Table Of Content

Astronomy&Astrophysicsmanuscriptno.AXP˙obs˙astropacca c ESO2008 (cid:13) February2,2008 Adaptive optics near-infrared observations of magnetars V.Testa1,N.Rea2,3,R.P.Mignani4,G.L.Israel1,R.Perna5,S.Chaty6,L.Stella1,S.Covino7, R.Turolla8,4,S.Zane4,G.LoCurto9,S.Campana7,G.Marconi9,S.Mereghetti10 1 INAF-OsservatorioAstronomicodiRoma,ViaFrascati33,00040MontePorzioCatone,Italy 2 UniversityofAmsterdam,AstronomicalInstitute“AntonPannekoek”,Kruislaan403,1098SJ,TheNetherlands 8 3 SRONNetherlandsInstituteforSpaceResearch,Sorbonnelaan,2,3584CA,Utrecht,TheNetherlands 0 4 MSSL,UniversityCollegeLondon,HolmburySt.Mary,DorkingSurreyRH56NT,UK 0 5 JILAandDepartmentofAstrophysicalandPlanetarySciences,UniversityofColorado,440UCB,Boulder,CO,80309,USA 2 6 LaboratoireAIM,CEA/DSM–CNRS–Universite´ ParisDiderot,DAPNIA/Serviced’Astrophysique,Baˆt.709,CEA–Saclay,FR- 91191Gif-sur-Yvette,France n 7 INAF-OsservatorioAstronomicodiBrera,viaBianchi46,23807Merate(Lc),Italy a J 8 UniversityofPadua,PhysicsDepartment,viaMarzolo8,35131,Padova,Italy 9 ESO-EuropeanSouthernObservatory,AlonsodeCordova3107,Vitacura,Santiago,Chile 4 10 INAF-IstitutodiAstrofisicaSpazialeeFisicaCosmica‘G.Occhialini’,viaBassini15,20133,Milano,Italy 2 Received/Accepted ] h ABSTRACT p - o Context.Wereportonnear-infrared(IR)observationsofthethreeanomalousX-raypulsarsXTEJ1810–197,1RXSJ1708–4009and r 1E1841–045,andthesoftgamma-rayrepeaterSGR1900+14,takenwiththeESO-VLT,theGemini,andtheCFHTtelescopes. st Aims.This work is aimed at identifying and/or confirming the IR counterparts of these magnetars, as well as at measuring their a possibleIRvariability. [ Methods.InordertoperformphotometryofobjectsasfaintasK 20,wehaveuseddatatakenwiththelargesttelescopes,equipped s∼ withthemostadvancedIRdetectorsandinmostofthecaseswithAdaptiveOpticsdevices.Thelatterarecriticaltoachievethesharp 2 spatialaccuracyrequiredtopinpointfaintobjectsincrowdedfields. v Results.We confirm with high confidence the identification of the IR counterpart to XTEJ1810–197, and its IR variability. For 1 1E1841–045 andSGR1900+14 weproposetwocandidateIRcounterpartsbasedonthedetectionofIRvariability.For1RXSJ1708– 7 4009 weshowthatnoneofthepotentialcounterpartswithinthesourceX-rayerrorcirclecanbeyetconvincinglyassociatedwith 1 thisAXP. 4 Conclusions.TheIRvariabilityoftheAXPXTEJ1810–197 doesnotfollowthesamemonotonicdecreaseofitspost-outburstX-ray . 2 emission. Instead, the IR variability appears more similar to the one observed in radio band, although simultaneous IR and radio 1 observations are crucial to draw any conclusion in this respect. For 1E1841–045 and SGR1900+14 follow-up observations are 7 neededtoconfirmourproposedcandidateswithhigherconfidence. 0 Keywords.stars:pulsars:individual:XTEJ1810–197,1RXSJ1708–4009,1E1841–045,SGR1900+14 : v i X 1. Introduction power-law(Γ 2–4;fora reviewsee e.g.Woods&Thompson r ∼ a 2006).Their X-ray energyoutputis much larger than the rota- In recent years, the study of isolated neutron stars has become tionalenergylossandimpliesthatthesesourcesarenotrotation- oneofthemostchallengingresearchareasinhigh-energyastro- allypowered,atvariancewithmostyoungisolatedneutronstars. physics,largelyasaresultofthediscoveryofseveralnewclasses Rather,AXPsandSGRsarebelievedtobepoweredbytheneu- ofsourcesbesidestheclassical,well-studiedradiopulsars.The tronstarultra-strongmagneticfield(Duncan&Thompson1992; most extreme objects are the so-called “magnetars”. This class Thompson & Duncan 1995). In the ”magnetar” model, crustal comprises the Anomalous X–ray Pulsars (AXPs) and the Soft deformations,drivenbymagneticstressesimpartedtothecrust Gamma–ray Repeaters (SGRs), observationallyvery similar in bythestronginternaltoroidalmagneticfield,areresponsiblefor manyrespects.TheyallareslowX–raypulsarswithperiodsin theobservedactivity,X/γ-rayburstsandgiantflares. a narrow range (P= 2–12s), relatively large period derivatives (P˙ = 10 13 10 10ss 1), spin-down ages of 103 104yr, and Until not long ago, the persistent emission of AXPs and − − − magnetic fie−lds, as inferred from the classical ma−gnetic dipole SGRs was detected only in the soft X-ray range. However, in spin-down formula1, of 1014 1015G, larger than the electron the last few yearsthe availability of more sensitive optical and quantumcritical field (B −4.4 1013G) abovewhich quan- IRtelescopes,aswellasofγ-raysatellites,openednewwindows cr tum effects become crucial.