Table Of Content

Astronomy&Astrophysicsmanuscriptno.final (cid:13)c ESO2008 February5,2008 The nature of the X-Ray Flash of August 24 2005. Photometric evidence for an on-axis z = 0.83 burst with continuous energy 7 ⋆ injection and an associated supernova? 0 0 2 J.Sollerman1,2,J.P.U.Fynbo1,J.Gorosabel3,J.P.Halpern4,J.Hjorth1,P.Jakobsson1,5,N.Mirabal4,D.Watson1, n D.Xu1,A.J.Castro-Tirado3,C.Fe´ron1,A.O.Jaunsen6,M.Jel´ınek3,B.L.Jensen1,D.A.Kann7,J.E.Ovaldsen6, a J A.Pozanenko8,M.Stritzinger1,C.C.Tho¨ne1,A.deUgartePostigo3,S.Guziy3,M.Ibrahimov9,S.P.Ja¨rvinen10,11, 5 A.Levan5,V.Rumyantsev12,andN.Tanvir5,13 2 1 DarkCosmologyCentre,NielsBohrInstitute,UniversityofCopenhagen,JulianeMariesVej30,DK–2100CopenhagenØ,Denmark 1 v 2 StockholmObservatory,DepartmentofAstronomy,AlbaNova,S-10691Stockholm,Sweden 6 3 InstitutodeAstrof´ısicadeAndaluc´ıa(IAA-CSIC),P.O.Box3.004,E-18.080Granada,Spain 3 4 ColumbiaAstrophysicsLaboratory,550West120thStreet,ColumbiaUniversity,NewYork,NY10027-6601,USA 7 5 CentreforAstrophysicsResearch,UniversityofHertfordshire,CollageLane,Hatfield,Herts,AL109AB,UK 1 6 InstituteofTheoreticalAstrophysics,POBox1029,N-0315Oslo,Norway 0 7 Thu¨ringerLandessternwarteTautenburg,Sternwarte5,D–07778Tautenburg,Germany 7 8 SpaceResearchInstitute(IKI),84/32ProfsoyuznayaStr,Moscow117997,Russia 0 9 UlughBegAstronomicalInstitute,Tashkent700052,Uzbekistan / h 10 AstrophysikalischesInstitutPotsdam,AnderSternwarte16,D-14482Potsdam,Germany p 11 AstronomyDivision,P.O.Box3000,FIN-90014UniversityofOulu,Finland - 12 CrimeanLaboratoryofSternbergAstronomicalInstituteMSU,Nauchny,Crimea,98409,Ukraine o 13 DepartmentofPhysicsandAstronomy,UniversityofLeicester,Leicester,LE17RH,UK r t s a : ABSTRACT v i X Aims.OuraimistoinvestigatethenatureoftheX-RayFlash(XRF)ofAugust24,2005. r Methods.WepresentcomprehensivephotometricR-bandobservationsofthefadingopticalafterglowofXRF050824,from11minutesto104 a daysaftertheburst.InadditionwepresentobservationstakenduringthefirstdayintheBRIKbandsandtwoepochsofspectroscopy.Wealso analyseavailableX-raydata. Results. TheR-bandlightcurveof theafterglow resemblesthelightcurvesof longduration Gamma-RayBursts(GRBs),i.e.,apower-law, albeitwitharathershallowslopeofα = 0.6(F ∝ t−α).OurlateR-bandimagesrevealthehostgalaxy.Therest-frameB-bandluminosityis ν ∼0.5L∗.Thestar-formationrateasdeterminedfromthe[OII]emissionlineis∼1.8M yr−1.Whenaccountingforthehostcontribution,the ⊙ slopeisα=0.65±0.01andabreakinthelightcurveissuggested.Apotentiallightcurvebumpat2weekscanbeinterpretedasasupernova onlyifthisisasupernovawithafastriseandafastdecay.However,theoverallfitstillshowsexcessscatterinthelightcurveintheformof wigglesandbumps.TheflatlightcurvesintheopticalandX-rayscouldbeexplainedbyacontinuousenergyinjectionscenario,withanon-axis viewingangleandawidejetopeningangle(θj∼>10◦).Iftheenergyinjectionsareepisodicthiscouldpotentiallyhelpexplainthebumpsand wiggles. Spectroscopyoftheafterglowgivesaredshiftofz=0.828±0.005frombothabsorptionandemissionlines.Thespectralenergydistribution (SED)oftheafterglowhasapower-law(F ∝ν−β)shapewithslopeβ=0.56±0.04.ThiscanbecomparedtotheX-rayspectralindexwhich ν isβ =1.0±0.1.ThecurvatureoftheSEDconstrainsthedustreddeningtowardsthebursttoA <0.5mag. X v Keywords. cosmology:observations—gammarays:bursts— ⋆ This paper is based on observations from a multitude of tele- scopes,forexampleonobservationsmadewithESOTelescopesatthe ParanalObservatory(programmeID075.D-0270)andwiththeNTT atedontheislandofLaPalmajointlybyDenmark,Finland,Iceland, and ESO/Danish 1.5-m telescope at the La Silla Observatory. Also Norway, andSweden, intheSpanishObservatorio delRoque delos basedonobservationsmadewiththeNordicOpticalTelescope,oper- MuchachosoftheInstitutodeAstrofisicadeCanarias. 2 J.Sollermanetal.:XRF050824 1. INTRODUCTION 2006; Wiersemaetal., 2007) andunusualburst, XRF060218, with a very low E ∼ 5 keV (Campanaetal., peak X-Ray Flashes (XRFs) are similar to Gamma-Ray Bursts 2006) showed unambiguous evidence for an associated su- (GRBs), but with most of the fluence of the prompt emis- pernova (e.g., Sollermanetal., 2006; Modjazetal., 2006; sion detected in the X-ray band. Their existence as a class Mirabaletal., 2006; Pianetal., 2006). SN2006aj associated was suggested by Heiseetal. (2001) based on data from the with XRF060218 clearly established the close link between BeppoSAXsatellite. SNe,(GRBs)andXRFs. The high energy spectra (νF ) of the prompt emission ν However, other XRFs with late time coverage did ap- fromGRBsarewelldescribedbytheso-calledBandfunction parently not show any clear evidence for a supernova bump (Bandetal., 1993), which is composedof two smoothly con- (Soderbergetal., 2005; Levanetal., 2005). This could point nected power-laws. The energy at which the two power-laws to a differencein theoriginofXRFs andthe longGRBs with connect, E , is where most of the energy is emitted. For peak accompanying SNe (but see the recent paper by Fynboetal., classicalGRBs,E istypicallyafew100keV(Preeceetal., peak 2006, for GRBs with no associated supernova emission), al- 2000). The spectra of the prompt emission of XRFs are also thoughanysuchclaimhastoawaitobservationsofXRFswith wellfittedbytheBandfunction,butwithvaluesofE below peak a betterdetermineddistancescale.Mostofthe XRFsclaimed .50keVandinsomecasesevenbelow10keV(Kippenetal., to lack SNe in the above-mentionedstudies had no measured 2003;Barraudetal.,2003). redshifts. Sakamotoetal. (2005) argue, in accordance with previ- We here present data for XRF050824 for which we have ous studies, that the spectral distributions of XRFs, X-ray obtainedawellmonitoredopticallightcurveandasecurespec- RichGRBsandGRBsformacontinuum,suggestingthatthey troscopicredshift. all arise from the same phenomenon (see also Barraudetal., 2003, 2005). There has also been growing evidence that (at least some) XRFs are the result of classical GRBs seen off- 1.1.XRF050824 axis(Yamazakietal.,2002,2003;Rhoads,2003;Fynboetal., XRF050824 was detected by the BAT instrument on-board 2004;Granotetal.,2005). theSwiftsatellite(Campanaetal.,2005a)onAugust24.9669 Other suggestions include XRFs as either dirty fireballs, 2005UT (UniversalTimeisusedthroughoutthearticle).The which are relativistic jets with a larger baryon load and burst duration (T ) was 25 s, i.e., this was a long-duration hence (assuming external shocks) lower Γ-factors than those burst.Thetotalflu9e0nceinthe15–150keVbandwas∼2.3×10−7 ofclassical GRBs (Dermeretal., 1999; Heiseetal., 2001), or ergs cm−2 (Krimmetal., 2005). The burst was also detected XRFs as clean fireballs where (assuming internal shocks) the by the FREGATE instrument onboard HETE-2 (Crewetal., spread in Γ-factors is small but the average Γ-factor is large 2005). As seen by HETE-2 the value of the fluence ratio (Barraudetal.,2005). S(2−30)/S(30−400)=2.7.Crewetal.(2005)estimatedE < peak However, there are still open questions regarding the ori- 12.7keV.GRB050824isthusclearlyanXRF. gin of XRFs. While the connection between long GRBs and The BAT on-board localization was reported to an accu- certain core-collapse supernovae appears to be well estab- racyof3arcminutes.EarlyROTSE-IIIdatadidnotdetectany lished (Hjorthetal., 2003; Mathesonetal., 2003; Zehetal., new object (Schaeferetal., 2005), but our observations start- 2004), the case has not been as well defined for XRFs. ing38minutesafterthetriggerrevealedanewobjectatR.A.= Fynboetal.(2004)performedthefirstcomprehensiveobserva- 00:48:56.1,Dec=+22:36:32(J2000,seeSect.3.1),whichwe tionalcampaignofanXRFopticalafterglow,forXRF030723 laterconfirmedasthecounterpart(Gorosabeletal.,2005). (see also Butleretal., 2005). The wellsampled lightcurvefor Inthispaperwefocusonourcomprehensiveopticalstudy XRF030723displayedseveralinterestingfeatures: of this afterglow. The article is organized as follows. Sect. 2 (i) Theveryearlylightcurvewasessentiallyflat, inaccor- outlines how the optical observations were obtained and re- dancewithmodelsforwhichXRFsareviewedawayfromthe duced.TheresultsarepresentedinSect.3,whichincludesthe jetaxis. astrometry,theopticallightcurve,thespectralenergydistribu- (ii)Thefollowingpower-lawdecaywasverysimilartothat tion, the spectroscopicresults, data on the hostgalaxyand an seenintypicalGRBs,suggestingacommonorigin. analysis of available X-ray data of the afterglow. Finally, we (iii)At∼16dayspastburstastrongbumpinthelightcurve endthepaperwithadiscussion(Sect.4)andsomeconclusions suggested the presence of a fast rising supernova. The super- (Sect.5). novainterpretationwasarguedto beconsistentwiththespec- tral energy distribution (SED) evolution (Fynboetal., 2004) and was later supported by modeling both of the afterglow 2. OBSERVATIONS (Granotetal., 2005) and of the supernova (Tominagaetal., 2.1.Photometry 2004). More evidence has now been presented arguing for a Our very first observations of XRF050824 were obtained common origin for GRBs and XRFs. XRF020903 had a with the BOOTES-1B 30 cm robotic telescope (e.g., late lightcurve and spectrum consistent with a supernova Castro-Tiradoetal., 2004) in southern Spain, which detected at z = 0.25 (Soderbergetal., 2005; Bersieretal., 2006). the burst from 10.6 minutes after the high energy event in More recently, the nearby (z = 0.0331, Mirabal&Halpern, several R-band exposures. However, since these images were J.Sollermanetal.:XRF050824 3 Table1.TelescopesandInstruments We extracted the spectrum using standard procedures within IRAF. Wavelength calibration was obtained using im- Telescope Instrument/ FOV Pixelscale ages of HeNeAr lamps obtained as part of the morning cali- CCD (arcminutes) (arcsecpixel−1) brations. Flux calibration was performedusing spectra of the BOOTES-1B ST8E 40×26 1.6 spectrophotometricstandardstarG138-31(Oke1990). OSN ROPER 7.92×7.92 0.232 NOT ALFOSC 6.3×6.3 0.189 NOT STANCAM 3×3 0.176 3. RESULTS D1.5m DFOSC 13.7×13.7 0.395 MDM1.3m SITeCCD 8.6×8.6 0.508 3.1.Astrometry MDM2.4m SITeCCD 9.4×9.4 0.275 CrAO2.6 FLI-IMG1001E 8.5×8.5 0.5 WedeterminedthecelestialpositionoftheXRF050824optical Maidanak1.5m SITeCCD 8.5×3.5 0.266 afterglowasthemeanastrometricsolutionfoundin10OSNR- WHT ULTRACAM 5×5 0.3 bandimages.Eachafterglowpositionisbasedon∼50USNO- NTT EMMI 9.9×9.1 0.33 A2.0referencestarsperimage.Themeanvalueofthecoordi- VLT FORS1 6.8×6.8 0.20 natesare: VLT FORS2 6.8×6.8 0.25 R.A.