296
theyarebasedonstellarmassesdirectlydeterminedfromobser- GGRRBB007600330066 11..45956957 1100.5.53 1170.61−++−48.214348.6 89..4152−+−+0000....20012888
vations, but on metallicities inferred from the mass-metallicity GRB080605 1.6408 10.53 47.0−+1117 8.46−+00..4028
relation(e.g.,Perleyetal.2016b);(iii)theyusemetallicitiesdi- GRB080602 1.8204 9.99 125.0−+11245 8.56−+00.0.28
rectlydeterminedfromtheobservations,butdonotconsiderthe GRB060814 1.9223 10.82 54.0+−8695 8.38+−00..134
−19 −0.28
stellarmasses(e.g.:Krühleretal.2015);and(iv)theyarebased
Notes.There are 4 LGRBs in the 1 < z < 2 sample for which we
on samples limited to small redshift ranges (e.g.,0 < z < 1) as
couldnotdeterminethemetallicityoftheirhostgalaxies:GRB050318,
inJapeljetal.(2016).
GRB050802, GRB060908, and GRB091208B. Indeed, there are no
In this paper we study the metallicity of the host galaxies usefulspectratothispurposeforthehostgalaxiesofGRB091208Band
of the complete Swift/BAT6 sample (Salvaterra et al. 2012) of GRB050318. For the host galaxies of GRB050802 and GRB060908
LGRBs at z < 2, visible from the southern hemisphere. Com- we obtained X-shooter spectroscopy (Prog. ID: 097.D-0672; PI: S.D.
bining the observed properties with simulations, we study their Vergani),butthespectradonotshowsufficientemissionlinestoallow
behaviorinthestellarmass-metallicityrelation(MZ)andfun- themetallicitydetermination.
damentalmetallicityrelation(FMR).Afterthedescriptionofthe ∗:fromnew/unpublishedX-shooterobservationspresentedinthispaper
(seeTable3).
sampleandnewdata(Section2),wepresenttheresultsinSec-
tion3anddiscusstheminSection4.
Allerrorsarereportedat1σconfidenceunlessstatedother-
wise.Weuseastandardcosmology(PlanckCollaborationetal. exceptionofthehostgalaxiesofGRB071117andGRB080602,
2014):Ωm = 0.315,ΩΛ = 0.685,andH0 = 67.3kms−1 Mpc−1. which are not part of the Perley et al. (2016b) sample, and for
Thestellarmassesandstarformationrates(SFR)aredetermined which we determined the stellar masses using Spitzer observa-
usingtheChabrierinitialmassfunction(Chabrier2003). tionsandthesameprescriptionasPerleyetal.(2016b).Thehost
ofGRB071117liesveryclose(∼2(cid:48)(cid:48))toaredgalaxy,and,there-
fore,thespatialresolutionoftheSpitzerobservationsallowedus
2. Thesample
to obtain only an upper limit on its infrared flux. We therefore
Our sample is composed of the 27 host galaxies of the also performed a spectral energy distribution fitting using the
Swift/BAT6 complete sample of LGRBs at z < 2 with declina- host galaxy photometry (see Table 2) following the same pre-
tion Dec< 30◦. As the spatial distribution of GRB is isotropic, scriptionsasVerganietal.(2015),andfoundlog(M /M )∼9.9.
(cid:63) (cid:12)
this restriction does not introduce any bias in our results. The
choice to select only the LGRBs that are well observable from Table2.ObservedABmagnitudes(correctedbytheMilkyWayextinc-
the southern hemisphere was due to the availability of the X- tion)ofGRB071117hostgalaxy.
shooterspectrograph(Vernetetal.2011)attheESOVLT(Very
LargeTelescope)facilities,which,thankstoitswidewavelength Hostgalaxy g r i z K
GRB071117 24.4±0.1 24.7±0.2 24.8±0.3 >24.4 22.9±0.2
coverage,makespossiblethedetectionoftheemissionlinesnec-
essarytodeterminetheSFRandmetallicityofthehostgalaxies Notes.The g, r, i, z magnitudes have been determined from GROND
at z < 2. In particular, metallicity is available for 81% of the (Greiner et al. 2008) observations, whereas for the K value we used
sample (an estimate of the metallicity was not possible for five VLT/HAWKIobservations(Prog.ID:095.D-0560;P.I.:S.D.Vergani).
hostgalaxiesonly).
