A(MNLTEXstylefilev1.4)
TheInfraredGlowofFirstStars
RubenSalvaterra1,ManuelaMagliocchetti2,AndreaFerrara2,RaffaellaSchneider3,4
1DipartimentodiFisicaeMatematica,Universit´adell’Insubria,ViaValleggio11,22100Como,Italy
2SISSA/InternationalSchoolforAdvancedStudies,ViaBeirut4,34100Trieste,Italy3OsservatorioAstrofisicodiArcetri,LargoEnricoFermi5,50125Firenze,Italy4“EnricoFermi”Centre,ViaPanisperna89/A,00184Roma,Italy
arXiv:astro-ph/0512403v2 16 Jan 20065February2008
ABSTRACT
Kashlinskyetal.(2005)findasignificantcosmicinfraredbackgroundfluctuationex-cessonangularscales>∼50arcsecthatcannotbeexplainedbyinstrumentalnoiseorlocalforegrounds.Theexcesshasbeententativelyattributedtoemissionfrompri-mordialverymassive(PopIII)starsformed≤200MyraftertheBigBang.Usinganevolutionarymodelmotivatedbyindependentobservationsandincludingvariousfeedbackprocesses,wefindthatPopIIIstarscancontribute<40%ofthetotalback-groundintensity(νJν∼1−2nWm−2sr−1inthe0.8-8µmrange)producedbyallgalaxies(hostingbothPopIIIandPopIIstars)atz≥5.Theinfraredfluctuationexcessisinsteadverypreciselyaccountedbytheclusteringsignalofgalaxiesatz≥5,predominantlyhostingPopIIstarswithmassesandpropertiessimilartothepresentones.
Keywords:galaxies:clustering-galaxies:infrared-cosmology:theory-large-scalestructure-cosmology:observations
1INTRODUCTION
ObservationsoftheinfraredbackgroundprovideimportantinformationontheemissionofcosmicluminoussourcesthroughoutthehistoryoftheUniverse.Ithasbeensug-gested(Santos,Bromm&Kamionkowski2003;Salvaterra&Ferrara2003)thatalargefractionofthemeasuredNear-InfraRed(1-10µm)cosmicBackground(NIRB)arisesfromredshiftedLyαlinephotonsandnebularemissionproducedbythefirstverymassivemetal-freestars.Thishypothesis,however,isverydemandingintermsoftherequiredconver-sionefficiencyofbaryonsintostars(Madau&Silk2005).AlargeNIRBcontributionfromsuchstarshasmorerecentlybeenrejectedbythepaucity(≤3)ofz∼10candidatesourcesinHubbleSpaceTelescopeultra-deepobservations(Salvaterra&Ferrara2005).Nevertheless,amoremodestcontributionfromveryhighredshiftgalaxies,whoseclus-teringshouldleaveadistinctsignatureonsmall-scalean-gularfluctuationsofthebackgroundlight(Magliocchetti,Salvaterra&Ferrara2003;Kashlinskyetal.2004;Coorayetal.2004),isstillpossible.
Kashlinskyetal.(2005)haverecentlyfoundsignifi-cantNIRBfluctuationsindeepexposuredataobtainedwithSpitzer/IRAC(Fazioetal.2004a,2004b)infourchannels(3.6,4.5,5.8,and8µm),afterGalacticstarsandgalaxiesbrightenoughtobeindividuallyresolvedbytheinstrumenthavebeencarefullysubtracted.Withtheonlyexceptionofthe8µmchannel,theshapeandamplitudeofthepower
c0000RAS
spectrumcannotbereproducedbyeithercontributionsfrom
interveningdusty,Galacticneutralhydrogengas(cirrus)orfromlocalinterplanetarydust(zodiacallight).Ordinarygalaxies(z<∼5)producefluctuationsduetotheirclusteringandshot-noise.Thefaintfluxlimits(≥0.3µJy)ofSpitzerdataallowtopushtheirresidualclusteringcontributionbe-lowtheleveloftheexcesssignalatrelativelylarge(>∼50arc-sec)angularscales(Kashlinskyetal.2005).Theshotnoisecomponent,estimateddirectlyfromgalaxycounts,fitstheobservedfluctuationsatsmallerangularscales,andrapidlyfadesawayatlargerangles.TheresiduallargescalesignalhasbeenascribedbyKashlinskyetal.(2005)ascomingfromverydistant(z≥5)sourcesprovidedtheirtotalfluxcontributionis>1nWm−2sr−1.TheaimofthisLetteristoshowthatthisisindeedthecase.