(J2000)=00:48:56.14±0.03s, VLT ISAAC 2.5×2.5 0.148 Dec(J2000)=+22:36:33.2±0.4′′ These astrometric errors include the 0.25′′ systematic error of the USNO-A2.0 catalogue (Assafinetal., 2001; Deutsch, not processed until later, the actual discovery was instead 1999). made via the studies we initiated 38 minutes after the burst (Gorosabeletal., 2005) using the 1.5 meter telescope at the 3.2.TheLightcurve ObservatoriodeSierraNevada(OSN) We conducteda comprehensivestudy of the optical after- The photometryof the XRF was carried outusing either PSF glowoverthefollowing100daysusingseveraltelescopesand fittingphotometry(whentheafterglowwasbright)oraperture instruments. In Table 1 we summarize the telescopes and in- photometry(at later times when the host started to contribute strumentsusedandprovidedetailsonthefield-of-view(FOV) significantly). We measured the magnitudes of the optical af- andpixelscaleoftheseinstruments. terglowaswellas4starsinthefield.Therelativemagnitudes WeusedtheESO/Danish1.54mtelescope(D1.5m)onLa weretransformedtothestandardsystemusingobservationsof SillaequippedwiththeDFOSCinstrument,the2.56mNordic photometricstandardstars(Landolt,1992).Thelocalstandard OpticalTelescope(NOT)onLaPalmaequippedwithALFOSC stars are marked in Fig. 1, and their magnitudes are given in and STANCAM. We also used the 1.3 m MDM telescope (in Table 3. The zeropoint uncertainties are of the order of 0.03 August 2005) and the 2.4 m MDM telescope (in September), mag. the Crimean Astrophysical Observatory (CrAO) 2.6 m tele- In Fig. 2 we have plotted the R-band lightcurve ranging scopeandthe1.5mtelescopeattheMaidanakobservatory. from11minutesto104daysaftertheXRF. Thisincludesthe Late observations were also obtained at the ESO New early detections from the BOOTES telescope (open circles). TechnologyTelescope(NTT)atLaSillaandattheVeryLarge Thebestfitpower-lawhasaslopeofα = 0.59±0.01(F ∝ R ν Telescope (VLT) on Paranal, Chile. A single epoch near in- t−α)andisalsoindicatedinthefigure. frared(near-IR)KsimagewasobtainedusingtheVLT/ISAAC TheI-bandlightcurvefollowstheR-bandverywellduring instrument. the first day when we have observations in both bands. The The full journal of observations is given in Table 2. The slopeisconsistentwiththeR-bandlightcurve(α =0.51±0.06, I datawerereducedusingstandardtechniquesforde-biasingand whereasα =0.57±0.03forthesameperiod). R flat-fielding. 3.3.SpectralEnergyDistribution 2.2.Spectroscopy ThemultibandobservationsofXRF050824allowedustocon- Spectra of the source were obtained with the VLT at two structthespectralenergydistribution(SED)oftheburstatan epochs.A 2×1500s spectrumwasobtainedonAugust25.4, epochof∼0.4days. about0.4 days past the burst, when the afterglow had a mag- Theresultis presentedin Fig. 3 wherewe haveconverted nitude of R≈ 20.7. We used the FORS2 spectrograph with a the B, R, I and K band magnitudesinto AB magnitudes. The 300V grism, the GG375 order separation filter and a 1.0 arc- optical and near-IR magnitudes were corrected for Galactic secwideslitprovidingadispersionof13.3Åoverthespectral reddeningofE(B−V)=0.035mag(Schlegeletal.,1998)and regionfrom3800Åto8900Å. transformedtofluxdensitiesusingtheconversionfactorsgiven The following night, on August 26.3, when the afterglow byFukugitaetal.(1995)andAllen(2000),respectively.Given hadfadedbyonemagnitude,weobtainedanotherspectrumof that the multibandobservationswere notall performedat the 6×1500sexposuretime.Theinstrumentalsetupwasidentical sameepoch,theircorrespondingfluxeswererescaledusingthe tothatusedonthefirstnight. bestfitpowerlaw. 4 J.Sollermanetal.:XRF050824 A power-lawfit in the formF ∝ ν−β providesa tolerable Gorosabeletal.(2003),i.e.,L=0.5±0.2L intherestframe ν ⋆ fitfortheSED.Thespectralindexisβ=0.56±0.04,assuming Bband. negligibleextinction. The host is extended with a size of roughly 0.8 arcsec, Unfortunately,withtheavailabledatawecannotsaymuch which at a distance of 5.24 Gpc correspondsto a linear scale about the extinction. The SMC, LMC or Milky Way extinc- of ∼ 6 kpc. Finally,fromthe [O II] lines we can estimate the tionlawsgiveequallygoodfitstothedata(Fig.3).Fixingthe star formation rate (SFR). The flux of this line, corrected for redshiftat z = 0.83 (see Sect. 3.4), a free fit with an intrinsic Galacticextinction,correspondstoaSFRof1.8 M⊙ yr−1,fol- power-lawshapeoftheSEDandanSMClikeextinctioncurve lowingKennicutt(1998). fromPei(1992)impliesanextinctionof A = 0.4±0.2mag. Infact,usingtheextinctioncorrectedvalueforthefluxof V However,thiswouldgiveanunrealisticallyflatβ∼0spectrum. Hβ, and assuming a case B recombination ratio for Hα ver- Giventhesparsedatasettheonlythingwecanfirmlyconclude sus Hβ, we can also use this line to estimate that the SFR is is thatA is less than 0.5mag. A low value of the globalex- 1.8 M yr−1,againfollowingKennicutt(1998).Asusual,any V ⊙ tinction is also implied by the blue color of the host galaxy slit-losseswouldincreasethisnumber.Wenotethattheconsis- (Sect.3.5). tencyoftheHαand[OII]predictionsoftheSFRalsosupports thenotionoflowextinctioninthehostgalaxy. This SFR compares rather well with the estimate for the 3.4.TheSpectra hostofGRB000210mentionedabove,whichhassimilarprop- erties and an estimated SFR from the UV light of 2.1 ± The combined flux-calibrated spectrum is also included in 0.2 M yr−1(Gorosabeletal.,2003). Fig.3.Theslopeisconsistentwiththecontemporaryphotom- ⊙ The specific star formation rate for the host galaxy of etry. XRF050824isthusonly∼ 4 M yr−1(L/L⋆)−1.Thisisrather We determinedtheredshiftfromthefirstnight’sspectrum ⊙ low, but not exceptional, and falls well within the popula- (Fynboetal.,2005)usingemissionlinessuchas[OII]λ3727, tion of smallstar-formingblue galaxiesas shownin Fig. 2 of [OIII]λλ4959,5007andHβ.AsnotedbyFynboetal.(2005), Sollermanetal.(2005),(seealsoChristensenetal.,2004). wealsodetectabsorptionlinesfromMgIIatthisredshift.We Finally,wecanestimatethemetallicityofthegalaxyusing discuss this further below (Sect. 4.1). The lines are shown in the R23 technique (Pagel, 1979). Using the results presented Fig. 4 and the measured positions and fluxes of the lines are in Table 4 and applying E(B − V) = 0.035 mag, we derive givenin Table 4. Theredshiftisz = 0.828±0.005.Notethat log(R )=1.0. This is quite high, and indicates a metallicity thefluxesgiveninTable4arenotcorrectedforGalacticorhost 23 (just)belowsolar(seee.g.,Fig.5byKewley&Dopita,2002). galaxyextinction.Theuncertaintiesinabsolutelinefluxescan However, the small number of emission lines in the analysis beupto30%. makesthisestimateratheruncertain. Using a cosmology where H = 70kms−1Mpc−1, Ω = 0 Λ The star formation rate and size thus indicates a 0.7 and Ω = 0.3, this redshift corresponds to a luminosity m fairly normal galaxy, similar to other GRB host galax- distanceof5.24Gpc.Thisdistancewillbeusedhereafter. ies (e.g., LeFloc’hetal., 2003; Christensenetal., 2004; Sollermanetal.,2005).Themetallicityconfirmsthetrendthat 3.5.TheHostGalaxy GRB host galaxies have sub-solar metallicities. The luminos- ityisnotparticularlylowcomparedtootherGRBhosts,butis OurlateR-bandimagefromDecember7,2005,104dayspast similartothehostofXRF050416A(Soderbergetal.,2006). explosion, shows an extended source with magnitude R = 23.70±0.15 at the position of the afterglow.This is the host 3.6.TheX-rays galaxy of XRF050824. An image obtained under very good seeing conditionsatVLT in October2005is shownin Fig. 5. Swift-XRTdidnotobservetheburstimmediatelyduetoalunar Notethatalllateobservations(past35days)areconsistentwith constraintand the XRT beganobservationsabout6000s after this beingthe host, with little contributionfromthe afterglow thebursttrigger(Campanaetal.,2005b).Wehaveanalysedthe (orasupernova). standardprocessedXRTdatastartingat0.4daysaftertheburst The host magnitudeis 23.6 when correctedfor a Galactic usingversion2.3oftheSwiftsoftware.Background-subtracted extinction of E(B− V) = 0.035 mag. At the redshift of this spectraandlightcurveswere extractedina standardwaywith galaxy the R band corresponds to the rest frame U band. circularsourceandbackgroundextractionaperturesof30′′and However, comparison of the absolute luminosity with other 60′′radiusforthePC-modedata.DatafromtheWT-modewere galaxiesis often made in the rest frame B band. To do so we notusedbecausetheyaddedverylittlesignal. need to make some assumption about the color of the host. The combined spectrum was fit with a single absorbed Here we note that the BVR magnitudes from our latest VLT power-lawwithabsorptionattheGalacticlevel(N = 3.62× H data, when the host is clearly dominating the emission, are 1020cm−2, Dickey&Lockman,1990)andgaveanacceptable very similar to the magnitudes of the host of GRB000210 fit(χ2 = 43.7for40degreesoffreedom).Addinganabsorber (Gorosabeletal.,2003)atasimilarredshift. at the redshiftof the host galaxy gives a better fit (χ2 = 30.4 We therefore conclude that the absolute luminosity of the for39degreesoffreedom)withanequivalenthydrogencolumn XRF050824 host is very similar to the one determined by densityofN =1.8+0.7×1021cm−2andapower-lawphotonin- H −0.6 J.Sollermanetal.:XRF050824 5 dexβ =1.0±0.1.Theabsorptionmodelhasabundanceratios TheflatearlypartofthelightcurveofXRF030723wasin- X fixed at solar values. The soft X-ray absorption is dominated terpreted in terms of geometry(Fynboetal., 2004), where an byα-chainelementsandisthereforeameasureofthemetalab- off-axisorientationcanmakeanincreasinglylargefractionof sorptionandisregardlessofwhethertheelementsareinthegas the jet visible andthusmaintaina constant(or evenbrighten- orsolidphase(seeWatsonetal.,2006;Turnsheketal.,2003). ing) lightcurve (Granotetal., 2005). We see no evidence for The flux decay of the afterglow followed a single power- thisinXRF050824.It’searlylightcurveisconsistentwiththe law with decay index, α = 0.75 ± 0.04, with the fit being same decay seen throughout the lightcurve. This can thus be X marginally acceptable (χ2 = 24.8 for 16 degrees of freedom, seen asevidenceforanon-axisburst,whichwouldmeanthat null hypothesis probability = 0.07). This decay rate is some- ageometricalinterpretationdoesnotexplainthedifferencebe- whatfasterthantheaverageslopeseenintheopticallightcurve tween XRFs andGRBs in all cases (see also Soderbergetal., (albeit there could be a break in the optical light curve, see 2006). Sect. 4.2.2). Note, however, that the X-ray data are not very constrainingatthelaterphases. 