As the original Swift/BAT6 sample is selected essentially TheSFRvaluesweretakenfromKrühleretal.(2015)with
onlyonthebasisoftheLGRBpromptγ-rayflux,andnoother theexceptionofthehostgalaxiesofGRB061007,GRB061121
selectioncriterionisappliedwhengatheringthegalaxysample and GRB071117, not included in that work. We obtained the
(exceptthesouthernhemispherevisibility),oursampledoesnot VLT/X-shooter spectroscopy of these three host galaxies (ESO
suffer of any flux bias. Indeed, no correlation has been found programs 095.D-0560 and 085.A-0795, PI: S.D. Vergani and
between the prompt γ-ray emission and host galaxy properties H.Flores,respectively).Weprocessedthespectrausingversion
(seee.g.:Levesqueetal.2010b;Japeljetal.2016).Furthermore, 2.6.0oftheX-shooterdatareductionpipeline(Modiglianietal.
darkburstsarecorrectlyrepresentedinthesample(seeMelandri 2010),followingtheproceduresdescribedinJapeljetal.(2015).
etal.2012).Therestrictiontothesouthernhemisphereatz < 2 The measured emission line fluxes are reported in Table3. We
maintainsthiscondition,with26%ofLGRBofthesamplebeing determinetheSFRfromtheHαfluxes(correctedbytheextinc-
dark. tion determined through the Balmer ratio), with the same pre-
Forthepartofthesampleatz<1,Verganietal.(2015)and scriptionsasKrühleretal.(2015).
Japeljetal.(2016)reportthetableswiththeobjectsinthesample Following the same prescription as in Japelj et al. (2016),
andtheirproperties(includingstellarmasses,SFRandmetallic- wedeterminedthemetallicityoftheobjectsinthesamplewith
ity).TherestrictiontotheDec<30◦ excludesGRB080430and the Maiolino et al. (2008) method on the strong emission line
GRB080319Bfromthesampleusedinthiswork. fluxesreportedintheliterature(Piranomonteetal.2015;Krühler
The properties (redshift, stellar mass, SFR, and metallicity) etal.2015)oronthosemeasuredbyus;intherelevantcases,the
ofthe1 < z < 2partofthesamplearereportedinTable1.The results are consistent within errors to those already reported in
stellar masses were taken from Perley et al. (2016b), with the theliterature.
Articlenumber,page2of4
S.D.Verganietal.:Thechemicalenrichmentoflong-GRBnurseriesuptoz=2
3. TheFMRandMZrelation host galaxies for which the relevant lines to use this metallic-
ity indicator are available. The resulting MZ plot confirms the
avoidanceofsuper-solarmetallicityandtheshiftofhighstellar
masshostgalaxiestowardlowermetallicitythanthosefoundfor
generalstar-forminggalaxypopulationsatsimilarstellarmasses
andredshifts.
We stress that the five galaxies in the sample for which we
couldnotdeterminethemetallicity(GRB050318,GRB050525,
GRB050802, GRB060908, and GRB091208B) are all faint
galaxies, not hosting dark GRBs, and with stellar masses
log(M /M )< 9.2 (three of these galaxies have log(M /M )<
∗ (cid:12) ∗ (cid:12)
8.7;seeVerganietal.2015;Perleyetal.2016b).Asuper-solar
metallicity for a large portion of these host galaxies is there-
foreextremelyunlikely.Fortwoofthesegalaxies(GRB050525
and GRB050802) SFR limits are available (Japelj et al. 2016;
Palmerio et al. in preparation). Under the conservative hypoth-
esis that they follow the FMR relation, we can derive lim-
its on their metallicities from their SFR and stellar masses of
12+log(O/H)<8.1,8.4,respectively.