Thelayoutofthepaperisasfollows:inSection2wewillbrieflydescribetheadoptedmodel,whileinSection3weprovidepredictionsfortheNIRBintensityandfluctuationsandcomparethelatteroneswiththeresultsofKashlinskyetal.(2005).Section4summarizesourconclusions.
2THEMODEL
Schneideretal.(2005)havepresentedascenariofortheformationofgalaxiesinaconcordanceΛCDMcosmolog-
2Salvaterra,Magliocchetti,Ferrara,Schneider
icalmodel⋆whichincludesaself-consistenttreatmentoftwokeyfeedbackprocesses:(i)radiativefeedback,suppress-ingstarformationinH2-coolinghalosandtheformationoflow-massgalaxiesduetotheeffectsofUVbackgroundradi-ationapproachingthereionizationepoch,and(ii)chemicalfeedback,whichcontrolsthetransitionfrommetal-freestars(PopIII)toordinarystars(PopII)throughtheprogressiveenrichmentofstarforminggaswithheavyelementsreleasedbysupernovaexplosions(Schneideretal.2002,2004,2005;Brommetal.2001).Chemicalfeedbackpropagatesthroughthehierarchyofgalaxymergersfromprogenitorstotheirdescendantssothat,ateachredshift,existinghaloswhichareallowedtoformstarsareclassifiedasPopII(PopIII)galaxiesdependingonwhetherthehaloitselforanyofitsprogenitorshave(havenot)alreadyexperiencedanepisodeofstarformation.
Withinthismodelwecancomputethecomovingspe-cificemissivity,ǫν,whichisgivenby∞
max(z′)
ǫν(z)=
dz′
lν(tz,z′)
Md2n
⋆
z
MMmin(z′)
τeff(ν0,z0,z)
dl
4π
∞
ǫν(z)e−z0
oldSd)isdefinedas:
C(θ)=δJ(θ′,φ′)δJ(θ′′,φ′′),
(3)
where(θ′,φ′)and(θ′′,φ′′)identifytwopositionsontheskyseparatedbyanangleθ.Theaboveexpressioncanbewrit-tenasthesumoftwoterms,CPandCC,thefirstoneduetoPoissonnoise(i.e.fluctuationsgivenbyrandomlydistributedobjects),andthesecondoneowingtosourceclustering.Itcanbeshownthattheshotnoisecontributionoriginatingfromz≥5galaxiesisnegligible,soweonlycon-centrateonfluctuationswhichstemfromtheclusteringofthesesources,i.e.weassumeC(θ)≡CC.
ThemethodadoptedhereissimilartothatpresentedinMagliocchettietal.(2003),wherebyangularfluctuationsareobtainedbymeansoftheexpression:C(θ)=
1
(1+z)2
e
−2τeff
dx
dMh
(Mh,z)f(Mh,z)
(M,(5)
dMh
h,z)f(Mh,z)wheredn/dMh(Mh,z)istheSheth&Tormen(1999)mass
functionofthedarkmatterhalos,f(Mh,z)isthefractionofhalosthatatanymassandredshifthostaPopII(PopIII)galaxyandMmin(z),Mmax(z)arethelimitingmassvaluesforsuchhalostohosteitheraPopIIoraPopIIIsource(seeSection2).