4.2.2. ABreakandaBumpintheLightcurve 4. DISCUSSION At first glance the R-band lightcurveshown in Fig. 2 appears consistentwitha singlepower-lawdeclinethroughoutthe en- 4.1.Absorptionlineredshift tire afterglow. However, since the final points are due to the host galaxy,the data do suggest a break in the lightcurve.As Asnotedin Sect.3.4thespectraalsoincludeabsorptionlines mentioned in Sect. 3.5 the host is extended, so most of the fromMgII.Theselinesareseenatbothepochs,butaremost light at these epochs is indeed from the galaxy. This means clearlydetectedinthefirstepoch,whichhasthebestsignal-to- that the single power law must be broken at an early time, or noiseratio.ThelinesaredisplayedinFig.4,andtheredshiftis the later pointswould have been much brighter. The break in consistentwiththeestimatefromtheemissionlines. thelightcurvealsomeansthatanextracomponentisneededto That the redshift can be determined from both emission explaintheexcesslightat10−20dayspasttheburst. linesandabsorptionlinesisofsomeimportance.Thedistance We embarked on simultaneously fitting two power-laws scale of the XRFs has only recently been shown to be cos- andastretchableSN1998bwtemplate,inaccordancewiththe mological. The first spectroscopic redshift of z = 0.25 for method outlined by Zehetal. (2004). Fixing the host galaxy XRF020903 (Soderbergetal., 2005) was based on emission magnitude, we were able to constrain a shallow break in the linesonly.Inprinciple,asinglecasecouldbeaffectedbyasu- lightcurvetooccurat∼0.5dayspastburst(Fig.6).Thiscould perposedandunrelatedgalaxy,butnowXRF050416Aalsohas be a cooling break.The requiredsupernovais rather unusual. ameasuredemissionlineredshift(z=0.65,Cenkoetal.,2005; In the notation of Zehetal. (2004) this is a supernova with Soderbergetal., 2006) as has GRB/XRF030528 (z = 0.78, k= 1.05±0.42 and s= 0.52±0.14. This is a bright and fast Rauetal.,2005). lightcurve and is different from the lightcurve of the canoni- The GRB030429 studied by Jakobssonetal. (2004) also calSN1998bw,whichisoftenassociatedwithlongGRBs,but displayed absorption lines. It showed an E of 35 keV at a peak is similar to, althoughbrighterthan, the supernovaassociated redshift of z = 2.66, and is thus a borderline case, consistent withXRF060218. with an X-ray Rich burst. More recently, the rather unusual XRF060218 had a secure redshift from both emission lines The actual peak luminosity of the potential supernova is, and absorption lines (z = 0.033, Mirabal&Halpern, 2006; however,highlyuncertain.Ifthereisinternalextinctioninthe Wiersemaetal.,2007;Pianetal.,2006). host galaxy the corresponding supernova would have to be Thesefindings,togetherwiththerobustredshiftdetermina- brighter, but our SED analysis shows that this can not be a tionfortherathernormalXRF050824,havethereforeproven verylargeeffect.Thisisalsosupportedbytheratherbluecolor thecosmologicaldistancescalefortheseobjectsbeyonddoubt. of the host galaxy, and by the deduced balmer line ratios. A largeruncertaintyarisesfromtheassumptionsonthecontribu- tionfromtheafterglow.WithSN1998bwejecting∼0.5 M of ⊙ 4.2.TheLightcurveoftheAfterglow 56Ni(e.g.,Woosleyetal.,1999;Sollermanetal.,2000),wecan estimate that a supernovaassociated with XRF050824would OurR-bandlightcurveofXRF050824isoneofthebestsam- havehadtoejectatleastontheorderof0.6±0.3 M of56Ni. pledopticallightcurvesforanXRF.Themostconspicuousas- ⊙ This is assuming that the peak brightness scales with nickel- pect of this lightcurve is that it is basically consistent with a mass. power-law for the entire duration (Fig. 2). The best fit power lawα=0.6isquiteaslowdecay. 4.2.3. Wiggles,bumps,humpsandjitter 4.2.1. TheEarlyTimes Infact,thefittothelightcurveisnotverysatisfactoryevenaf- Oneof themoreinterestingaspectsofthe lightcurveis thatit terinvokingabreakandahypotheticalsupernovabump.Thisis declinessteadilyfromveryearlyon.Thisisinstarkcontrastto seeninFig.6,wherethereducedχ2is1.8.Theentirelightcurve thelightcurveofXRF030723(Fig.6),whichapparentlyhada of XRF050824 displays wiggles and humps throughout the constantlightcurveforthefirstdayaftertheburst. timeoftheobservations.Thelargestdeviationsarefromasys- 6 J.Sollermanetal.:XRF050824 tematicdipintheNOTdatarelativetotheoverallfitjustbefore jet model may be unable to account for these variabilities 0.1days,andanincreaseinthescatterfrom2–5days.These (Kumar&Piran,2000),whichmayinsteadbeattributedtothe deviationscannotbesatisfactorilyfittedbyabrokenpower-law re-activity of the central engine (e.g., Fanetal., 2005), or, as schemethatissupposedtomodelasimpleimpulsiveshockor mentionedabove,toasupernova(Sect.4.2.2). jet. Anotherexampleisat0.2–0.5dayspasttheburstwhenwe 4.3.Amatirelation haveawellmonitoredlightcurve,inparticularfromtheMDM. ThisisshownintheinsetinFig.2.Alineardecayisnotafor- Several XRFs have been shown to follow the same relation mallygoodfittothesedata.Thereappeartobewigglesaround as GRBs, that Epeak ∝ Ei1s/o2 (Amatietal., 2002), where Eiso thesteadylineardecline.Sincemostoftheseobservationswere is the isotropic-equivalent radiated energy. For XRF050824, obtained at the same telescope, and have been reduced in the wecannotdetermineaprecisetotalenergyduetothelackof samewayagainstthesamelocalstandards,wedonotbelieve knowledgeconcerningthe peak energyof the BAT spectrum. this is purely an instrumental effect. Although the statistical However,wecancalculatelowerandupperlimitsbyintegrat- significanceisratherlowinourlightcurve,wenotethatsimilar ing the best fit power law spectral energy distributions in the jittering has been observed previously in GRBs, both in long (15−150)×(1+z)keVbandandinthethefull1−104 keV GRBs(e.g.,Gorosabeletal.,2006;Mathesonetal.,2003)and band.Weobtained4.1×1050erg< Eiso <3.4×1051erg,which in short GRBs (Hjorthetal., 2007), and can be interpreted in whenusing the Amatirelation,wouldprovidea constrainton termsofvariationsinthesurroundingcircumburstmediumor the observed peak energy 11keV < Eobs < 32keV. This is peak asdue to prolongedactivityof thecentralengine.Similar ex- thusonlyjustin agreementwiththatfromthe spectralfitting, planationscanthusbeputforwardalsoforthisXRF. Eobs <13keV. peak Besides this event, the Amati relation is also applicable to XRFs020903 and possibly 030723 4.2.4. Continuousenergyinjection (Lamb,Donaghy&Graziani, 2005), XRF050416A (Sakamotoetal., 2005) and XRF050406 (Schadyetal., Prolonged central engine activity with multiple energy injec- 2006).(SeealsoAmatietal.,2006;Ghisellinietal.,2006,for tions could thus be the explanation for the deviations from a XRF060218.) perfectpower-law(Bjo¨rnssonetal., 2004). Prolongedactivity ThatthisrelationholdsforbothXRFsandclassicalGRBs intermsofcontinuousenergyinjectioncouldalsoexplainthe not only implies that both classes of bursts can be on-axis slow decline rates in both optical (α = 0.65 when corrected R events(Amatietal.,2006)butalsosupportstheideathatboth for host galaxy contribution) and in X-rays (α = 0.75). In X canbeinterpretedunderaunifiedphysicalmechanism. Fig.7weshowanexampleofamodelfortheafterglowemis- sioninthe X-rayandtheR bandforcontinuousenergyinjec- tion.Thismodelisdetailedbelow. 5. CONCLUSIONS Wehavemodeledtheafterglowintermsofalong-termcon- We have seen that the afterglows of XRFs can appear quite tinual energy injection in the forward shock. We consider an different.TheearlyflatopticallightcurveofXRF030723was uniform relativistic jet undergoing the energy injection from consistent with predictions of an off-axis burst (Fynboetal., the central source and sweeping up its surrounding uniform 2004; Granotetal., 2005). On the contrary, XRF050824 dis- medium.Thedynamicalevolutionoftheoutflowiscalculated plays an optical lightcurve which is decaying at a fairly con- usingtheformulaeinHuangetal.(2000)andaddinganenergy stant, but slow, rate from 10 minutes after the burst. Our injectionprocesswiththeformdE /dt= A(t/t )−qfort <t< inj 0 0 afterglow model indicates this to be an on-axis burst. We t .Thefractionsofshockenergygiventotheelectronsǫ and end e have also found some evidence for a bump in the lightcurve, tothemagneticfieldǫ areassumedtobeconstant. B whichisconsistentwithasupernovaasfastasthatassociated The modelfits shown in Fig. 7 have the followingjet pa- withXRF060218,i.e,considerablyfasterthantheSN1998bw rameters:theisotropickineticenergyEk = 1052erg,ǫe = 0.4, lightcurve. ǫB = 0.003,thecircumburstdensityn = 0.1cm−3,theelectron Most well observed XRFs with a redshift where a super- index p= 2.05,thehalf-openingangleθj = 0.2,andtheview- nova could be expected to emerge do show some evidence ingangleθobs=0(i.e.,on-beamviewing),togetherwiththeen- for this. This is similar to the case for ordinary long GRBs ergyinjectionparameters:A=3×1049erg/s,q=0.8,t0 =100 (Zehetal., 2004). A common origin for XRFs and GRBs is s, and tend = 2 × 106 s. This rather large amount of ejected therefore likely but there also seems to be several parame- energyis needed to explain the long and shallow decline;the ters affecting the observables of the burst. In the context of amountissimilartothatfoundforGRB050315(Zhangetal., the four-field diagram presented by Sollermanetal. (2006), 2006). Since there is no proper jet break until possibly after XRF050824 should be in the same category as XRF020903 a week, the constrainton the jet opening angle of θj∼>10◦ is and XRF030723; XRFs with an associated supernova but quiterobust.Wedidnotattempttofittheveryearlylightcurve. where the opticallight is dominatedby the afterglowat early At these phases it is likely that a reverse shock componentis