Wefurther investigatetheimplications ofour observational
resultsbycomparingthemwiththeexpectationsofadedicated
numerical simulation of the LGRB host galaxy population pre-
sented in Campisi et al. (2009, 2011), coupling high resolution
numerical simulation of dark matter with the semi-analytical
models of galaxy formation described in De Lucia & Blaizot
(2007). Previous work (De Lucia et al. 2004) has shown that
thesimulatedgalaxypopulationprovidesagoodmatchwiththe
observed local galaxies properties and relations among stellar
mass,gasmassandmetallicity.Moreover,Campisietal.(2011)
shows that the simulations nicely reproduce the observed FMR
ofSDSSgalaxiesanditsspread.FollowingCampisietal.(2011)
wecomputetheexpectednumberofLGRBshostedineachsim-
Fig.1. Toppanel:MZplot.Thedotscorrespondtothehostgalaxies
ulatedgalaxy,assumedtobeproportionaltothenumberofshort-
oftheSwift/BAT6sampleofLGRBsatz < 2,colorcodeddepending
living massive stars (i.e., star particles less than 5 × 107 yr in
ontheirredshiftasshownintherightbar.Thelinescorrespondtothe
age),applyingdifferentmetallicitythresholds(Z )fortheGRB
relationsfoundforfieldgalaxiesattheredshiftindicatednexttoeach th
line. Bottom panel: The FMR plane. The dots correspond to the host progenitor,withprobabilityequaltoonebelowZthandzerooth-
galaxies of the Swift/BAT6 sample of LGRBs at z < 2, color coded erwise.WeconstructtheFMRofsimulatedhostsintheredshift
depending on their redshift as shown in the right bar. The gray line range z = 0.3−2 and we determined the best-fit value of Z
th
corresponds to the FMR found by Mannucci et al. (2010, 2011). The by minimizing the χ2 against the BAT6 host data in the same
darkbluecurveandareacorrespondtoFMRrelationandofitsquartiles redshift interval. The best-fit model (see Fig.1) is obtained for
obtainedusingthesimulationofCampisietal.(2011).Thecyancurve Z = 0.73+0.08 Z (1σ errors). This is consistent with indirect
andareacorrespondtothebest-fitmodelresults. th −0.07 (cid:12)
results inferred from the distribution of the LGRB host stellar
massesatz<1(Verganietal.2015)oroftheinfraredluminosi-
In Fig.1 we plot the host galaxies of our sample in the MZ tiesoverawiderredshiftrange(Perleyetal.2016b).
and FMR spaces. The dearth of high metallicity galaxies is ev-
identaswellasthefactthattherearemoremassivegalaxiesat
4. Discussionandconclusions
thehigherredshifts(1<z<2)thanatz<1.
At low stellar masses (log(M∗/M(cid:12)) < 9.5) there is some In this paper we considered the properties of the host galax-
agreementwiththeMZrelationandFMRfoundforgeneralstar- ies of the complete Swift/BAT6 sample of LGRBs (Salvaterra
forminggalaxypopulations(seealsoJapeljetal.2016),whereas et al. 2012) that are visible from the southern hemisphere and
massive LGRB host galaxies are clearly shifted toward lower at z < 2. We studied them with respect to the MZ and FMR
metallicitiesthanpredictedbythegeneralrelations. relationoffieldstar-forminggalaxies.Thisisthefirststudycon-
WhiletheMZrelationevolvesinredshift,theFMRhasthe sideringatthesametimetheSFR,metallicity(bothdirectlyde-
advantage that it is redshift independent in the redshift range terminedfromthehostgalaxyspectroscopy),andstellarmasses
considered here, hence strengthening the statistics of our re- foracompletesampleofLGRBsandonalargeredshiftrange.
sults. For the general population of star-forming galaxies with Furthermore,weuseLGRBhostgalaxysimulationstointerpret
log(M )−0.32log(SFR)(cid:38) 9.2,theFMRisvaliduptoz ∼ 2.2, ourresults.