≃AXPs×and SGRs are strong X-ray inthestudyofAXPsandSGRs,unveilingtheirmulti-bandprop- emitters, with X-ray luminosities of about 1034 1036ergs 1. erties.Inparticular,observationsfromESO,Gemini,CFHTand − Their0.1–10keVpersistentemissionhasrelativel−ysoftspectra Keck (see Israel et al. 2004a for a review) led to the discovery usuallymodeledbyanabsorbedblackbody(kT 0.2–0.6)plusa of faint (Ks 20), and in some cases variable, IR counterparts ∼ to five out o∼f seven confirmed AXPs (one of which was also Sendoffprintrequeststo:V.Testa:[email protected] detected in the opticalband by Hulleman et al. (2000)),and to 1 B 3 1019√PP˙ G one(outoffour)confirmedSGR(Israeletal.2005,Kosugietal. ∼ × 2 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars Telescope/Instrument Target Date(YY.MM.DD) MJD Filter ExposureTime(s) FWHM(′′) airmass VLT/NACO XTEJ1810–197 2003.10.08 52920 H 2520 0.189 1.17 2003.10.08 52920 Ks 2520 0.243 1.49 2004.03.14 53078 H 1080 0.108 1.47 2004.03.14 53078 Ks 1200 0.094 1.20 2004.03.14 53078 J 1080 0.081 1.13 2004.09.09 53257 J 1800 0.170 1.01 2004.09.09 53257 H 3600 0.140 1.08 2004.09.13 53261 Ks 3600 0.099 1.01 1RXSJ1708–4009 2003.05.20 52779 Ks 2400 0.083 1.04 2003.06.19 52809 H 2400 0.122 1.40 2003.06.20 52810 J 2400 0.122 1.04 2004.03.25 53089 L’ 2100 0.119 1.04 SGR1900+14 2006.04.01 53826 Ks 1560 0.125 1.43 2006.07.21 53937 Ks 2040 0.111 1.21 Gemini/NIRI XTEJ1810–197 2003.09.18 52900 Ks 285 0.397 1.62 2003.09.18 52900 H 285 0.397 1.99 2003.09.18 52900 J 285 0.397 1.78 2004.06.10 53166 Ks 510 0.631 1.32 2004.06.10 53166 H 510 0.526 1.30 2004.06.10 53166 J 510 0.655 1.32 2004.07.28 53214 Ks 510 0.468 1.39 2004.07.28 53214 H 510 0.526 1.53 CFHT/AOBIR 1E1841–045 2002.08.17 52503 K’ 1260 0.137 1.11 2002.08.17 52503 H 1260 0.161 1.31 Table1.Observationssummary. Fig.1.XTEJ1810–197:Leftpanel:VLT/NACOK-bandimage(March2004).Theradioposition(Camiloetal.2007c)ismarked togetherwithitscandidatecounterpart(objectA).Twelvenearbyobjects,someofwhichusedasareferenceforrelativephotometry, arealsomarkedinthefigure.Rightpanel:K vs.H-K colour-magnitudediagramofthesourcesdetectedintheNACOfieldofview S S (filledcircles).ObjectA ismarkedwitharectangleindicatingtherangeincolourandmagnitudecoveredthroughouttheepochs, whilesomeofthenearbyobjectsmarkedinthemapandlistedinTab.2areplottedinredandlabeled.Magnitudesandcoloursfor allthesourcesbutobject’A’havebeenaveragedoveralltheobservations. 2005). Furthermore, deep INTEGRAL observations in the 20– ofthesesourcesisconcerned.Theextrapolationofthecanonical 200keV band revealed that most of these highly magnetized blackbodyplus power-law,used to model the soft X-ray emis- sourcesarealsohardX-rayemitters(Kuiperetal.2004,2006). sion of these sources, largelyover-predictstheir opticalandIR Therecentdiscoveryofpulsedradioemissionfromthetransient emission. Recently, new attempts have been made at modeling magnetars XTEJ1810–197 (Halpernetal. 2005, Camiloetal. the multi-bandspectra of magnetarsto overcomethis problem. 2006)and1E1547.0 5408(Camiloetal.2007a),theonlycases In particular, Reaetal. (2007a) showed that AXP spectra can − sofar(Burgayetal.2006,2007),isintriguingsinceradioemis- be fitted by a resonant cyclotron scattering model or two log- sion was believed to be quenched in magnetic fields above the parabolicfunctions,whichdonotover-predicttheirdimoptical quantumcriticallimit(Baring&Harding1998). and IR magnitudes. However, despite solving this issue, these newspectralmodelsarestillfarfromgivinganyphysicalinter- Despite the more complete observational picture we have pretationfortheopticalandIRemission. nowforAXPsandSGRs,thephysicalprocessesatthebasisof theiremissioninthedifferentbandsarenotfullyunderstoodas Some possibilities are that either the magnetars opti- yet.ThisisparticularlytrueasfarastheopticalandIRemission cal/IR emission arise from the star surface/magnetosphere Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 3 (Beloborodov& Thompson2006),or fromthe reprocessingof the X-ray emission via a fossil disk around the neutron star (Chatterjee et al. 2000, Perna et al. 2000). Indeed, the Spitzer mid-IR detection of the AXP 4U0142+614(Wangetal. 2006) hasbeeninterpretedasthefirstevidenceofapassivefossildisk aroundanAXP,believedtoemitthroughreprocessingoftheX- rayradiationfromthemagnetar. In this paper we report on Gemini and ESO-VLT observa- tionsofthetransientAXPXTEJ1810–197,onESO-VLTobser- vations of the AXP 1RXSJ1708–4009 and of SGR1900+14, andCFHTobservationsoftheAXP1E1841–045( 2).There- § sultsarepresentedanddiscussedinthecontextofdifferentemis- sionscenarios,andcomparedwiththeopticalandIRproperties ofothermagnetars(see 4). § 2. Observationsanddatareduction 2.1.Observations Observationswereperformedwiththe8mVeryLargeTelescope (VLT)andthe3.5mCanada-France-HawaiiTelescope(CFHT), bothequippedwithadaptiveoptics(AO)andnear-IRcameras.In addition,weusedarchivalobservationsperformedwiththe8m Gemini telescope, also equipped with an IR camera but with- outAO, thatwere offeredstartingfrom2004only.A summary of the telescopesand instrumentsused, as well as the observa- tionslogs,isreportedinTab.1.Inallcasesimageswereobtained throughthestandardobservingtechniquecommonlyappliedto IR arrays, i.e. by stacking repeated exposuresalong each node of a pre-defined dithering pattern, with each node jittered by a few arcsec with respectto the previousone. Thisallowsto ob- Fig.2.XTEJ1810–197:Toppanel:X-rayflux(showninunitsof tain a set of dithered exposures that are used to produce sky- 10 11ergcm 2s 1)historyinthe0.5–10k˙eVenergyrange(from − − − backgroundimages. Gotthelf & Halpern 2007). Lower panels: from top to bottom: VLT/YepunobservationsofXTEJ1810–197,1RXSJ1708– K ,H,andJbandrelativemagnitudevariationforXTEJ1810– s 4009 andSGR1900+14 werecarriedoutindifferentobserving 197 counterpart(red star) and for two nearby, bright compari- runs (see Tab. 1) using NAos COnica (NACO)2 with the S27 son objects. Magnitude variations are computed relative to the camera (0′.