phases. required. ThemountingevidenceforsupernovaeinGRBsandXRFs We note again the late re-brightening. At such a late also shows that there is a large variety in supernova prop- time, the ejecta is only moderately relativistic. The patchy erties (Woosley&Bloom, 2006). The emergence of super- J.Sollermanetal.:XRF050824 7 novae much fainter (Pianetal., 2006; Sollermanetal., 2006) and much faster (this work) than the canonical SN1998bw put constraints on the underlying explosion model. Recently, Fynboetal. (2006) also reported two GRBs where no su- pernova emission is seen, down to 100 times fainter than SN1998bw.It is still not clear whether we see two (or more) fundamentallydifferentexplosionmechanisms,orifthereisa verywidecontinuumofpropertiesfortheseblasts. Acknowledgements. This paper is based on observations collected by the Gamma-Ray Burst Collaboration at ESO (GRACE) at the EuropeanSouthernObservatory,ParanalandLaSilla,Chile.Wethank theESOstafffortheirhelpinsecuringtheservicemodedatareported here.WeacknowledgebenefitsfromcollaborationwithintheEUFP5 Research TrainingNetwork”Gamma-Ray Bursts: AnEnigmaanda Tool”. This work was also made at the DARK Cosmology Centre funded by The Danish National Research Foundation. JS acknowl- edges support from Danmarks Nationalbank, from the Anna-Greta andHolgerCrafoordfundandfromKnut&AliceWallenbergfoun- dation. CrAO and IKI acknowledge support from Russian Ministry of Education and Science. JPH and NM acknowledge support from theNationalScienceFoundationunder grant0206051. Theresearch activities of JG and MJ are supported by the Spanish Ministry of SciencethroughprojectsAYA2004-01515andESP2005-07714-C03- 03. Based in part on observations made with the BOOTES instru- ments in South Spain (ESAt-INTA/CEDEA, Huelva) and with the 1.5m Telescope of Observatorio de Sierra Nevada (OSN), operated byIAA/CSIC.Someofthedatapresentedherehavebeentakenusing ALFOSC,whichisownedbytheInstitutodeAstrof´ısicadeAndaluc´ıa (IAA) and operated at the Nordic Optical Telescope under agree- mentbetweenIAAandtheDepertmentofAstronomyofCopenhagen University. Thanks to Tamara Davis for very careful reading of the manuscript,andfordetaileddiscussionsonK-corrections.Lastbutnot least, we acknowledge support from several key observers that con- tributedtothiswork,namelyV.BirykovandD.Sharapovaswellas D.R.Rafferty&J.R.Thorstensen.FinallythelateHugoE.Schwarz contributedwithobservationsandcommentsonanearlierversionof thispaper. 8 J.Sollermanetal.:XRF050824 References Hjorth, J., Sollerman, J., Møller, P., et al. 2003, Nature, 423, 847 Allen, C. W. 2000, Allen’s AstrophysicalQuantities, 4th edi- Hjorth,J.,etal.2007,inprep. tion2000,ed.A.N.Cox Huang,Y. F., Gou,L. J., Dai, Z. G. & Lu,T. 2000,ApJ, 543, Amati,L.,Frontera,F.,Tavani,M.,etal.2002,A&A,390,81 90 Amati,L.,DellaValle,M.,Frontera,F.,Malesani,D.,Guidorzi, Jakobsson,P.,Hjorth,J.,Fynbo,J.P.U.2004,A&A,425,785 C.,Montanari,E.,&Pian,E.2006,astro-ph/0607148 Kennicutt,R.C.1998,ARA&A,36,189 Assafin, M., Andrei,A. H., Martins,R., Vieira da Silva Neto, Kewley,L.J.,&Dopita,M.A.2002,ApJS,142,35 D.N.,Camargo,J.I.B.,Teixeira,R.&Benevides-Soares,P. Kippen, R. M., Woods, P. M., Heise, J., in’t Zand, J. J. M., 2001,ApJ,552,380 Briggs, M. S., & Preece,R. D. 2003,AIP Conf.Proc.662: Band,D.,Matteson,J.,Ford,L.,etal.1993,ApJ,413,281 Gamma-Ray Burst and Afterglow Astronomy 2001: A Barraud,C.,Olive,J.-F.,Lestrade,J.P.,etal.2003,A&A,400, WorkshopCelebratingtheFirstYearoftheHETEMission, 1021 662,244 Barraud, C., Daigne, F., Mochkovitch, R., Atteia, J. L. 2005, Krimm,H.,Barbier,L.,Barthelmy,S.,etal.2005,GCN,3871 A&A,440,809 Kumar,P.&Piran,T.2000,ApJ,535,152 Bersier, D., Fruchter, A. S., Strogler, L.-G., et al. 2006, ApJ, Landolt,A.U.1992,AJ,104,340 643,284 Lamb,D.Q.,Donaghy,T.Q.,&Graziani,C.2005,ApJ,620, Bjo¨rnsson, G., Gudmundsson, E. H., & Johannesson, E. H. 355L 2004,ApJ,615,L77 LeFloch,E.,Duc,P.-A.,Mirabel,I.F.,etal.2003,A&A,400, Butler,N.R.,Sakamoto,T.,Suzuki,M.,etal.2005,ApJ,621, 499 884 Levan,A.,Patel,S.,Kouveliotou,C.,etal.2005,ApJ,622,977 Campana, S., Barthelmy,S., Burrows, D., et al. 2005a,GCN, Matheson,T.,Garnavich,P.M.,Stanek,K.Z.,etal.2003,ApJ, 3866 599,394 Campana,S.,Mangano,V.,Chincarini,G.,etal.2005b,GCN, Mirabal,N.,&Halpern,J.P.2006,GCN4792 3872 Mirabal,N.&Halpern,J.P,An,D.,etal.2006,ApJ,643,99 Campana,S.,Mangano,V.,Blustin,A.J.,etal.2006,Nature, Modjaz,Stanek,K.Z.,Garnavich,P.M.,etal.2006,ApJ,645, 442,1008 21 Castro-Tirado,A.J.,Jelinek,M.,MateoSanguino,T.J.,etal. Oke,J.B.1990,AJ,99,1621 2004,AstronomischeNachrichten,325,679 Pagel, B. E. J., Edmunds,M. G., Blackwell, D. E., Chun, M. Cenko,S.B.,Kulkarni,S.R.,Gal-Yam,A.,&Berger,E.2005, S.,&Smith,G.1979,MNRAS,189,95 GCN3542 Pei,Y.C.1992,ApJ,395,130 Christensen, L., Hjorth, J., & Gorosabel, J. 2004, A&A, 425, Pian,E.,Mazzali,P.,Masetti,N.,etal.2006,Nature,442,1011 913 Preece,R.D.,Mallozzi,R.S.,Pendleton,G.N.,&PaciesasW. Crew,G.,Ricker,G.,Atteia,J-L.,etal.2005,GCN,3890 S.2000,ApJS,126,19 Dermer,C.D.,Chiang,J.,&Bo¨ttcher,M.1999,ApJ,513,656 Rau,A.,Salvato,M.,&Greiner,J.2005,A&A,444,425 Deutsch,E.W.1999,AJ,118,1882 Rhoads,J.E.2003,ApJ,591,1097 Dickey,J.M.,&Lockman,F.J.1990,ARA&A,28,215 Sakamoto,T., Lamb,D. Q., Kawai,N., etal. 2005,ApJ, 629, Fan,Y.Z.,&Wei,D.M.2005,MNRAS,364,L42 311 Fukugita,M.,Shimasaku,K.,&Ichikawa,T.1995,PASP,107, Schady,P.,Mason,K.O.,Osborne,J.P.,etal.2006,ApJ,643, 945 276 Fynbo,J.,Sollerman,J.,Hjorth,J.,etal.2004,ApJ,609,962 Schaefer,B.E.,Yuang,F.,&Alatalo,K.2005,GCN,3867 Fynbo,J.P.U.,Jensen,B.L.,Sollerman,J.,etal.2005,GCN, Schlegel,D.J.,Finkbeiner,D.P.,&Davis,M.1998,ApJ,500, 3874 525. Fynbo, J. P. U., Watson, D., Thoene, C., et al. 2006, Nature, Soderberg,A.M.,Kulkarni,S.R.,Fox,D.B.,etal.2005,ApJ, 444,1047 627,877 Ghisellini, G., Ghirlanda, G., Mereghetti, S., Bosnjak, Z., Soderberg, A. M., Nakar, E., Cenko, S. B., et al. 2006, Tavecchio,F.,&Firmani,C.2006,MNRAS,372,1699 astro-ph/0607511 Gorosabel,J.,Casanova,V., Garrido,R., Castro-Tirado,A. J., Sollerman, J., Kozma, C., Fransson, C., Leibundgut, B., Jelinek,M.,&deUgartePostigo,A.2005,GCN,3865 Lundqvist,P.,Ryde,F.,&Woudt,P.2000,ApJ,537,L127 Gorosabel, J., Christensen, L., Hjorth, J., et al. 2003, A&A, Sollerman,J., O¨stlin, G., Fynbo,J. P. U., Hjorth, J., Fruchter, 400,127 A.,&Pedersen,K.2005,NewAstronomy,11,103 Gorosabel, J., Castro-Tirado, A. J., Ramirez-Ruiz, E., et al. Sollerman,J.,Jaunsen,A.O.,Fynbo,J.P.U.,etal.2006,A&A, 2006,ApJ,641,L13. 454,503 Granot,J.,Ramirez-Ruiz,E.,&Perna,R.2005,ApJ,630,1003 Tominaga, N., Deng, J., Mazzali, P. A., Maeda, K., Nomoto, Heise, J., in’tZand,J., Kippen,R. M., & Woods, P. M. 2001, K., Pian, E., Hjorth, J., & Fynbo, J. P. U. 2004, ApJ, 612, in the proc. of the conference, Gamma-ray Bursts in the L105 AfterglowEra,16 Turnshek, D. A., Rao, S. M., Ptak, A. F., Griffiths, R. E., & Monier,E.M.2003,ApJ,590,730 J.Sollermanetal.:XRF050824 9 Watson,D.,Fynbo,J.P.U.,Ledoux,C.,etal.2006,ApJ,652, 1011 Wiersema, K., Savaglio, S., Vreeswijk, P. M., et al., 2007, A&A,inpress,astro-ph/0701034 Woosley, S. E., Eastman, R. G., & Schmidt, B. P. 1999,ApJ, 516,788 Woosley,S.E.,&Bloom,J.2006,ARA&A,44,507 Yamazaki,R.,Ioka,K.,&Nakamura,T.2002,ApJ,571,L31 Yamazaki,R.,Ioka,K.,&Nakamura,T.2003,ApJ,593,941 Zeh,A.,Klose,S.,&Hartmann,D.H.2004,ApJ,609,952 Zhang, B., Fan, Y. Z., Dyks, J., Kobayashi, S., Me´sza´ros, P., Burrows,D.N.,Nousek,J.A.&Gehrels,N.2006,ApJ,642, 354 10 J.Sollermanetal.:XRF050824 Table2.LogofobservationsandphotometryoftheafterglowofXRF050824. Date ∆t Magnitude MagnitudeError PassBand Telescope (UT) (days) (1σ) Aug25.0973 0.130449 20.72 0.04 B-band NOT Aug25.1017 0.134850 20.72 0.05 B-band NOT Aug25.1061 0.139250 20.65 0.04 B-band NOT Oct6.1943 42.2274 24.43 0.16 B-band VLT Oct6.1898 42.2229 24.26 0.17 V-band VLT Aug24.9742 0.007346 18.22 0.35 R-band BOOTES Aug24.9813 0.014429 19.11 0.32 R-band BOOTES Aug24.9935 0.0266991 18.94 0.03 R-band OSN Aug24.9971 0.0302391 19.04 0.04 R-band OSN Aug25.0007 0.0338001 19.07 0.03 R-band OSN Aug25.0096 0.042705 19.67 0.33 R-band BOOTES Aug25.0166 0.0496998 19.31 0.03 R-band OSN Aug25.0337 0.0668297 19.56 0.05 R-band OSN Aug25.0372 0.0703697 19.59 0.07 R-band OSN Aug25.0546 0.0877495 19.80 0.04 R-band NOT Aug25.0609 0.0940495 19.85 0.05 R-band NOT Aug25.0653 0.0984497 19.93 0.04 R-band NOT Aug25.0881 0.121212 19.33 0.20 R-band BOOTES Aug25.1496 0.182749 20.15 0.04 R-band OSN Aug25.1599 0.193050 20.21 0.05 R-band OSN Aug25.1700 0.203150 20.23 0.07 R-band OSN Aug25.1813 0.214449 20.38 0.08 R-band OSN Aug25.2003 0.233429 20.42 0.10 R-band MDM Aug25.2042 0.237391 20.22 0.08 R-band MDM Aug25.2082 0.241360 20.42 0.10 R-band MDM Aug25.2122 0.245300 20.34 0.09 R-band MDM Aug25.2161 0.249279 20.39 0.09 R-band MDM Aug25.2201 0.253250 20.47 0.11 R-band MDM Aug25.2241 0.257210 20.51 0.12 R-band MDM Aug25.2251 0.258249 20.25 0.05 R-band D1.5m Aug25.2280 0.261179 20.39 0.11 R-band MDM Aug25.2303 0.263451 20.47 0.06 R-band D1.5m Aug25.2320 0.265150 20.35 0.11 R-band MDM Aug25.2360 0.269110 20.24 0.10 R-band MDM Aug25.2402 0.273399 20.35 0.07 R-band MDM Aug25.2477 0.280849 20.30 0.07 R-band MDM Aug25.2551 0.288280 20.49 0.08 R-band MDM Aug25.2626 0.295719 20.35 0.08 R-band MDM Aug25.2700 0.303150 20.39 0.07 R-band MDM Aug25.2783 0.311409 20.62 0.11 R-band MDM Aug25.2966 0.329741 20.43 0.05 R-band MDM Aug25.3040 0.337179 20.55 0.08 R-band MDM Aug25.3062 0.339350 20.64 0.09 R-band D1.5m Aug25.3111 0.344250 20.69 0.10 R-band D1.5m Aug25.3115 0.344610 20.51 0.08 R-band MDM Aug25.3288 0.361910 20.49 0.06 R-band MDM Aug25.3362 0.369339 20.62 0.07 R-band MDM Aug25.3436 0.376780 20.55 0.06 R-band MDM Aug25.3511 0.384220 20.60 0.07 R-band MDM Aug25.3587 0.391870 20.58 0.06 R-band MDM Aug25.3632 0.396349 20.55 0.05 R-band VLT Aug25.3661 0.399300 20.60 0.06 R-band MDM Aug25.3736 0.406740 20.60 0.06 R-band MDM

ebook img

The nature of the X-Ray Flash of August 24 2005 PDF

0.55 MB·English
by  J. Sollerman
#additional_collections #journals #arxiv
Save to my drive
Quick download
Download
Upgrade Premium
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview The nature of the X-Ray Flash of August 24 2005