(cid:63)
hasbeendefinedoverSFRandstellarmassrangesencompass- Thanks to the sample extension to z ≈ 2, we could double
ing those of the host galaxies in our sample, and has a smaller thesamplesizecomparedtoJapeljetal.(2016)andshowforthe
scatter (0.06dex) than the MZ relation Mannucci et al. (2010, first time that LGRB host galaxies do not follow the FMR. We
2011). findthatLGRBsuptoz ≈ 2tendtoexplodeinapopulationof
Toverifythatourresultsareindependentofthemethodused galaxies with subsolar metallicity (Z∼ 0.5-0.8Z ). Our results
(cid:12)
to determine the metallicity, we used the Kobulnicky & Kew- are well reproduced by LGRB host galaxy simulations with a
ley (2004) R23 method to determine the metallicities of the 21 metallicitythresholdfortheLGRBproductionofZ ∼0.7Z .
th (cid:12)
Articlenumber,page3of4
A&Aproofs:manuscriptno.VerganiHGz2LE1c
Table3.Emissionlinefluxes(correctedforMWabsorption)ofthehostgalaxiesofGRB061007,GRB061121,andGRB071117inunits10−17erg
s−1cm−2.Upperlimitsaregivenatthe3σconfidencelevel.
Hostgalaxy [Oii]λ3726 [Oii]λ3729 [Neiii]λ3869 Hδ Hγ Hβ [Oiii]λ4959 [Oiii]λ5007 Hα [Nii]λ6583
GRB061007 -(a) 2.4±0.3 <0.7 <0.7 <1.7 1.0±0.4 1.3±0.8 9.5±1.4 4.4±0.4 <2.4
GRB061121 8.3±1.0 18.4±1.0 2.5±0.5 0.7±0.2 4.2±1.4 7.9±1.6 7.9±1.6 26.6±1.4 40.0±0.9 4.5±0.8
GRB071117 2.0±0.7 3.4±0.3 <0.4 <0.8 -(b) -(b) 3.0±0.6 6.6±1.0 5.6±1.0(c) <1.2
Notes.(a).Linestronglyaffectedbyaskyline.Todeterminethehostgalaxypropertieswefixeditsvalueto[Oii]λ3729/1.5(lowelectrondensity
case;Osterbrock1989).(b)Linesfallingontoonoisyregionstodetermineasignificantupperlimit.(c).Thelineiscontaminatedbyaskyline.
ThefluxhasbeendeterminedbyaGaussianfit,usingthepartofthelinenotcontaminatedbythesky.
Although strong metallicity gradients (> 0.1−0.2dex) are References
unlikely(onthe basisoflow-redshift,spatiallyresolved LGRB
Berger,E.,Cowie,L.L.,Kulkarni,S.R.,etal.2003,ApJ,588,99
host galaxies observations; Christensen et al. 2008; Levesque
Campisi,M.A.,DeLucia,G.,Li,L.-X.,Mao,S.,&Kang,X.2009,MNRAS,
etal.2011;Kruhleretal.inpreparation),wecannotexcludethat 400,1613
they are at play in the couple of galaxies showing evidences of Campisi,M.A.,Tapparello,C.,Salvaterra,R.,Mannucci,F.,&Colpi,M.2011,
MNRAS,417,1013
super-solarmetallicities(as,e.g.,inthecaseofGRB060306;see
Cano,Z.,Wang,S.-Q.,Dai,Z.-G.,&Wu,X.-F.2016,ArXiv:1604.03549
alsoNiinoetal.2015).Theexistenceofsomesuper-solarhosts Chabrier,G.2003,PASP,115,763
may as well indicate, however, that the formation of LGRBs Christensen,L.,Vreeswijk,P.M.,Sollerman,J.,etal.2008,A&A,490,45
DeLucia,G.&Blaizot,J.2007,MNRAS,375,2
is also possible above the general threshold, although at much