′027 pixel size, 28′′ 28′′ field of view). Gemini- September2004observation.Tobetterfitintheplot,magnitude × North observations of XTEJ1810–197 were carried out using differencesforobjects#2and#8havebeenshiftedby+1.1and- the Near-InfraRed Imager (NIRI)3 with the f/6 camera (0′.′117 1.1mag,respectively.BottompanelshowstheH-Kscolourvari- pixel scale, 120′′ 120′′ field of view). 1E1841–045 was ob- ation.Magnitudeerrorsat(1σ)areshown.Theseincludeinter- served at the CFH×T using the AdaptiveOptics Bonnet(AOB)4 nalerrors,aperturecorrectionerrorsandzeropointerrors. InfraRedcamera(0.035pixelsize,36 36 fieldofview). ′′ ′′ ′′ × 2.2.Datareductionandcalibration in the VLT-NACO images the PSF was found to be variable acrossthefieldofview,whileintheCFHT-AOBimagesitwas TheVLTdatawerepre-reducedwiththeESONACOpipeline5, found to feature an unusual profile with a clearly visible sec- whichisbasedonthepackageeclipse,whilethenativeversion ondarydiffractionring.Botheffectsaredifficultto accountfor, oftheeclipse6packagewasusedfortheGeminiandCFHTdata. even by using more sophisticated and position-dependentPSF Inallcases,theimagepre-reductionproducedafinalco-added, models.Ontheotherhand,intheGemini-NIRIimagesthePSF sky-subtractedandflat-fieldedimageforeachband. was found to be more stable, although worse sampled due to Dependingonthecase,fluxeswerecomputedeitherthrough thelargerpixelsize.Forthesereasons,wedecidedasageneral PSF or aperture photometry using the IRAF7 version of the strategytouse aperturephotometryforourAO images,i.e. the daophotpackage(Stetson1992).ForAOobservationswefound NACOandAOBones,andtousePSFphotometryfortheNIRI someinstrument-dependenteffectswhichhamperedthePSFfit- images.Fortheformer,aperturephotometrywasperformedus- ting and discouraged the use of PSF photometry.In particular, ing an aperture or diameter twice the measured PSF. Aperture andPSFphotometrieshavebeencomparedbyapplyingaperture 2 www.eso.org/instruments/naco correctionsusingtheIRAFdaogrowalgorithm,whichcalculates 3 www.gemini.edu/sciops/instruments/niri the aperturecorrectionextrapolatingto anideally infinite aper- 4 www.cfht.hawaii.edu/Instruments ture. 5 www.eso.org/instruments/naco/ 6 www.eso.org/eclipse For the VLT images the photometric calibration was per- 7 IRAF is distributed by the National Optical Astronomy formed by observing standard stars from the Perssonetal. Observatories, which are operated by the Association of Universities (1998) catalog, usually observed under similar conditions(and forResearchinAstronomy,Inc.,undercooperativeagreementwiththe ataboutthesameepoch)asourtargets.ForalltheNACOruns NationalScienceFoundation. thezeropointuncertaintyhasbeenfoundtobeverysmall,i.e.in 4 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars SourceId. R.A.(2000) Dec.(2000) 52900 52920 53078 53166 53214 53257 53261 Ks A 272.46283 -19.73110 20.81 0.08 20.82 0.09 21.18 0.05 21.21 0.10 21.04 0.09 1 272.46390 -19.73016 13.87±0.07 13.85±0.04 14.26±0.04 14.01±0.04 14.01±0.08 2 272.46391 -19.73061 17.23±0.07 17.25±0.04 17.23±0.04 17.36±0.04 17.30±0.08 3 272.46338 -19.73070 19.09±0.07 19.21±0.05 19.47±0.04 19.15±0.05 19.41±0.08 4 272.46335 -19.73110 19.43±0.07 19.37±0.05 19.46±0.04 19.42±0.04 19.47±0.08 5 272.46399 -19.73098 19.52±0.07 19.77±0.06 19.67±0.04 19.62±0.04 19.70±0.08 6 272.46384 -19.73145 18.61±0.07 18.61±0.04 18.60±0.04 18.66±0.04 18.67±0.08 7 272.46327 -19.72952 17.69±0.07 17.70±0.04 17.73±0.04 17.75±0.04 17.75±0.08 8 272.46535 -19.72897 15.48±0.07 15.40±0.04 15.45±0.04 15.51±0.04 15.53±0.08 9 272.46264 -19.72882 15.25±0.07 15.15±0.04 15.23±0.04 15.27±0.04 15.23±0.08 10 272.46082 -19.73013 14.43±0.07 14.19±0.04 14.55±0.04 14.44±0.04 14.29±0.08 11 272.46430 -19.73252 14.83±0.07 14.74±0.04 14.91±0.04 14.87±0.04 14.90±0.08 12 272.46502 -19.73141 16.59±0.07 16.50±0.04 16.51±0.04 16.63±0.04 16.54±0.08 ± ± ± ± ± H A 21.67 0.12 21.81 0.16 22.55 0.07 21.48 0.12 21.73 0.11 22.29 0.18 1 14.27±0.05 14.20±0.04 14.39±0.04 14.24±0.09 14.24±0.08 14.24±0.08 2 17.66±0.05 17.68±0.04 17.66±0.04 17.69±0.09 17.71±0.08 17.67±0.08 3 20.00±0.06 20.11±0.05 20.07±0.04 19.90±0.09 19.78±0.09 20.07±0.08 4 22.24±0.14 22.05±0.11 22.06±0.09 21.66±0.23 21.87±0.10 22.64±0.17 5 22.28±0.28 22.74±0.42 ±- 21.79±0.21 21.68±0.10 22.80±0.11 6 21.17±0.09 21.38±0.07 21.38 0.05 21.11±0.10 21.15±0.09 21.34±0.09 7 18.16±0.06 18.19±0.04 18.14±0.04 