Astronomy&Astrophysicsmanuscriptno.final (cid:13)c ESO2008 February5,2008 The nature of the X-Ray Flash of August 24 2005. Photometric evidence for an on-axis z = 0.83 burst with continuous energy 7 ⋆ injection and an associated supernova? 0 0 2 J.Sollerman1,2,J.P.U.Fynbo1,J.Gorosabel3,J.P.Halpern4,J.Hjorth1,P.Jakobsson1,5,N.Mirabal4,D.Watson1, n D.Xu1,A.J.Castro-Tirado3,C.Fe´ron1,A.O.Jaunsen6,M.Jel´ınek3,B.L.Jensen1,D.A.Kann7,J.E.Ovaldsen6, a J A.Pozanenko8,M.Stritzinger1,C.C.Tho¨ne1,A.deUgartePostigo3,S.Guziy3,M.Ibrahimov9,S.P.Ja¨rvinen10,11, 5 A.Levan5,V.Rumyantsev12,andN.Tanvir5,13 2 1 DarkCosmologyCentre,NielsBohrInstitute,UniversityofCopenhagen,JulianeMariesVej30,DK–2100CopenhagenØ,Denmark 1 v 2 StockholmObservatory,DepartmentofAstronomy,AlbaNova,S-10691Stockholm,Sweden 6 3 InstitutodeAstrof´ısicadeAndaluc´ıa(IAA-CSIC),P.O.Box3.004,E-18.080Granada,Spain 3 4 ColumbiaAstrophysicsLaboratory,550West120thStreet,ColumbiaUniversity,NewYork,NY10027-6601,USA 7 5 CentreforAstrophysicsResearch,UniversityofHertfordshire,CollageLane,Hatfield,Herts,AL109AB,UK 1 6 InstituteofTheoreticalAstrophysics,POBox1029,N-0315Oslo,Norway 0 7 Thu¨ringerLandessternwarteTautenburg,Sternwarte5,D–07778Tautenburg,Germany 7 8 SpaceResearchInstitute(IKI),84/32ProfsoyuznayaStr,Moscow117997,Russia 0 9 UlughBegAstronomicalInstitute,Tashkent700052,Uzbekistan / h 10 AstrophysikalischesInstitutPotsdam,AnderSternwarte16,D-14482Potsdam,Germany p 11 AstronomyDivision,P.O.Box3000,FIN-90014UniversityofOulu,Finland - 12 CrimeanLaboratoryofSternbergAstronomicalInstituteMSU,Nauchny,Crimea,98409,Ukraine o 13 DepartmentofPhysicsandAstronomy,UniversityofLeicester,Leicester,LE17RH,UK r t s a : ABSTRACT v i X Aims.OuraimistoinvestigatethenatureoftheX-RayFlash(XRF)ofAugust24,2005. r Methods.WepresentcomprehensivephotometricR-bandobservationsofthefadingopticalafterglowofXRF050824,from11minutesto104 a daysaftertheburst.InadditionwepresentobservationstakenduringthefirstdayintheBRIKbandsandtwoepochsofspectroscopy.Wealso analyseavailableX-raydata. Results. TheR-bandlightcurveof theafterglow resemblesthelightcurvesof longduration Gamma-RayBursts(GRBs),i.e.,apower-law, albeitwitharathershallowslopeofα = 0.6(F ∝ t−α).OurlateR-bandimagesrevealthehostgalaxy.Therest-frameB-bandluminosityis ν ∼0.5L∗.Thestar-formationrateasdeterminedfromthe[OII]emissionlineis∼1.8M yr−1.Whenaccountingforthehostcontribution,the ⊙ slopeisα=0.65±0.01andabreakinthelightcurveissuggested.Apotentiallightcurvebumpat2weekscanbeinterpretedasasupernova onlyifthisisasupernovawithafastriseandafastdecay.However,theoverallfitstillshowsexcessscatterinthelightcurveintheformof wigglesandbumps.TheflatlightcurvesintheopticalandX-rayscouldbeexplainedbyacontinuousenergyinjectionscenario,withanon-axis viewingangleandawidejetopeningangle(θj∼>10◦).Iftheenergyinjectionsareepisodicthiscouldpotentiallyhelpexplainthebumpsand wiggles. Spectroscopyoftheafterglowgivesaredshiftofz=0.828±0.005frombothabsorptionandemissionlines.Thespectralenergydistribution (SED)oftheafterglowhasapower-law(F ∝ν−β)shapewithslopeβ=0.56±0.04.ThiscanbecomparedtotheX-rayspectralindexwhich ν isβ =1.0±0.1.ThecurvatureoftheSEDconstrainsthedustreddeningtowardsthebursttoA <0.5mag. X v Keywords. cosmology:observations—gammarays:bursts— ⋆ This paper is based on observations from a multitude of tele- scopes,forexampleonobservationsmadewithESOTelescopesatthe ParanalObservatory(programmeID075.D-0270)andwiththeNTT atedontheislandofLaPalmajointlybyDenmark,Finland,Iceland, and ESO/Danish 1.5-m telescope at the La Silla Observatory. Also Norway, andSweden, intheSpanishObservatorio delRoque delos basedonobservationsmadewiththeNordicOpticalTelescope,oper- MuchachosoftheInstitutodeAstrofisicadeCanarias. 2 J.Sollermanetal.:XRF050824 1. INTRODUCTION 2006; Wiersemaetal., 2007) andunusualburst, XRF060218, with a very low E ∼ 5 keV (Campanaetal., peak X-Ray Flashes (XRFs) are similar to Gamma-Ray Bursts 2006) showed unambiguous evidence for an associated su- (GRBs), but with most of the fluence of the prompt emis- pernova (e.g., Sollermanetal., 2006; Modjazetal., 2006; sion detected in the X-ray band. Their existence as a class Mirabaletal., 2006; Pianetal., 2006). SN2006aj associated was suggested by Heiseetal. (2001) based on data from the with XRF060218 clearly established the close link between BeppoSAXsatellite. SNe,(GRBs)andXRFs. The high energy spectra (νF ) of the prompt emission ν However, other XRFs with late time coverage did ap- fromGRBsarewelldescribedbytheso-calledBandfunction parently not show any clear evidence for a supernova bump (Bandetal., 1993), which is composedof two smoothly con- (Soderbergetal., 2005; Levanetal., 2005). This could point nected power-laws. The energy at which the two power-laws to a differencein theoriginofXRFs andthe longGRBs with connect, E , is where most of the energy is emitted. For peak accompanying SNe (but see the recent paper by Fynboetal., classicalGRBs,E istypicallyafew100keV(Preeceetal., peak 2006, for GRBs with no associated supernova emission), al- 2000). The spectra of the prompt emission of XRFs are also thoughanysuchclaimhastoawaitobservationsofXRFswith wellfittedbytheBandfunction,butwithvaluesofE below peak a betterdetermineddistancescale.Mostofthe XRFsclaimed .50keVandinsomecasesevenbelow10keV(Kippenetal., to lack SNe in the above-mentionedstudies had no measured 2003;Barraudetal.,2003). redshifts. Sakamotoetal. (2005) argue, in accordance with previ- We here present data for XRF050824 for which we have ous studies, that the spectral distributions of XRFs, X-ray obtainedawellmonitoredopticallightcurveandasecurespec- RichGRBsandGRBsformacontinuum,suggestingthatthey troscopicredshift. all arise from the same phenomenon (see also Barraudetal., 2003, 2005). There has also been growing evidence that (at least some) XRFs are the result of classical GRBs seen off- 1.1.XRF050824 axis(Yamazakietal.,2002,2003;Rhoads,2003;Fynboetal., XRF050824 was detected by the BAT instrument on-board 2004;Granotetal.,2005). theSwiftsatellite(Campanaetal.,2005a)onAugust24.9669 Other suggestions include XRFs as either dirty fireballs, 2005UT (UniversalTimeisusedthroughoutthearticle).The which are relativistic jets with a larger baryon load and burst duration (T ) was 25 s, i.e., this was a long-duration hence (assuming external shocks) lower Γ-factors than those burst.Thetotalflu9e0nceinthe15–150keVbandwas∼2.3×10−7 ofclassical GRBs (Dermeretal., 1999; Heiseetal., 2001), or ergs cm−2 (Krimmetal., 2005). The burst was also detected XRFs as clean fireballs where (assuming internal shocks) the by the FREGATE instrument onboard HETE-2 (Crewetal., spread in Γ-factors is small but the average Γ-factor is large 2005). As seen by HETE-2 the value of the fluence ratio (Barraudetal.,2005). S(2−30)/S(30−400)=2.7.Crewetal.(2005)estimatedE < peak However, there are still open questions regarding the ori- 12.7keV.GRB050824isthusclearlyanXRF. gin of XRFs. While the connection between long GRBs and The BAT on-board localization was reported to an accu- certain core-collapse supernovae appears to be well estab- racyof3arcminutes.EarlyROTSE-IIIdatadidnotdetectany lished (Hjorthetal., 2003; Mathesonetal., 2003; Zehetal., new object (Schaeferetal., 2005), but our observations start- 2004), the case has not been as well defined for XRFs. ing38minutesafterthetriggerrevealedanewobjectatR.A.= Fynboetal.(2004)performedthefirstcomprehensiveobserva- 00:48:56.1,Dec=+22:36:32(J2000,seeSect.3.1),whichwe tionalcampaignofanXRFopticalafterglow,forXRF030723 laterconfirmedasthecounterpart(Gorosabeletal.,2005). (see also Butleretal., 2005). The wellsampled lightcurvefor Inthispaperwefocusonourcomprehensiveopticalstudy XRF030723displayedseveralinterestingfeatures: of this afterglow. The article is organized as follows. Sect. 2 (i) Theveryearlylightcurvewasessentiallyflat, inaccor- outlines how the optical observations were obtained and re- dancewithmodelsforwhichXRFsareviewedawayfromthe duced.TheresultsarepresentedinSect.3,whichincludesthe jetaxis. astrometry,theopticallightcurve,thespectralenergydistribu- (ii)Thefollowingpower-lawdecaywasverysimilartothat tion, the spectroscopicresults, data on the hostgalaxyand an seenintypicalGRBs,suggestingacommonorigin. analysis of available X-ray data of the afterglow. Finally, we (iii)At∼16dayspastburstastrongbumpinthelightcurve endthepaperwithadiscussion(Sect.4)andsomeconclusions suggested the presence of a fast rising supernova. The super- (Sect.5). novainterpretationwasarguedto beconsistentwiththespec- tral energy distribution (SED) evolution (Fynboetal., 2004) and was later supported by modeling both of the afterglow 2. OBSERVATIONS (Granotetal., 2005) and of the supernova (Tominagaetal., 2.1.Photometry 2004). More evidence has now been presented arguing for a Our very first observations of XRF050824 were obtained common origin for GRBs and XRFs. XRF020903 had a with the BOOTES-1B 30 cm robotic telescope (e.g., late lightcurve and spectrum consistent with a supernova Castro-Tiradoetal., 2004) in southern Spain, which detected at z = 0.25 (Soderbergetal., 2005; Bersieretal., 2006). the burst from 10.6 minutes after the high energy event in More recently, the nearby (z = 0.0331, Mirabal&Halpern, several R-band exposures. However, since these images were J.Sollermanetal.:XRF050824 3 Table1.TelescopesandInstruments We extracted the spectrum using standard procedures within IRAF. Wavelength calibration was obtained using im- Telescope Instrument/ FOV Pixelscale ages of HeNeAr lamps obtained as part of the morning cali- CCD (arcminutes) (arcsecpixel−1) brations. Flux calibration was performedusing spectra of the BOOTES-1B ST8E 40×26 1.6 spectrophotometricstandardstarG138-31(Oke1990). OSN ROPER 7.92×7.92 0.232 NOT ALFOSC 6.3×6.3 0.189 NOT STANCAM 3×3 0.176 3. RESULTS D1.5m DFOSC 13.7×13.7 0.395 MDM1.3m SITeCCD 8.6×8.6 0.508 3.1.Astrometry MDM2.4m SITeCCD 9.4×9.4 0.275 CrAO2.6 FLI-IMG1001E 8.5×8.5 0.5 WedeterminedthecelestialpositionoftheXRF050824optical Maidanak1.5m SITeCCD 8.5×3.5 0.266 afterglowasthemeanastrometricsolutionfoundin10OSNR- WHT ULTRACAM 5×5 0.3 bandimages.Eachafterglowpositionisbasedon∼50USNO- NTT EMMI 9.9×9.1 0.33 A2.0referencestarsperimage.Themeanvalueofthecoordi- VLT FORS1 6.8×6.8 0.20 natesare: VLT FORS2 6.8×6.8 0.25 R.A.