DeLucia,G.,Kauffmann,G.,&White,S.D.M.2004,MNRAS,349,1101
lowerrate.Applyingsmoothercutoffstothemetallicity,instead Fruchter,A.S.,Levan,A.J.,Strolger,L.,etal.2006,Nature,441,463
of the step function used here, shifts Z toward lower values Fryer,C.L.,Mazzali,P.A.,Prochaska,J.,etal.2007,PASP,119,1211
th
depending on the functional shape used. The present statistics Georgy,C.,Ekström,S.,Meynet,G.,etal.2012,A&A,542,A29
Graham,J.F.&Fruchter,A.S.2013,ApJ,774,119
doesnotallowustodiscriminatebetweendifferentcutoffshapes, Greiner,J.,Bornemann,W.,Clemens,C.,etal.2008,PASP,120,405
therefore we do not go into further detail. We point out how- Guetta,D.&DellaValle,M.2007,ApJ,657,L73
ever that none of them succeed in reproducing the super-solar Hirschi,R.,Meynet,G.,&Maeder,A.2005,A&A,443,581
Japelj,J.,Covino,S.,Gomboc,A.,etal.2015,A&A,579,A74
metallicityvalue.ItshouldalsobestressedthattheGRB060306 Japelj,J.,Vergani,S.D.,Salvaterra,R.,etal.2016,A&A,590,A129
metallicityisveryuncertainwithprettylargeerrorbars. Kelly,P.L.,Filippenko,A.V.,Modjaz,M.,&Kocevski,D.2014,ApJ,789,23
Kennicutt,Jr.,R.C.,Bresolin,F.,&Garnett,D.R.2003,ApJ,591,801
The relatively high metallicity threshold found in this work Kewley,L.J.&Dopita,M.A.2002,ApJS,142,35
Kewley,L.J.&Ellison,S.L.2008,ApJ,681,1183
is much higher than required from standard collapsar models
Kobulnicky,H.A.&Kewley,L.J.2004,ApJ,617,240
(but see Georgy et al. 2012). Binary stars are a possible solu- Krühler,T.,Malesani,D.,Fynbo,J.P.U.,etal.2015,A&A,581,A125
tion as progenitors, although detailed models studying the role Levesque,E.M.,Berger,E.,Soderberg,A.M.,&Chornock,R.2011,ApJ,739,
23
of metallicity on the fates of binary stars are missing. How-
Levesque,E.M.,Kewley,L.J.,Berger,E.,&Zahid,H.J.2010a,AJ,140,1557
ever,itisimportanttonotethatthemetallicitiesdeterminedus- Levesque,E.M.,Soderberg,A.M.,Kewley,L.J.,&Berger,E.2010b,ApJ,725,
ing strong emission lines are not absolute values (see Kewley 1337
López-Sánchez,Á.R.,Dopita,M.A.,Kewley,L.J.,etal.2012,MNRAS,426,
&Ellison2008).Inourcase,theyarerelativetotheKewley&
2630
Dopita (2002) photoionization models on which the Maiolino
Maiolino,R.,Nagao,T.,Grazian,A.,etal.2008,A&A,488,463
et al. (2008) method is based. On the one hand, some works Mannucci,F.,Cresci,G.,Maiolino,R.,Marconi,A.,&Gnerucci,A.2010,MN-
seem to indicate that those models may overestimate oxygen RAS,408,2115
Mannucci,F.,Salvaterra,R.,&Campisi,M.A.2011,MNRAS,414,1263
abundances by ∼ 0.2-0.5dex compared to the metallicity de-
Melandri,A.,Sbarufatti,B.,D’Avanzo,P.,etal.2012,MNRAS,421,1265
rived using the so-called direct Te method (see e.g., Kennicutt Metzger,B.D.,Margalit,B.,Kasen,D.,&Quataert,E.2015,MNRAS,454,