18.14±0.09 18.15±0.08 18.20±0.08 8 18.95±0.06 19.00±0.05 18.90±0.04 18.90±0.09 18.93±0.08 18.95±0.08 9 18.42±0.06 18.42±0.04 18.34±0.04 18.44±0.09 18.41±0.08 18.44±0.08 10 14.62±0.05 14.51±0.04 14.65±0.05 14.61±0.09 14.61±0.08 -± 11 15.15±0.05 15.12±0.04 15.19±0.04 15.13±0.09 15.14±0.08 15.12 0.08 12 16.91±0.05 16.89±0.04 16.83±0.04 16.93±0.09 16.91±0.08 16.90±0.08 ± ± ± ± ± ± J A 22.92 0.22 23.89 0.23 23.45 0.12 1 ± 15.37±0.06 15.04±0.04 2 18.85−0.05 18.83±0.06 18.88±0.04 3 21.06±0.06 21.35±0.07 21.35±0.04 7 19.34±0.05 19.39±0.06 19.36±0.04 9 18.06±0.04 18.00±0.06 18.01±0.04 10 ± 15.54±0.06 ± 11 15.98−0.04 16.01±0.06 15.97− 0.04 12 19.04±0.05 18.94±0.06 18.99±0.04 ± ± ± Table 2.XTEJ1810–197:Fromtoptobottom:K ,HandJmagnitudesforthecandidateIRcounterpartofXTEJ1810–197 and s some of the twelve nearby objects, for the 6 differentobservingruns(see Tab.1). Numbersatop the magnitudecolumnsgive the observationsepochsinModifiedJulianDays(MJD).Magnitudeerrorsaregivenat1σandincludezeropointerrors,eitherabsolute orrelative,thatare,forthebrightestobjects,thedominanterrorsource. the0.02-0.05magnituderange.Theresidualuncertaintyintro- totheSeptember2004one.Therelativephotometricuncertainty, ducedbytheextinctiontermisalmostwipedoutbythecloseness basedonabout100starsincommonamongthevariousimages intimeandairmassoftargetfieldsandstandards.Sinceitturns is of the order of 0.05 magnitudes for the NACO-to-NACO outtobe<0.01magnitudesithasbeenneglected. registration,andsli≈ghtlyworsefortheGemini-to-NACOone( ∼ 0.08magnitudes). ≈ The Gemini images of XTEJ1810–197 have been at first FortheCFHT photometry,forwhichnoNACO imagesare instancecalibratedwiththeavailablestandardstars,buttheun- available, the only optionwas to use the 2MASS catalog,after certainty in the zero points and the zero point differences usu- carefully accounting for all the caveats mentioned above. For ally found between standards taken on the same nights sug- theL observationof1RXSJ1708–4009 wehaveusedstandard gested us to discard the primarystandard calibration.A photo- ′ starsfromtheUKIRTIRstandardcatalog(Leggettetal.2003), metric calibration was also tried using stars from the 2MASS yieldingaphotometricaccuracyof 0.1magnitudes. (Skrutskieetal. 2006) catalog identified in the target fields. ∼ However, such an on-the-frame calibration suffered from two In all cases, astrometric calibration was performed using 2MASS stars as reference, yielding an average uncertainty of main problems. Firstly, the numberof suitable 2MASS stars is 0.2, after accounting for the intrinsic astrometric accuracy often quite small and the brighter ones were found to be gen- ′′ ∼ of 2MASS ( 200 mas) and the r.m.s. of the astrometric fits. erallysaturatedinourimages.Furthermore,the2MASScatalog ∼ However,duetothesmallinstrumentpixelscales,thelatterhas wasbuiltfromimageswitha2 pixelsizesothat,inmanycases, ′′ beenfoundtobemuchsmallerthanthe2MASSastrometricac- the measured magnitudescorrespondto blended objectswhich curacyandhenceithasbeenneglected. are, instead, resolved in our higher resolution images. Thus, given the accuracy of the NACO zeropoints and the very low r.m.s. of the photometric solution, we chose to re-calibrate the 3. Dataanalysisandresults Geminiphotometryof the XTEJ1810–197 field on the NACO onebyusing,assecondaryphotometriccalibrators,asetofwell- 3.1.XTEJ1810–197 suited field stars in common between the NACO and the NIRI images. As a reference, we chose the September 2004 NACO Fig.1(left)showstheXTEJ1810–197 field(K band)observed s observationsbecause of their overallbetter quality. In order to withNACOonOctober2003(seealsoIsraeletal.2004b).The ensurefullconsistencyamongobservationsperformedatdiffer- XTEJ1810–197 candidate IR counterpart, labeled A, is coin- ent epochs (Tab. 1), also the NACO October 2003 and March cident with the radio position (RA 18h 09m 51s.087, DEC - 2004photometryoftheXTEJ1810–197 fieldwasre-calibrated 19 43 51.93; error = 0.01; Camilo et al. 2006). For both ◦ ′ ′′ ′′ Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 5 the NACO and NIRI observations, photometry was computed Id RA(◦) DEC(◦) J H Ks andcalibratedasdescribedintheprevioussection.Foreachin- 12 225577..1199555110 --4400..1144789165 2202..9233±00..0197 1189..6804±00..0069 1178..5831±00..0045 strument, the resulting J, H and Ks band catalogues were then 3 257.19560 -40.14799 22.11±0.25 20.01±0.12 18.85±0.05 matched and compared with the multi-band ones. The multi- 4 257.19503 -40.14792 22.03±0.17 20.19±0.09 19.63±0.07 5 257.19498 -40.14800 ± 20.50±0.10 19.95±0.08 bandmagnitudesoftheXTEJ1810–197 candidatecounterpart, 6 257.19495 -40.14800 − ± 20.03±0.09 all calibrated to the NACO system, are reported in Tab.2 to- 7 257.19494 -40.14793 − 20.90−0.11 20.25±0.08 gether with those of a number of field objects, chosen among 8 257.19563 -40.14790 − ± 20.83±0.10 9 257.19502 -40.14786 − 21.55−0.18 20.92±0.11 those used as a referencefor