(J2000)=00:48:56.14±0.03s, VLT ISAAC 2.5×2.5 0.148 Dec(J2000)=+22:36:33.2±0.4′′ These astrometric errors include the 0.25′′ systematic error of the USNO-A2.0 catalogue (Assafinetal., 2001; Deutsch, not processed until later, the actual discovery was instead 1999). made via the studies we initiated 38 minutes after the burst (Gorosabeletal., 2005) using the 1.5 meter telescope at the 3.2.TheLightcurve ObservatoriodeSierraNevada(OSN) We conducteda comprehensivestudy of the optical after- The photometryof the XRF was carried outusing either PSF glowoverthefollowing100daysusingseveraltelescopesand fittingphotometry(whentheafterglowwasbright)oraperture instruments. In Table 1 we summarize the telescopes and in- photometry(at later times when the host started to contribute strumentsusedandprovidedetailsonthefield-of-view(FOV) significantly). We measured the magnitudes of the optical af- andpixelscaleoftheseinstruments. terglowaswellas4starsinthefield.Therelativemagnitudes WeusedtheESO/Danish1.54mtelescope(D1.5m)onLa weretransformedtothestandardsystemusingobservationsof SillaequippedwiththeDFOSCinstrument,the2.56mNordic photometricstandardstars(Landolt,1992).Thelocalstandard OpticalTelescope(NOT)onLaPalmaequippedwithALFOSC stars are marked in Fig. 1, and their magnitudes are given in and STANCAM. We also used the 1.3 m MDM telescope (in Table 3. The zeropoint uncertainties are of the order of 0.03 August 2005) and the 2.4 m MDM telescope (in September), mag. the Crimean Astrophysical Observatory (CrAO) 2.6 m tele- In Fig. 2 we have plotted the R-band lightcurve ranging scopeandthe1.5mtelescopeattheMaidanakobservatory. from11minutesto104daysaftertheXRF. Thisincludesthe Late observations were also obtained at the ESO New early detections from the BOOTES telescope (open circles). TechnologyTelescope(NTT)atLaSillaandattheVeryLarge Thebestfitpower-lawhasaslopeofα = 0.59±0.01(F ∝ R ν Telescope (VLT) on Paranal, Chile. A single epoch near in- t−α)andisalsoindicatedinthefigure. frared(near-IR)KsimagewasobtainedusingtheVLT/ISAAC TheI-bandlightcurvefollowstheR-bandverywellduring instrument. the first day when we have observations in both bands. The The full journal of observations is given in Table 2. The slopeisconsistentwiththeR-bandlightcurve(α =0.51±0.06, I datawerereducedusingstandardtechniquesforde-biasingand whereasα =0.57±0.03forthesameperiod). R flat-fielding. 3.3.SpectralEnergyDistribution 2.2.Spectroscopy ThemultibandobservationsofXRF050824allowedustocon- Spectra of the source were obtained with the VLT at two structthespectralenergydistribution(SED)oftheburstatan epochs.A 2×1500s spectrumwasobtainedonAugust25.4, epochof∼0.4days. about0.4 days past the burst, when the afterglow had a mag- Theresultis presentedin Fig. 3 wherewe haveconverted nitude of R≈ 20.7. We used the FORS2 spectrograph with a the B, R, I and K band magnitudesinto AB magnitudes. The 300V grism, the GG375 order separation filter and a 1.0 arc- optical and near-IR magnitudes were corrected for Galactic secwideslitprovidingadispersionof13.3Åoverthespectral reddeningofE(B−V)=0.035mag(Schlegeletal.,1998)and regionfrom3800Åto8900Å. transformedtofluxdensitiesusingtheconversionfactorsgiven The following night, on August 26.3, when the afterglow byFukugitaetal.(1995)andAllen(2000),respectively.Given hadfadedbyonemagnitude,weobtainedanotherspectrumof that the multibandobservationswere notall performedat the 6×1500sexposuretime.Theinstrumentalsetupwasidentical sameepoch,theircorrespondingfluxeswererescaledusingthe tothatusedonthefirstnight. bestfitpowerlaw. 4 J.Sollermanetal.:XRF050824 A power-lawfit in the formF ∝ ν−β providesa tolerable Gorosabeletal.(2003),i.e.,L=0.5±0.2L intherestframe ν ⋆ fitfortheSED.Thespectralindexisβ=0.56±0.04,assuming Bband. negligibleextinction. The host is extended with a size of roughly 0.8 arcsec, Unfortunately,withtheavailabledatawecannotsaymuch which at a distance of 5.24 Gpc correspondsto a linear scale about the extinction. The SMC, LMC or Milky Way extinc- of ∼ 6 kpc. Finally,fromthe [O II] lines we can estimate the tionlawsgiveequallygoodfitstothedata(Fig.3).Fixingthe star formation rate (SFR). The flux of this line, corrected for redshiftat z = 0.83 (see Sect. 3.4), a free fit with an intrinsic Galacticextinction,correspondstoaSFRof1.8 M⊙ yr−1,fol- power-lawshapeoftheSEDandanSMClikeextinctioncurve lowingKennicutt(1998). fromPei(1992)impliesanextinctionof A = 0.4±0.2mag. Infact,usingtheextinctioncorrectedvalueforthefluxof V However,thiswouldgiveanunrealisticallyflatβ∼0spectrum. Hβ, and assuming a case B recombination ratio for Hα ver- Giventhesparsedatasettheonlythingwecanfirmlyconclude sus Hβ, we can also use this line to estimate that the SFR is is thatA is less than 0.5mag. A low value of the globalex- 1.8 M yr−1,againfollowingKennicutt(1998).Asusual,any V ⊙ tinction is also implied by the blue color of the host galaxy slit-losseswouldincreasethisnumber.Wenotethattheconsis- (Sect.3.5). tencyoftheHαand[OII]predictionsoftheSFRalsosupports thenotionoflowextinctioninthehostgalaxy. This SFR compares rather well with the estimate for the 3.4.TheSpectra hostofGRB000210mentionedabove,whichhassimilarprop- erties and an estimated SFR from the UV light of 2.1 ± The combined flux-calibrated spectrum is also included in 0.2 M yr−1(Gorosabeletal.,2003). Fig.3.Theslopeisconsistentwiththecontemporaryphotom- ⊙ The specific star formation rate for the host galaxy of etry. XRF050824isthusonly∼ 4 M yr−1(L/L⋆)−1.Thisisrather We determinedtheredshiftfromthefirstnight’sspectrum ⊙ low, but not exceptional, and falls well within the popula- (Fynboetal.,2005)usingemissionlinessuchas[OII]λ3727, tion of smallstar-formingblue galaxiesas shownin Fig. 2 of [OIII]λλ4959,5007andHβ.AsnotedbyFynboetal.(2005), Sollermanetal.(2005),(seealsoChristensenetal.,2004). wealsodetectabsorptionlinesfromMgIIatthisredshift.We Finally,wecanestimatethemetallicityofthegalaxyusing discuss this further below (Sect. 4.1). The lines are shown in the R23 technique (Pagel, 1979). Using the results presented Fig. 4 and the measured positions and fluxes of the lines are in Table 4 and applying E(B − V) = 0.035 mag, we derive givenin Table 4. Theredshiftisz = 0.828±0.005.Notethat log(R )=1.0. This is quite high, and indicates a metallicity thefluxesgiveninTable4arenotcorrectedforGalacticorhost 23 (just)belowsolar(seee.g.,Fig.5byKewley&Dopita,2002). galaxyextinction.Theuncertaintiesinabsolutelinefluxescan However, the small number of emission lines in the analysis beupto30%. makesthisestimateratheruncertain. Using a cosmology where H = 70kms−1Mpc−1, Ω = 0 Λ The star formation rate and size thus indicates a 0.7 and Ω = 0.3, this redshift corresponds to a luminosity m fairly normal galaxy, similar to other GRB host galax- distanceof5.24Gpc.Thisdistancewillbeusedhereafter. ies (e.g., LeFloc’hetal., 2003; Christensenetal., 2004; Sollermanetal.,2005).Themetallicityconfirmsthetrendthat 3.5.TheHostGalaxy GRB host galaxies have sub-solar metallicities. The luminos- ityisnotparticularlylowcomparedtootherGRBhosts,butis OurlateR-bandimagefromDecember7,2005,104dayspast similartothehostofXRF050416A(Soderbergetal.,2006). explosion, shows an extended source with magnitude R = 23.70±0.15 at the position of the afterglow.This is the host 3.6.TheX-rays galaxy of XRF050824. An image obtained under very good seeing conditionsatVLT in October2005is shownin Fig. 5. Swift-XRTdidnotobservetheburstimmediatelyduetoalunar Notethatalllateobservations(past35days)areconsistentwith constraintand the XRT beganobservationsabout6000s after this beingthe host, with little contributionfromthe afterglow thebursttrigger(Campanaetal.,2005b).Wehaveanalysedthe (orasupernova). standardprocessedXRTdatastartingat0.4daysaftertheburst The host magnitudeis 23.6 when correctedfor a Galactic usingversion2.3oftheSwiftsoftware.Background-subtracted extinction of E(B− V) = 0.035 mag. At the redshift of this spectraandlightcurveswere extractedina standardwaywith galaxy the R band corresponds to the rest frame U band. circularsourceandbackgroundextractionaperturesof30′′and However, comparison of the absolute luminosity with other 60′′radiusforthePC-modedata.DatafromtheWT-modewere galaxiesis often made in the rest frame B band. To do so we notusedbecausetheyaddedverylittlesignal. need to make some assumption about the color of the host. The combined spectrum was fit with a single absorbed Here we note that the BVR magnitudes from our latest VLT power-lawwithabsorptionattheGalacticlevel(N = 3.62× H data, when the host is clearly dominating the emission, are 1020cm−2, Dickey&Lockman,1990)andgaveanacceptable very similar to the magnitudes of the host of GRB000210 fit(χ2 = 43.7for40degreesoffreedom).Addinganabsorber (Gorosabeletal.,2003)atasimilarredshift. at the redshiftof the host galaxy gives a better fit (χ2 = 30.4 We therefore conclude that the absolute luminosity of the for39degreesoffreedom)withanequivalenthydrogencolumn XRF050824 host is very similar to the one determined by densityofN =1.8+0.7×1021cm−2andapower-lawphotonin- H −0.6 J.Sollermanetal.:XRF050824 5 dexβ =1.0±0.1.Theabsorptionmodelhasabundanceratios TheflatearlypartofthelightcurveofXRF030723wasin- X fixed at solar values. The soft X-ray absorption is dominated terpreted in terms of geometry(Fynboetal., 2004), where an byα-chainelementsandisthereforeameasureofthemetalab- off-axisorientationcanmakeanincreasinglylargefractionof sorptionandisregardlessofwhethertheelementsareinthegas the jet visible andthusmaintaina constant(or evenbrighten- orsolidphase(seeWatsonetal.,2006;Turnsheketal.,2003). ing) lightcurve (Granotetal., 2005). We see no evidence for The flux decay of the afterglow followed a single power- thisinXRF050824.It’searlylightcurveisconsistentwiththe law with decay index, α = 0.75 ± 0.04, with the fit being same decay seen throughout the lightcurve. This can thus be X marginally acceptable (χ2 = 24.8 for 16 degrees of freedom, seen asevidenceforanon-axisburst,whichwouldmeanthat null hypothesis probability = 0.07). This decay rate is some- ageometricalinterpretationdoesnotexplainthedifferencebe- whatfasterthantheaverageslopeseenintheopticallightcurve tween XRFs andGRBs in all cases (see also Soderbergetal., (albeit there could be a break in the optical light curve, see 2006). Sect. 4.2.2). Note, however, that the X-ray data are not very constrainingatthelaterphases. 