etal.2003;Yinetal.2007).Ontheotherhand,otherworks(see 3311
Modigliani, A., Goldoni, P., Royer, F., et al. 2010, in Proc. SPIE, Vol. 7737,
e.g.,López-Sánchezetal.2012;Nichollsetal.2012)foundthat
ObservatoryOperations:Strategies,Processes,andSystemsIII,773728
the oxygen abundances determined using temperatures derived Modjaz,M.,Kewley,L.,Kirshner,R.P.,etal.2008,AJ,135,1136
fromcollisional-excitedlinescouldbeunderestimatedby∼0.2- Nicholls,D.C.,Dopita,M.A.,&Sutherland,R.S.2012,ApJ,752,148
0.3dex. In principle, the simulations should be independent of Niino,Y.,Nagamine,K.,&Zhang,B.2015,MNRAS,449,2706
Nuza,S.E.,Tissera,P.B.,Pellizza,L.J.,etal.2007,MNRAS,375,665
thesemodelsandthereforethecurvesderivedinthisworkfrom
Osterbrock,D.E.1989,Astrophysicsofgaseousnebulaeandactivegalacticnu-
simulationsshouldnotbeaffectedbythisissue. clei(MillValley,CA,UniversityScienceBooks)
Perley,D.A.,Niino,Y.,Tanvir,N.R.,Vergani,S.D.,&Fynbo,J.P.U.2016a,
TheZ ∼ 0.7 Z thresholdshouldnotbeconsidered,there- SpaceSci.Rev.[arXiv:1602.00770]
th (cid:12)
Perley,D.A.,Tanvir,N.R.,Hjorth,J.,etal.2016b,ApJ,817,8
fore,asanabsolutevalue.Nonetheless,tobeinagreementwith
Piranomonte,S.,Japelj,J.,Vergani,S.D.,etal.2015,MNRAS,452,3293
themetallicities(Z≤0.2Z(cid:12))neededinmostLGRBsinglemas- PlanckCollaboration,Ade,P.A.R.,Aghanim,N.,etal.2014,A&A,571,A16
sive star progenitor models, all the metallicities presented here Salvaterra,R.,Campana,S.,Vergani,S.D.,etal.2012,ApJ,749,68
Schulze,S.,Klose,S.,Björnsson,G.,etal.2011,A&A,526,A23
shouldbesystematicallyoverestimated,mostofthembyatleast
Smartt,S.J.2015,PASA,32[arXiv:1504.02635]
∼0.5dex. Svensson,K.M.,Levan,A.J.,Tanvir,N.R.,Fruchter,A.S.,&Strolger,L.-G.
2010,MNRAS,405,57
Tremonti,C.A.,Heckman,T.M.,Kauffmann,G.,etal.2004,ApJ,613,898
Acknowledgements. Thisworkisbasedinpartonobservationsmadewiththe
Trenti,M.,Perna,R.,&Jimenez,R.2015,ApJ,802,103
SpitzerSpaceTelescope(programs90062and11116),whichisoperatedbythe vandenHeuvel,E.P.J.&PortegiesZwart,S.F.2013,ApJ,779,114
JetPropulsionLaboratory,CaliforniaInstituteofTechnologyunderacontract vandenHeuvel,E.P.J.&Yoon,S.-C.2007,Ap&SS,311,177
withNASA.SDVthanksM.Rodrigues,H.FloresandF.Hammerforuseful Vergani,S.D.,Salvaterra,R.,Japelj,J.,etal.2015,A&A,581,A102
discussions.JJacknowledgesfinancialcontributionfromthegrantPRINMIUR Vernet,J.,Dekker,H.,D’Odorico,S.,etal.2011,A&A,536,A105
2012201278X4FL002.TKacknowledgessupportfromaSofjaKovalevskaja Woosley,S.E.&Heger,A.2006,ApJ,637,914
AwardtoPatriciaSchady.WethanksG.CupaniforsharinghisexpertiseonX- Yin,S.Y.,Liang,Y.C.,Hammer,F.,etal.2007,A&A,462,535
shooterdatareduction. Yoon,S.-C.&Langer,N.2005,A&A,443,643
Articlenumber,page4of4