TheaboveprescriptionsthenallowtoestimateC(θ),i.e.thecontributionfromtheclusteringofunresolvedz≥5(PopIIandPopIII)galaxiestothebackgroundfluctuationsatdifferentwavelengths.Animportantfeaturetomention
c0000RAS,MNRAS000,000–000TheInfraredGlowofFirstStars
3
Figure2.PowerspectrumoffluctuationsattheIRACbands.Differentsymbolsshowdataofdifferentobservedfields(Kashlin-skyetal.2005):filledcirclescorrespondstotheQSO1700field,opencirclestoEGS.Dataareaveragedoverwidebinstoincrease
thesignal-to-noiseratio.TheerrorsareN−1/2
q,whereNqisthenumberofFourierelementsatthegivenq-bin.Dotted(dashed)linesshowthecontributionfromPopII(PopIII)galaxies.Dot-dashedlinesshowthecontributionofshotnoisefromremaininggalaxiesfainterthanthelimitingmagnitude(Kashlinskyetal.2005).Thesolidlineisthesumofthedifferentcomponents.
abouttheresultingC(θ)isthat–independentofthecon-sideredfrequencyandonlyduetothehighredshiftofthesourcesinexam–itfeaturesasharpdropataround300arcsecondbeyondwhichitrapidlyapproacheszero.Thisim-pliesthatsignalsonscaleslargerthantheabovevaluesuchasthosedetectedbysomeexperiments(e.g.Hauser&Dwek2001;Matsumotoetal.2005)havetorelyonthepresenceofamorelocalpopulationofunresolvedsources.
Finally,wecanmovetothecalculationofthepowerspectrumP(q)originatingfromtheabovefluctuations,∞
P(q)=2π
C(θ)J0(θ,q)θdθ,(6)
0
whereJ0isthezero-thorderBesselfunction.
InordertoconfrontourpredictionswiththeKashlinskyetal.(2005)results,P(q)hasbeenevaluatedinthe3.6,4.5,5.8and8µmIRACbands.OurfindingsaresummarizedinFigure2,wherethedottedlinesrepresentthefluctuationsoriginatingfromPopIIgalaxies,thedashedlinesthoseduetoPopIIIsourcesandthedot-dashedcurvesindicatetheshotnoisecontributionfromgalaxiesfainterthanthelim-itingmagnitudeasestimatedbyKashlinskyetal.(2005).Itisclearfromtheplotthatatthelowestfrequenciesthecontributionofz≥5galaxiescanprovideanexcellentfittotheobservedfluctuations;inparticularitdominatesovertheshot-noisecomponentatallangularscalesgreaterthan∼50arcsec.At5.8µm,theirimportanceisreducedbutthepresenceofz≥5sourcesisstillrequiredtoproperlyaccount
4Salvaterra,Magliocchetti,Ferrara,Schneider
fortheamplitudeofthesignalatthelargestprobedscales.Finally,highredshiftgalaxiesandshot-noisealonecannotreproducethepowerspectrumat8µm,wherethecirrusprobablybecomesthedominantcomponent(Kashlinskyetal.2005).
Asalastpointitisworthmentioningthat,asexpectedwithinthepresentscenario,thecontributionofPopIIIgalax-iesisnegligibleinallIRACbands;theamplitudeoftheirsignalisinfact∼50timeslowerthanthatofPopIIgalaxiesinthemostfavorablecase(3.6µm),whilethePopIII-to-PopIIcontributionratiocangoaslowas10−3atthehighestfrequencyprobedbyIRAC.
4CONCLUSIONS
Usingaphysically-motivated,observationally-testedmodeloftheearlyUniverse(Schneideretal.2005),wecomputetheexpectedbackgroundradiationintheNIRbysourcesformingwhentheUniversewas<1Gyrold.Wefindthatthebackgroundintensity,νJν∼1−2nWm−2sr−1,isalmostconstantinthe0.8-8µmrange.PopIIgalaxiesdomi-natetheNIRBintheentirewavelengthrange,whilePopIIIgalaxiescontributeatmost40%ofthetotalintensity(atλ∼1.5µm),viatheirstrongLyαlineemission.Finally,wefoundthattheinfraredfluctuationexcessonangularscales≥50arcsecdetectedbySpitzer/IRAC(Kashlinskyetal.2005)isaccountedverypreciselybytheclusteringsignalofgalaxiesatz≥5predominantlyhostingstarswithmassesandpropertiessimilartothepresentones.