4.2.2. ABreakandaBumpintheLightcurve 4. DISCUSSION At first glance the R-band lightcurveshown in Fig. 2 appears consistentwitha singlepower-lawdeclinethroughoutthe en- 4.1.Absorptionlineredshift tire afterglow. However, since the final points are due to the host galaxy,the data do suggest a break in the lightcurve.As Asnotedin Sect.3.4thespectraalsoincludeabsorptionlines mentioned in Sect. 3.5 the host is extended, so most of the fromMgII.Theselinesareseenatbothepochs,butaremost light at these epochs is indeed from the galaxy. This means clearlydetectedinthefirstepoch,whichhasthebestsignal-to- that the single power law must be broken at an early time, or noiseratio.ThelinesaredisplayedinFig.4,andtheredshiftis the later pointswould have been much brighter. The break in consistentwiththeestimatefromtheemissionlines. thelightcurvealsomeansthatanextracomponentisneededto That the redshift can be determined from both emission explaintheexcesslightat10−20dayspasttheburst. linesandabsorptionlinesisofsomeimportance.Thedistance We embarked on simultaneously fitting two power-laws scale of the XRFs has only recently been shown to be cos- andastretchableSN1998bwtemplate,inaccordancewiththe mological. The first spectroscopic redshift of z = 0.25 for method outlined by Zehetal. (2004). Fixing the host galaxy XRF020903 (Soderbergetal., 2005) was based on emission magnitude, we were able to constrain a shallow break in the linesonly.Inprinciple,asinglecasecouldbeaffectedbyasu- lightcurvetooccurat∼0.5dayspastburst(Fig.6).Thiscould perposedandunrelatedgalaxy,butnowXRF050416Aalsohas be a cooling break.The requiredsupernovais rather unusual. ameasuredemissionlineredshift(z=0.65,Cenkoetal.,2005; In the notation of Zehetal. (2004) this is a supernova with Soderbergetal., 2006) as has GRB/XRF030528 (z = 0.78, k= 1.05±0.42 and s= 0.52±0.14. This is a bright and fast Rauetal.,2005). lightcurve and is different from the lightcurve of the canoni- The GRB030429 studied by Jakobssonetal. (2004) also calSN1998bw,whichisoftenassociatedwithlongGRBs,but displayed absorption lines. It showed an E of 35 keV at a peak is similar to, althoughbrighterthan, the supernovaassociated redshift of z = 2.66, and is thus a borderline case, consistent withXRF060218. with an X-ray Rich burst. More recently, the rather unusual XRF060218 had a secure redshift from both emission lines The actual peak luminosity of the potential supernova is, and absorption lines (z = 0.033, Mirabal&Halpern, 2006; however,highlyuncertain.Ifthereisinternalextinctioninthe Wiersemaetal.,2007;Pianetal.,2006). host galaxy the corresponding supernova would have to be Thesefindings,togetherwiththerobustredshiftdetermina- brighter, but our SED analysis shows that this can not be a tionfortherathernormalXRF050824,havethereforeproven verylargeeffect.Thisisalsosupportedbytheratherbluecolor thecosmologicaldistancescalefortheseobjectsbeyonddoubt. of the host galaxy, and by the deduced balmer line ratios. A largeruncertaintyarisesfromtheassumptionsonthecontribu- tionfromtheafterglow.WithSN1998bwejecting∼0.5 M of ⊙ 4.2.TheLightcurveoftheAfterglow 56Ni(e.g.,Woosleyetal.,1999;Sollermanetal.,2000),wecan estimate that a supernovaassociated with XRF050824would OurR-bandlightcurveofXRF050824isoneofthebestsam- havehadtoejectatleastontheorderof0.6±0.3 M of56Ni. pledopticallightcurvesforanXRF.Themostconspicuousas- ⊙ This is assuming that the peak brightness scales with nickel- pect of this lightcurve is that it is basically consistent with a mass. power-law for the entire duration (Fig. 2). The best fit power lawα=0.6isquiteaslowdecay. 4.2.3. Wiggles,bumps,humpsandjitter 4.2.1. TheEarlyTimes Infact,thefittothelightcurveisnotverysatisfactoryevenaf- Oneof themoreinterestingaspectsofthe lightcurveis thatit terinvokingabreakandahypotheticalsupernovabump.Thisis declinessteadilyfromveryearlyon.Thisisinstarkcontrastto seeninFig.6,wherethereducedχ2is1.8.Theentirelightcurve thelightcurveofXRF030723(Fig.6),whichapparentlyhada of XRF050824 displays wiggles and humps throughout the constantlightcurveforthefirstdayaftertheburst. timeoftheobservations.Thelargestdeviationsarefromasys- 6 J.Sollermanetal.:XRF050824 tematicdipintheNOTdatarelativetotheoverallfitjustbefore jet model may be unable to account for these variabilities 0.1days,andanincreaseinthescatterfrom2–5days.These (Kumar&Piran,2000),whichmayinsteadbeattributedtothe deviationscannotbesatisfactorilyfittedbyabrokenpower-law re-activity of the central engine (e.g., Fanetal., 2005), or, as schemethatissupposedtomodelasimpleimpulsiveshockor mentionedabove,toasupernova(Sect.4.2.2). jet. Anotherexampleisat0.2–0.5dayspasttheburstwhenwe 4.3.Amatirelation haveawellmonitoredlightcurve,inparticularfromtheMDM. ThisisshownintheinsetinFig.2.Alineardecayisnotafor- Several XRFs have been shown to follow the same relation mallygoodfittothesedata.Thereappeartobewigglesaround as GRBs, that Epeak ∝ Ei1s/o2 (Amatietal., 2002), where Eiso thesteadylineardecline.Sincemostoftheseobservationswere is the isotropic-equivalent radiated energy. For XRF050824, obtained at the same telescope, and have been reduced in the wecannotdetermineaprecisetotalenergyduetothelackof samewayagainstthesamelocalstandards,wedonotbelieve knowledgeconcerningthe peak energyof the BAT spectrum. this is purely an instrumental effect. Although the statistical However,wecancalculatelowerandupperlimitsbyintegrat- significanceisratherlowinourlightcurve,wenotethatsimilar ing the best fit power law spectral energy distributions in the jittering has been observed previously in GRBs, both in long (15−150)×(1+z)keVbandandinthethefull1−104 keV GRBs(e.g.,Gorosabeletal.,2006;Mathesonetal.,2003)and band.Weobtained4.1×1050erg< Eiso <3.4×1051erg,which in short GRBs (Hjorthetal., 2007), and can be interpreted in whenusing the Amatirelation,wouldprovidea constrainton termsofvariationsinthesurroundingcircumburstmediumor the observed peak energy 11keV < Eobs < 32keV. This is peak asdue to prolongedactivityof thecentralengine.Similar ex- thusonlyjustin agreementwiththatfromthe spectralfitting, planationscanthusbeputforwardalsoforthisXRF. Eobs <13keV. peak Besides this event, the Amati relation is also applicable to XRFs020903 and possibly 030723 4.2.4. Continuousenergyinjection (Lamb,Donaghy&Graziani, 2005), XRF050416A (Sakamotoetal., 2005) and XRF050406 (Schadyetal., Prolonged central engine activity with multiple energy injec- 2006).(SeealsoAmatietal.,2006;Ghisellinietal.,2006,for tions could thus be the explanation for the deviations from a XRF060218.) perfectpower-law(Bjo¨rnssonetal., 2004). Prolongedactivity ThatthisrelationholdsforbothXRFsandclassicalGRBs intermsofcontinuousenergyinjectioncouldalsoexplainthe not only implies that both classes of bursts can be on-axis slow decline rates in both optical (α = 0.65 when corrected R events(Amatietal.,2006)butalsosupportstheideathatboth for host galaxy contribution) and in X-rays (α = 0.75). In X canbeinterpretedunderaunifiedphysicalmechanism. Fig.7weshowanexampleofamodelfortheafterglowemis- sioninthe X-rayandtheR bandforcontinuousenergyinjec- tion.Thismodelisdetailedbelow. 5. CONCLUSIONS Wehavemodeledtheafterglowintermsofalong-termcon- We have seen that the afterglows of XRFs can appear quite tinual energy injection in the forward shock. We consider an different.TheearlyflatopticallightcurveofXRF030723was uniform relativistic jet undergoing the energy injection from consistent with predictions of an off-axis burst (Fynboetal., the central source and sweeping up its surrounding uniform 2004; Granotetal., 2005). On the contrary, XRF050824 dis- medium.Thedynamicalevolutionoftheoutflowiscalculated plays an optical lightcurve which is decaying at a fairly con- usingtheformulaeinHuangetal.(2000)andaddinganenergy stant, but slow, rate from 10 minutes after the burst. Our injectionprocesswiththeformdE /dt= A(t/t )−qfort <t< inj 0 0 afterglow model indicates this to be an on-axis burst. We t .Thefractionsofshockenergygiventotheelectronsǫ and end e have also found some evidence for a bump in the lightcurve, tothemagneticfieldǫ areassumedtobeconstant. B whichisconsistentwithasupernovaasfastasthatassociated The modelfits shown in Fig. 7 have the followingjet pa- withXRF060218,i.e,considerablyfasterthantheSN1998bw rameters:theisotropickineticenergyEk = 1052erg,ǫe = 0.4, lightcurve. ǫB = 0.003,thecircumburstdensityn = 0.1cm−3,theelectron Most well observed XRFs with a redshift where a super- index p= 2.05,thehalf-openingangleθj = 0.2,andtheview- nova could be expected to emerge do show some evidence ingangleθobs=0(i.e.,on-beamviewing),togetherwiththeen- for this. This is similar to the case for ordinary long GRBs ergyinjectionparameters:A=3×1049erg/s,q=0.8,t0 =100 (Zehetal., 2004). A common origin for XRFs and GRBs is s, and tend = 2 × 106 s. This rather large amount of ejected therefore likely but there also seems to be several parame- energyis needed to explain the long and shallow decline;the ters affecting the observables of the burst. In the context of amountissimilartothatfoundforGRB050315(Zhangetal., the four-field diagram presented by Sollermanetal. (2006), 2006). Since there is no proper jet break until possibly after XRF050824 should be in the same category as XRF020903 a week, the constrainton the jet opening angle of θj∼>10◦ is and XRF030723; XRFs with an associated supernova but quiterobust.Wedidnotattempttofittheveryearlylightcurve. where the opticallight is dominatedby the afterglowat early At these phases it is likely that a reverse shock componentis phases. required. ThemountingevidenceforsupernovaeinGRBsandXRFs We note