the photometrycross-calibrations 10 257.19531 -40.14795 − ± 20.94±0.11 (seeprevioussection).Wenotethatthecandidatecounterpartis 11 257.19500 -40.14804 − − 20.69±0.10 not detected in the J and Ks-band Gemini observations of the 1132 225577..1199552468 --4400..1144786112 −− −− 2211..2132±±00..1149 June 2004 run which were performedin non-photometriccon- 14 257.19545 -40.14774 − − 21.34±0.21 ditions. Given the short integration time, the overall poor data 15 257.19523 -40.14807 − − 21.52±0.17 16 257.19509 -40.14787 − − 21.58±0.12 quality,andthelackofstandardobservations,wecouldonlyde- 17 257.19510 -40.14799 − − 21.59±0.16 rive tentative estimates on the limiting magnitudes (3σ upper 18 257.19538 -40.14816 − 21.34−0.15 21.66±0.14 limits of J>21.9 and K >20) which are of no use to constrain 19 257.19550 -40.14813 − ± 21.74±0.16 s 20 257.19530 -40.14792 − − 21.88±0.19 thefluxevolutionofoursource. − − ± The time-averaged (Ks, H-Ks) colour-magnitude diagram Table3.1RXSJ1708–4009:J,H,andKsmagnitudesforallthe (CMD) of the field is shown in Fig.1 (right). A well-defined objectsdetectedwithin1′.′1ofthesourceX-rayposition,ordered main sequence group of stars, probably tracking a young star bydecreasingKs magnitude.Noobjecthasbeendetectedinthe cluster, is recognizable in the left part of the diagram, while L′band.Magnitudeerrorsaregivenat1σ,andincludezeropoint a second red clump of stars, most likely red giants or highly uncertainties. absorbed main sequence stars, is present on the right. The XTEJ1810–197 counterpart (green star) appears much fainter than the nearby comparison stars (red dots), and its colour in- deedsuggeststhatitmightbeapeculiarobject. The variability of the XTEJ1810–197 IR counterpart is shown in Fig.2, together with the X-ray variability of the X- ray source (see 4 for details). While in the X-rays the source § features a clear, almost monotonic flux decay, in the IR there is no clear trend in the flux evolution, and hence no correla- tion with the X-rays. Indeed, the flux seems to vary in an er- ratic way in all the threeIR bands.Inparticular,we foundthat the overallvariabilityis apparentlylargerin the H andJ bands (1.07 0.24and0.97 0.32magnitudes,respectively)thaninthe ± ± K one (0.40 0.13 magnitudes). We note that our detailed re- s ± analysis of the October 2003 and March 2004 observations of XTEJ1810–197 yields a K band variability slightly smaller s (0.36 0.10) than that originally reported by Rea et al. (2004) ± usingthesameobservations(0.50 0.10),althoughitisstillcon- ± sistent within the errors.We attribute this (nonsignificant)dif- ference to the more accurate relative photometry between the two epochs, performed using a larger set of secondary photo- Fig.3.1RXSJ1708–4009:NACOKs-bandimageofthesource metric calibrators than the one used in Rea et al. (2004). The fieldwiththe1′.′1radius(99%confidencelevel)X-rayerrorcir- last panelshows the (H-K ) colourvariabilityof the source.In cleoverlaied.Objectsdetectedat 3σarelabeled. s ≥ general,we note that the sourceseems to becomeredderwhen itsfluxdecreases,althoughnoclear trendcanbe recognizedin oursparsephotometry.Weneedtowarn,however,thatanycon- increase the overall uncertaintieson the measured magnitudes. clusion on the source flux and colourvariability must be taken Toquantifytheserandomerrorswehavecross-checkedthepho- with due care. First of all, the apparent lack of correlation be- tometryofanumberofteststarswithbrightnesscomparableto tween the variability in the H and K bands is only suggested theoneofourtarget.Wefoundthatallteststarsshowascatterof s by the July 2004 Gemini Ks band observation. If we take this ∼0.30magnitudesinJandof∼0.10magnitudesintheHandKs observationoutandweconsideronlytheco-evalfluxmeasure- bandsfortheNACO-to-NACOregistration,whilethescatterin- ments,theK bandlightcurvebecomesmoreconsistentwiththe creasesto 0.20magnitudesfortheGemini-to-NACOregistra- s ∼ H(aswellaswiththeJ)bandone.Moregenerally,althoughspe- tion.Hence,wetakethesevaluesasrepresentativeoftherandom cial care was devotedto check the cross-instrumentcalibration errorsofourrelativephotometry.Byaddingthemtotheformal (see 2.2),itispossiblethatourrelativephotometryisaffected errorswe findthatonly the H bandvariability(1.07 0.26)can byra§ndomerrors.Thesemaybeinduced,e.g.byfluctuationsin beconsideredformallysignificant,anditisconsisten±twiththat the atmosphericconditions, sky background,seeing, quality of measured in the J band (0.97 0.43)and, marginally,also with ± theprimarycalibrationframes(darkandflatfields),andglitches theKs bandone(0.40 0.16). ± in the detector performance.Of course, while these effects are marginal for the relatively bright stars that we have used as a 3.2.1RXSJ1708–4009 referencetocomputeourcross-instrumentphotometriccalibra- tion, they are indeed significant for much fainter stars, such as Fig.3 shows the NACO K -band image of the field around the s the XTEJ1810–197 candidate counterpart (K 21), and can Chandrapositionof1RXSJ1708–4009(RA17h08m46s.9,DEC s ∼ 6 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars Fig.4.1RXSJ1708–4009:Colour-magnitudediagramsfortheobjectsinthefieldofview.Sourcesdetectedwithinorclosetothe X-rayerrorcircle(seeFig.3)areshowninredandlabeled. -40 08 52.4).ThesizeoftheX-rayerrorcirclehasbeenesti- ROSATX-rayposition,liesmorethantwoerrorradiiawayfrom ◦ ′ ′′ matedconsideringtheuncertaintyontheChandraposition,that the updated Chandra position. The CMD shows a well-defined is 0.7 (Israeletal. 