Twoadditionalpointsareworthnoticing:(i)verymas-sivestars(M≥100M⊙)donotneedtobeinvokedtoex-plainNIRBfluctuationsandreionizationhistory;(ii)be-causeoftheirsmallcontribution(P(q)≤10−10nW2m−4sr−1)totheobservedpowerspectruminallchannels,ex-tractingthesignalofthe(very)firstPopIIIstarsisex-tremelychallenging.Futureinstruments(astheJamesWebbSpaceTelescope)willbeabletodirectlyidentifythesesourcesuptoz=10orabove.Finally,theintensityoftheNIRBprovidedbyz≥5galaxiesfallsshortofaccountingfortheexcessmeasuredbyIRTS(Matsumotoetal.2005)andDIRBE(Hauser&Dwek2001)experiments.Theori-ginofthiscomponentremainsverypuzzling(Salvaterra&Ferrara2005)andmightrequireeitherarevisionofcurrentmodelofzodiacallightsubtractionortheexistenceofalargepopulationoffaintgalaxieslocatedatz≃2−3(orboth).ImportantinsightsontheseissuesareexpectedfromtheupcomingCIBERexperiment(Bocketal.2005),thatwillbeabletosimultaneouslymeasurethetotalNIRBinten-sityandfluctuationpowerspectruminthepoorlyknownwavelengthrange0.8-2µm.Suchinstrument,inaddition,willallowaclearseparationofthecosmologicalsignalfromlocalforegrounds(i.e.zodiacallight).
REFERENCES
BockJ.etal.,2005,proceedingsofUCIrvineMay2005workshop
on”FirstLight&Reionization”,eds.E.Barton&A.Cooray,NewAstronomyReviews,inpress,astro-ph/0510587
Bouwens,R.J.,Illingworth,G.D.,Thompson,R.I.&Franx,M.,
2005a,ApJ,624,L5.
Bouwens,R.J.,Illingworth,G.D.,Blakeslee,J.P.&Franx,M.,
2005b,ApJinpress.
Bromm,V.,Ferrara,A.,Coppi,P.S.,Larson,R.B.,2001,MN-RAS,328,969
Bruzual,G.&Charlot,S.,2003,MNRAS,344,1000.
Cooray,A.,Bock,J.,Keating,B.,Lange,A.&Matsumoto,T.
2004,ApJ,606,611.
Fazio,G.G.etal.,2004a,ApJS,154,10.Fazio,G.G.etal.,2004b,ApJS,154,39.
Hauser,M.G.&Dwek,E.,2001,ARA&A,39,249.
Kashlinsky,A.,Arendt,R.,Gardner,J.P.,Mather,J.&Moseley,
S.H.,2004,ApJ,608,1.
Kashlinsky,A.,ArendtR.G.,Mather,J.&Moseley,S.H.,2005,
Nature,438,45.
Kogut,A.etal.,2003,ApJS,148,161.
Madau,P.&Silk,J.,2005,MNRAS,359,L37.
Magliocchetti,M.,Salvaterra,R.&Ferrara,A.,2003,MNRAS,
342,L25.
MagliocchettiM.,PorcianiC.,2003,MNRAS,346,186Matsumoto,T.etal.,2005,ApJ,626,31.
Peacock,J.A.&Dodds,S.J.1996,MNRAS,280,L19.Salvaterra,R.&Ferrara,A.,2003,MNRAS,339,973.
Salvaterra,R.&Ferrara,A.,2005,MNRASinpress,astro-ph/0509338.
Santos,M.,BrommV.&Kamionkowski,M.,2003,MNRAS,336,
1082.
Schaerer,D.,2002,A&A,382,28.
Schneider,R.,Ferrara,A.,Natarajan,P.,Omukai,K.,2002,ApJ,
571,30.
Schneider,R.,FerraraA.,Salvaterra,R.,Omukai,K.,Bromm,
V.,2004,Nature,422,869.
Schneider,R.,Salvaterra,R.,Ferrara,A.&Ciardi,B.,2005,sub-mittedtoMNRAS,astro-ph/0510685.
Sheth,R.K.&Tormen,G.,1999,MNRAS,308,119.Spergel,D.N.etal.,2003,ApJS,148,175.
c0000RAS,MNRAS000,000–000
因篇幅问题不能全部显示，请点此查看更多更全内容