again the late re-brightening. At such a late also shows that there is a large variety in supernova prop- time, the ejecta is only moderately relativistic. The patchy erties (Woosley&Bloom, 2006). The emergence of super- J.Sollermanetal.:XRF050824 7 novae much fainter (Pianetal., 2006; Sollermanetal., 2006) and much faster (this work) than the canonical SN1998bw put constraints on the underlying explosion model. Recently, Fynboetal. (2006) also reported two GRBs where no su- pernova emission is seen, down to 100 times fainter than SN1998bw.It is still not clear whether we see two (or more) fundamentallydifferentexplosionmechanisms,orifthereisa verywidecontinuumofpropertiesfortheseblasts. Acknowledgements. This paper is based on observations collected by the Gamma-Ray Burst Collaboration at ESO (GRACE) at the EuropeanSouthernObservatory,ParanalandLaSilla,Chile.Wethank theESOstafffortheirhelpinsecuringtheservicemodedatareported here.WeacknowledgebenefitsfromcollaborationwithintheEUFP5 Research TrainingNetwork”Gamma-Ray Bursts: AnEnigmaanda Tool”. This work was also made at the DARK Cosmology Centre funded by The Danish National Research Foundation. JS acknowl- edges support from Danmarks Nationalbank, from the Anna-Greta andHolgerCrafoordfundandfromKnut&AliceWallenbergfoun- dation. CrAO and IKI acknowledge support from Russian Ministry of Education and Science. JPH and NM acknowledge support from theNationalScienceFoundationunder grant0206051. Theresearch activities of JG and MJ are supported by the Spanish Ministry of SciencethroughprojectsAYA2004-01515andESP2005-07714-C03- 03. Based in part on observations made with the BOOTES instru- ments in South Spain (ESAt-INTA/CEDEA, Huelva) and with the 1.5m Telescope of Observatorio de Sierra Nevada (OSN), operated byIAA/CSIC.Someofthedatapresentedherehavebeentakenusing ALFOSC,whichisownedbytheInstitutodeAstrof´ısicadeAndaluc´ıa (IAA) and operated at the Nordic Optical Telescope under agree- mentbetweenIAAandtheDepertmentofAstronomyofCopenhagen University. Thanks to Tamara Davis for very careful reading of the manuscript,andfordetaileddiscussionsonK-corrections.Lastbutnot least, we acknowledge support from several key observers that con- tributedtothiswork,namelyV.BirykovandD.Sharapovaswellas D.R.Rafferty&J.R.Thorstensen.FinallythelateHugoE.Schwarz contributedwithobservationsandcommentsonanearlierversionof thispaper. 8 J.Sollermanetal.:XRF050824 References Hjorth, J., Sollerman, J., Møller, P., et al. 2003, Nature, 423, 847 Allen, C. W. 2000, Allen’s AstrophysicalQuantities, 4th edi- Hjorth,J.,etal.2007,inprep. tion2000,ed.A.N.Cox Huang,Y. F., Gou,L. J., Dai, Z. G. & Lu,T. 2000,ApJ, 543, Amati,L.,Frontera,F.,Tavani,M.,etal.2002,A&A,390,81 90 Amati,L.,DellaValle,M.,Frontera,F.,Malesani,D.,Guidorzi, Jakobsson,P.,Hjorth,J.,Fynbo,J.P.U.2004,A&A,425,785 C.,Montanari,E.,&Pian,E.2006,astro-ph/0607148 Kennicutt,R.C.1998,ARA&A,36,189 Assafin, M., Andrei,A. H., Martins,R., Vieira da Silva Neto, Kewley,L.J.,&Dopita,M.A.2002,ApJS,142,35 D.N.,Camargo,J.I.B.,Teixeira,R.&Benevides-Soares,P. Kippen, R. M., Woods, P. M., Heise, J., in’t Zand, J. J. M., 2001,ApJ,552,380 Briggs, M. S., & Preece,R. D. 2003,AIP Conf.Proc.662: Band,D.,Matteson,J.,Ford,L.,etal.1993,ApJ,413,281 Gamma-Ray Burst and Afterglow Astronomy 2001: A Barraud,C.,Olive,J.-F.,Lestrade,J.P.,etal.2003,A&A,400, WorkshopCelebratingtheFirstYearoftheHETEMission, 1021 662,244 Barraud, C., Daigne, F., Mochkovitch, R., Atteia, J. L. 2005, Krimm,H.,Barbier,L.,Barthelmy,S.,etal.2005,GCN,3871 A&A,440,809 Kumar,P.&Piran,T.2000,ApJ,535,152 Bersier, D., Fruchter, A. S., Strogler, L.-G., et al. 2006, ApJ, Landolt,A.U.1992,AJ,104,340 643,284 Lamb,D.Q.,Donaghy,T.Q.,&Graziani,C.2005,ApJ,620, Bjo¨rnsson, G., Gudmundsson, E. H., & Johannesson, E. H. 355L 2004,ApJ,615,L77 LeFloch,E.,Duc,P.-A.,Mirabel,I.F.,etal.2003,A&A,400, Butler,N.R.,Sakamoto,T.,Suzuki,M.,etal.2005,ApJ,621, 499 884 Levan,A.,Patel,S.,Kouveliotou,C.,etal.2005,ApJ,622,977 Campana, S., Barthelmy,S., Burrows, D., et al. 2005a,GCN, Matheson,T.,Garnavich,P.M.,Stanek,K.Z.,etal.2003,ApJ, 3866 599,394 Campana,S.,Mangano,V.,Chincarini,G.,etal.2005b,GCN, Mirabal,N.,&Halpern,J.P.2006,GCN4792 3872 Mirabal,N.&Halpern,J.P,An,D.,etal.2006,ApJ,643,99 Campana,S.,Mangano,V.,Blustin,A.J.,etal.2006,Nature, Modjaz,Stanek,K.Z.,Garnavich,P.M.,etal.2006,ApJ,645, 442,1008 21 Castro-Tirado,A.J.,Jelinek,M.,MateoSanguino,T.J.,etal. Oke,J.B.1990,AJ,99,1621 2004,AstronomischeNachrichten,325,679 Pagel, B. E. J., Edmunds,M. G., Blackwell, D. E., Chun, M. Cenko,S.B.,Kulkarni,S.R.,Gal-Yam,A.,&Berger,E.2005, S.,&Smith,G.1979,MNRAS,189,95 GCN3542 Pei,Y.C.1992,ApJ,395,130 Christensen, L., Hjorth, J., & Gorosabel, J. 2004, A&A, 425, Pian,E.,Mazzali,P.,Masetti,N.,etal.2006,Nature,442,1011 913 Preece,R.D.,Mallozzi,R.S.,Pendleton,G.N.,&PaciesasW. Crew,G.,Ricker,G.,Atteia,J-L.,etal.2005,GCN,3890 S.2000,ApJS,126,19 Dermer,C.D.,Chiang,J.,&Bo¨ttcher,M.1999,ApJ,513,656 Rau,A.,Salvato,M.,&Greiner,J.2005,A&A,444,425 Deutsch,E.W.1999,AJ,118,1882 Rhoads,J.E.2003,ApJ,591,1097 Dickey,J.M.,&Lockman,F.J.1990,ARA&A,28,215 Sakamoto,T., Lamb,D. Q., Kawai,N., etal. 2005,ApJ, 629, Fan,Y.Z.,&Wei,D.M.2005,MNRAS,364,L42 311 Fukugita,M.,Shimasaku,K.,&Ichikawa,T.1995,PASP,107, Schady,P.,Mason,K.O.,Osborne,J.P.,etal.2006,ApJ,643, 945 276 Fynbo,J.,Sollerman,J.,Hjorth,J.,etal.2004,ApJ,609,962 Schaefer,B.E.,Yuang,F.,&Alatalo,K.2005,GCN,3867 Fynbo,J.P.U.,Jensen,B.L.,Sollerman,J.,etal.2005,GCN, Schlegel,D.J.,Finkbeiner,D.P.,&Davis,M.1998,ApJ,500, 3874 525. Fynbo, J. P. U., Watson, D., Thoene, C., et al. 2006, Nature, Soderberg,A.M.,Kulkarni,S.R.,Fox,D.B.,etal.2005,ApJ, 444,1047 627,877 Ghisellini, G., Ghirlanda, G., Mereghetti, S., Bosnjak, Z., Soderberg, A. M., Nakar, E., Cenko, S. B., et al. 2006, Tavecchio,F.,&Firmani,C.2006,MNRAS,372,1699 astro-ph/0607511 Gorosabel,J.,Casanova,V., Garrido,R., Castro-Tirado,A. J., Sollerman, J., Kozma, C., Fransson, C., Leibundgut, B., Jelinek,M.,&deUgartePostigo,A.2005,GCN,3865 Lundqvist,P.,Ryde,F.,&Woudt,P.2000,ApJ,537,L127 Gorosabel, J., Christensen, L., Hjorth, J., et al. 2003, A&A, Sollerman,J., O¨stlin, G., Fynbo,J. P. U., Hjorth, J., Fruchter, 400,127 A.,&Pedersen,K.2005,NewAstronomy,11,103 Gorosabel, J., Castro-Tirado, A. J., Ramirez-Ruiz, E., et al. Sollerman,J.,Jaunsen,A.O.,Fynbo,J.P.U.,etal.2006,A&A, 2006,ApJ,641,L13. 454,503 Granot,J.,Ramirez-Ruiz,E.,&Perna,R.2005,ApJ,630,1003 Tominaga, N., Deng, J., Mazzali, P. A., Maeda, K., Nomoto, Heise, J., in’tZand,J., Kippen,R. M., & Woods, P. M. 2001, K., Pian, E., Hjorth, J., & Fynbo, J. P. U. 2004, ApJ, 612, in the proc. of the conference, Gamma-ray Bursts in the L105 AfterglowEra,16 Turnshek, D. A., Rao, S. M., Ptak, A. F., Griffiths, R. E., & Monier,E.M.2003,ApJ,590,730 J.Sollermanetal.:XRF050824 9 Watson,D.,Fynbo,J.P.U.,Ledoux,C.,etal.2006,ApJ,652, 1011 Wiersema, K., Savaglio, S., Vreeswijk, P. M., et al., 2007, A&A,inpress,astro-ph/0701034 Woosley, S. E., Eastman, R. G., & Schmidt, B. P. 1999,ApJ, 516,788 Woosley,S.E.,&Bloom,J.2006,ARA&A,44,507 Yamazaki,R.,Ioka,K.,&Nakamura,T.2002,ApJ,571,L31 Yamazaki,R.,Ioka,K.,&Nakamura,T.2003,ApJ,593,941 Zeh,A.,Klose,S.,&Hartmann,D.H.2004,ApJ,609,952 Zhang, B., Fan, Y. Z., Dyks, J., Kobayashi, S., Me´sza´ros, P., Burrows,D.N.,Nousek,J.A.&Gehrels,N.2006,ApJ,642, 354 10 J.Sollermanetal.:XRF050824 Table2.LogofobservationsandphotometryoftheafterglowofXRF050824. Date ∆t Magnitude MagnitudeError PassBand Telescope (UT) (days) (1σ) Aug25.0973 0.130449 20.72 0.04 B-band NOT Aug25.1017 0.134850 20.72 0.05 B-band NOT Aug25.1061 0.139250 20.65 0.04 B-band NOT Oct6.1943 42.2274 24.43 0.16 B-band VLT Oct6.1898 42.2229 24.26 0.17 V-band VLT Aug24.9742 0.007346 18.22 0.35 R-band BOOTES Aug24.9813 0.014429 19.11 0.32 R-band BOOTES Aug24.9935 0.0266991 18.94 0.03 R-band OSN Aug24.9971 0.0302391 19.04 0.04 R-band OSN Aug25.0007 0.0338001 19.07 0.03 R-band OSN Aug25.0096 0.042705 19.67 0.33 R-band BOOTES Aug25.0166 0.0496998 19.31 0.03 R-band OSN Aug25.0337 0.0668297 19.56 0.05 R-band OSN Aug25.0372 0.0703697 19.59 0.07 R-band OSN Aug25.0546 0.0877495 19.80 0.04 R-band NOT Aug25.0609 0.0940495 19.85 0.05 R-band NOT Aug25.0653 0.0984497 19.93 0.04 R-band NOT Aug25.0881 0.121212 19.33 0.20 R-band BOOTES Aug25.1496 0.182749 20.15 0.04 R-band OSN Aug25.1599 0.193050 20.21 0.05 R-band OSN Aug25.1700 0.203150 20.23 0.07 R-band OSN Aug25.1813 0.214449 20.38 0.08 R-band OSN Aug25.2003 0.233429 20.42 0.10 R-band MDM Aug25.2042 0.237391 20.22 0.08 R-band MDM Aug25.2082 0.241360 20.42 0.10 R-band MDM Aug25.2122 0.245300 20.34 0.09 R-band MDM Aug25.2161 0.249279 20.39 0.09 R-band MDM Aug25.2201 0.253250 20.47 0.11 R-band MDM Aug25.2241 0.257210 20.51 0.12 R-band MDM Aug25.2251 0.258249 20.25 0.05 R-band D1.5m Aug25.2280 0.261179 20.39 0.11 R-band MDM Aug25.2303 0.263451 20.47 0.06 R-band D1.5m Aug25.2320 0.265150 20.35 0.11 R-band MDM Aug25.2360 0.269110 20.24 0.10 R-band MDM Aug25.2402 0.273399 20.35 0.07 R-band MDM Aug25.2477 0.280849 20.30 0.07 R-band MDM Aug25.2551 0.288280 20.49 0.08 R-band MDM Aug25.2626 0.295719 20.35 0.08 R-band MDM Aug25.2700 0.303150 20.39 0.07 R-band MDM Aug25.2783 0.311409 20.62 0.11 R-band MDM Aug25.2966 0.329741 20.43 0.05 R-band MDM Aug25.3040 0.337179 20.55 0.08 R-band MDM Aug25.3062 0.339350 20.64 0.09 R-band D1.5m Aug25.3111 0.344250 20.69 0.10 R-band D1.5m Aug25.3115 0.344610 20.51 0.08 R-band MDM Aug25.3288 0.361910 20.49 0.06 R-band MDM Aug25.3362 0.369339 20.62 0.07 R-band MDM Aug25.3436 0.376780 20.55 0.06 R-band MDM Aug25.3511 0.384220 20.60 0.07 R-band MDM Aug25.3587 0.391870 20.58 0.06 R-band MDM Aug25.3632 0.396349 20.55 0.05 R-band VLT Aug25.3661 0.399300 20.60 0.06 R-band MDM Aug25.3736 0.406740 20.60 0.06 R-band MDM

See more

The list of books you might like

Upgrade Premium
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.
DMCA & Copyright: Dear all, most of the website is community built, users are uploading hundred of books everyday, which makes really hard for us to identify copyrighted material, please contact us if you want any material removed.
Copyright @ PDFdrive.to, 2022 | We love our users