2003), combined with the overall accuracy mainsequenceextendinguptoK 12,withthecandidatesoc- ′′ ′ ∼ ofourastrometriccalibration(see 2.2),yieldinga finaluncer- cupyingthe fainter regionandstretchingovera broadrangeof § tainty of 1.1 at 99% confidence level. The magnitudes of all colours. ′′ theobjectsdetectedwithintheX-rayerrorcirclearereportedin Bycomparingourphotometricresultswiththosereportedin Tab.3.Nootherobjectisdetecteddowntothe3σlimitingmag- literaturefor1E1841–045 (Durant2005),itisevidentthatour nitudesof22.98,22.20,22.26intheJ,H,K bands,respectively. source #9 (source B in Durant (2005)), shows a >4σ variabil- s IntheL bandnoobjectisdetectedwithintheX-rayerrorcircle ity (∆Ks =1.30 0.24).As a comparison,the average ∆Ks and ′ ± down to 17.8 (the deepest limit ever obtained for an AXPs in ∆HbetweenourphotometryandthatofDurant(2005)for9ob- thisband).Fig.4showsthe(K ,J-K )and(H,H-K )CMDsof jectsinthefieldofviewis 0.2magnitudes.Wethententatively s s s ∼ alltheobjectsdetectedinthefield.ThetwoCMDsshowafairly proposethisobjectastheIRcounterparttothisAXP. scattered sequence.Despite previousclaims (Israelet al. 2003; Safi-Harb&West2005;Durant&vanKerkwijk2006),noneof 3.4.SGR1900+14 these objects show significant (>3σ) IR variability to justify a safe identification as the IR counterpart of 1RXSJ1708–4009 Fig.6 shows the NACO K band image around the s (seealso 4). SGR1900+14 radio position (RA 19h 07m 14s.33, DEC § +09 19 20.1, with a 1 σ uncertainty of 0.15 in each coor- ◦ ′ ′′ ′′ dinate; Frail et al. 1999). Our astrometry yields an overall 3σ 3.3.1E1841–045 positionuncertaintyof0.81,whichcompareswellwiththeone ′′ An updated determination of the 1E1841–045 position of0.79(99%confidencelevel)quotedbyKaplanetal.(2002). ′′ (Wachteretal.2004)wasobtainedwithChandraandgivesRA We notethat the PSFs of field objectsare slightly elongatedin 18h 41m 19s.336, DEC -04 56 10.83, with a 3σ uncertainty the NACO images, and one of the objects detected within the ◦ ′ ′′ radius of 0.9. However, the latter uncertaintyis based only on radioerrorcircle (our#3and objectC of Kaplanet al. (2002)) ′′ five 2MASS reference stars used for the boresight correction looksapparentlyextended.However,its PSF is consistentwith (Wachter et al. 2004). We decided to use a conservative nom- that measured for most field objects. In this case, photometric inal Chandra positional uncertainty of 0.7 at 90% confidence calibrationwasperformedusingasreferencethemagnitudesof ′′ level. Fig.5 shows a revised finding chart of the field around a few objects detected aroundthe targetposition (Kaplanetal. the updated X-ray position. The size of the X–ray error circle, 2002, Frailetal. 1999) (see Fig.6). The three new objects 1.2(99%c.l.)accountsfortheoverallaccuracyofourastromet- detected in our NACO images are also marked (# 7, 8 and 9). ′′ riccalibration(see 2.2).Themagnitudesofallobjectsdetected In particular, object #7 falls very close to the position of #3, § withinorclosetotheX-rayerrorcirclearereportedinTab.4.No whichpromptedusto usePSF photometryratherthanaperture other objecthas beendetected down to the 3σ limiting magni- photometryasdone,instead,fortheotherNACO observations. tudesof21.54and21.00forHandK,respectively.Allobjects The magnitudes of all the detected objects are listed in Tab. ′ havebeendetectedinboththeHandK bands,exceptfor#9and 5.No other objectis detectedclose or within the radio position ′ #10.Fig.5showsthepositionoftheseobjectsinthe(K,H-K) error circle down to a 3σ limiting magnitude of K = 22.60 ′ ′ s colour-magnitude diagram compared to other objects detected and 22.25 for the March 31 and for the July 20 observations, inthefield.Wenotethatobject#19ofMereghettietal.(2001), respectively.Wenotethatamongournewlydetectedobjects,#7 originally considered a potential candidate on the basis of the (K 19.7)istheonlynewsourcewithinthe99%errorcircle, s ∼ Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 7 Fig.5. Left panel: 1E1841–045: CFHT K-band image of field, with the 1.2 radius (99% confidence level) X-ray error circle ′ ′′ overlaied.Objectsdetectedat 3σare labeled.Object#19of Mereghettietal. (2001)is labeledasa reference.Rightpanel:K , ′ ≥ H-K CMDfor1E1841–045 field.ObjectsdetectedwithinorclosetotheX-rayerrorcircle(left)areshowninredandlabeled. ′ anditdisplayssomeevidenceofvariability(∆K =0.47 0.11) with respectto the X/IR ratio thatan AXP orSGR is expected s ± between the two epochs. Unfortunately,its proximity to object tohave,andiii)withtheexceptionoftheirIRvariability,theIR #3 prevents a better determination of the actual magnitude counterparttotheseneutronstarsare,atourcurrentknowledge, variation, even through PSF photometry. As we did in 3.1, consistentwiththeIRemissionofverylowmassstars,whichif § we evaluated the effects of random errors on our relative layinginthesourceerrorcirclewillthenresultinthesameX/IR photometry by measuring the magnitude scatter of a number ratio.Thisiswhyaslongassignificantvariabilityisnotdetected, oftest starsofbrightnesscomparableto object#7andselected ifmorethanonefaintIRobjectispresentinthepositionalcircle, close to the field center in order not to introduce biases in our anyproposedIRconterpartneedsafurtherconfirmation. PSFphotometryduetotheposition-dependentNACOPSF. We foundthatfortheteststarsthescatteris 0.1magnitudes.Even ≈ 4.1.XTEJ1810–197 IRvariability adding the random error we found that the measured variation is still significant, although only at the 3σ level. Based on IR observations with the VLT–NACO camera of XTEJ1810– ∼ variability,wethenproposeobject#7asatentativeIRcandidate 197 performed after the outburst, unveiled its IR counter- counterparttoSGR1900+14. part (Israeletal. 2004b). Follow-up observations performed six months later detected IR variability from the candidate counterpart(Reaetal.2004),withafluxdecreasebyaboutafac- 4. Discussion torof2.SimultaneousX–rayobservationsoverthesameperiod Beforediscussingourresults,wewouldliketoclarifyherehow revealeda similar X-ray flux decrease between the two epochs wedefineareliableidentificationofanIRcounterparttoanAXP (Rea et al. 2004; Gotthelf et al. 2004). This suggested that the orSGR. Allofthe currentwelldefinedcounterpartshavebeen correlatedIR/X-rayvariabilityisacharacteristicofthissource, establishedeitherthroughaveryaccuratepositionalcoincidence asexpectediftheIRemissionweredominatedbythereprocess- (noothercandidatesinthepositionalerrorcircle)orbythede- ingoftheX-rayemissionfromafossildiskaroundthemagnetar tectionofsignificantIRvariability.Insomecases,whatareusu- (Perna et al. 2000;Perna & Hernquist2000).However,our re- allydefined(notveryclearly)as”strange”colourswithrespect centIRobservationsshowthatsuchcorrelationisnotasobvious totheotherstarsinthefieldofviewhasbeenusedasamethod as thoughtbefore (see also Camilo et al. 2007b).While the IR toclaimthedetectionofa counterpart.However,weprefernot flux is indeed variable, it does not follow the same monotonic toadoptthis”colour”criterionbecausewefinditrathermislead- decrease of the X-ray flux observed in the post-outburstphase ing.Infact,i)almostallcounterpartsreliablyidentifiedthrough (see Fig.2 and Gotthelf & Halpern 2007). Instead, the IR vari- IR variability detection or positional coincidence have colours ability appears more similar to the one observed in radio band not much at variance with some of the their field stars, and ii) (Camiloetal.2006,2007c).Unfortunately,theIRobservations ourcurrentlackofknowledgeofthetheexactphysicsbehindthe are too sparse for any firm conclusion to be drawn regarding IR emission of AXPs and SGRs does not allow any prediction a possible connection between the radio and the IR variabil- on theirIR spectra.Anothercriterion,oftenused to proposeor ity. Furthermore, we note that the onset of the radio emission strengthenanidentificationistheX/IRfluxratio.Webelievethis (January2005;Camiloetal.2006)occurredafterourIRobser- methodratherroughandnotconstrainingatallforthefollowing vations. reasons: i) AXPs and SGRs are variable both in the X-ray and Inthefossildiskscenario,theXTEJ1810–197 IRvariabil- IRbands,andwedonothaveyetaclearideaontheconnection ity can be hardly explainedif the IR emission were dominated betweenthesetwobands,ii)therearenotheoreticalpredictions by reprocessing of the X-ray radiation from the pulsar. In this 8 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars Id. RA(◦) DEC(◦) H K′ Id. RA(◦) DEC(◦) K(2006.03.31) K(2006.07.20) 1 280.330811 -4.936699 20.70 0.31 19.53 0.15 1 286.81000 9.32198 19.36 0.09 19.39 0.06 2 280.330444 -4.936720 20.62±0.29 20.95±0.49 2 286.80949 9.32190 18.14±0.07 18.05±0.05 3 280.330597 -4.936695 19.58±0.10 18.19±0.03 3 286.80957 9.32212 17.31±0.07 17.25±0.05 4 280.330658 -4.936555 18.48±0.03 18.56±0.05 4 286.80948 9.32216 18.42±0.07 18.48±0.05 5 280.330475 -4.936538 19.01±0.06 19.86±0.18 5 286.80989 9.32189 20.10±0.11 19.96±0.08 6 280.330627 -4.936442 19.42±0.08 19.24±0.11 6 286.80974 9.32220 20.73±0.26 20.85±0.22 7 280.330933 -4.936414 17.88±0.02 17.07±0.02 7 286.80963 9.32215 19.21±0.08 19.68±0.08 8 280.330353 -4.936280 20.69±0.29 18.31±0.04 8 286.80997 9.32206 20.86±0.19 20.55±0.13 9 280.330760 -4.936186 >21±.54 20.47±0.28 9 286.81016 9.32210 20.59±0.20 20.46±0.15 10 280.330670 -4.936334 >21.54 19.93±0.16 ± ± ± Table 5. SGR1900+14: K magnitudes of the March and July Table4.1E1841–045:HandK′ magnitudesforalltheobjects 2006 observations,for all objects within 1′.′6 from the position detectedwithin1′.′1ofthesourceX–rayposition.Magnitudeer- (see text for details). Errorson the magnitudesare givenat 1σ rorsaregivenat1σconfidencelevel. confidencelevel. Note that the magnitudes of both the previous candidates caseinfacttheIRfluxshoulddecreaseastheX-rayfluxdrops (our#1and#3;Israeletal.2003,Safi-Harb&West2005,Durant (see Rea et al. 2004 for details), and no increase in any of the & van Kerkwijk 2006) would make 1RXSJ1708–4009 much IR bandswould be expected. If, on the other hand,the IR flux brighterintheIRbandthananyotherAXP(seee.g.Israeletal. were dominated by the disk emissivity resulting from viscous 2004a). Furthermore,all the other fainter candidates we report dissipation (which could be the case if the disc inner radius is here,layingwithinthesourcepositionalerrors(seeFig.3),have fartherout),thenfluxvariabilitywouldbetheresultofavaria- brightness in better agreementwith the IR emission properties tioninthediskmassinflowrate.Giventhenon-correlatedIRand of AXPs than both objects #1 and #3. Durant&vanKerkwijk X-rayvariabilityweobserve,thislatterscenariocouldbemore (2006)proposedsource#3asa possiblecandidateonthebasis viable.However,whileavariabilityintheIRemissioniseasily ofIRvariability,thestrangecoloursandtheresultingX/IRratio predictable as resulting from a mass inflow rate variation, this of this object. We believe that given the large number of faint shouldtakeplacesimilarlyinalltheIRbands.Asnotedin 3.1, IR sources laying within the 1RXSJ1708–4009 positional er- § weseeahintforanIR-banddependentvariability,althoughno rorcircle,andtheratherbrightmagnitudesofsource#3,neither significantconclusioncanbedrawnonthisaspectyet. the 2.5σIRvariability,noritscolours(seeFig.4)oritsX/IR Within themagnetarsscenario,a fewdifferentmodelshave ratio∼can reliablytightthis objectto the AXP (see also above). beenproposedfortheproductionoftheIRradiation(e.g.Heyl We would like to stress that we can not exclude any of these & Hernquist2005;Beloborodov& Thompson2007),but none candidates,butthatafirmidentificationofanIRcounterpartto of them makes (yet) specific predictions for the IR variability 1RXSJ1708–4009 isfarfrombeingasettledissue. thatwouldpermitadirectcomparisonwithourdata. This AXP has been recently observed to have a vari- IR variability has been observed for other magnetars, able X-ray emission (Rea et al. 2005; 2007b; Campana et i.e. 1E1048.1–5937 (Israel et al. 2002; Tam et al. 2007), al. 2007), which, if correlated somehow with the IR as for 4U0142+614(Hulleman et al. 2004), SGR1806–20 (Israel et 1E2259+586(Tam et al. 2004), would imply a variable IR al. 2005), and 1E2259+586 (Tam et al. 2004). Unfortunately, counterpart. In particular, the X-ray variability observed for withthecurrentdatanofirmconclusioncanbedrawnyetabout 1RXSJ1708–4009 in X-ray observations sparse over several theIRvariabilityinconnectionwithvariabilitiesinotherobserv- years was of the order of 50%, which would roughlyimply (if ingbands,exceptfor1E2259+586 andSGR1806–20 (Tamet X-rayandIR are directlycorrelated)anIR variabilityof 0.5 ∼ al.2004;Israeletal2005),whereacorrelationwiththesource magnitudes,notdetectedfromanyoftheobjectsneartheposi- burstingactivityandX-rayfluxenhancementhasbeendetected. tion of 1RXSJ1708–4009. However, the lack of simultaneous IRandX-rayobservationsofthissourcepreventusfromdraw- inganyfirmconclusionsaboutpossiblecorrelatedvariability. 4.2.Othersources:1E1841–045,SGR1900+14 and 1RXSJ1708–4009 5. Conclusions NoIRcounterparthadbeenclaimedsofarfor1E1841–045 and SGR1900+14, likely because of the crowding of the field in TheIRcounterparttoXTEJ1810–197 isconfirmedtobevari- whichthesemagnetarshappenstolay.Bycomparingourphoto- able in time (as previosuly proposed by Rea et al. (2004)). metricresultswiththosereportedinliteraturefor1E1841–045, However,despitepreviousclaims,itsvariabilitymightbemore it is evident that there is one object, our source #9 (source B similartoitsradiobehaviorthantoitsX–rayvariability,which in Durant (2005)), with magnitudes similar to the other AXP appears to decay smoothly after the outburst. Simultaneous IR counterparts,andshowinga>4σvariability.Furthermore,com- andradioobservationsofthisAXPareneededtoreliablyassess paringourtwoobservationsofSGR1900+14 wefounda 3σ thispossibility. ∼ variability in source #7 (not detected by Kaplan et al. (2002) To date, IR counterparts to AXPs and SGRs have been becauseitwastoofaintfortheirlimitingmagnitudes).Wethere- confirmed only for XTEJ1810–197 (Israel et al. 2004b), foreconsider these objectsas possible IR counterpartsto these 1E1048.1–5937(Wang&Chakrabarty2002;Israeletal.2002), twoneutronstars. 1E2259+586(Hulleman et al. 2001), 4U0142+614(Hulleman The new deep observation of 1RXSJ1708–4009 show 20 et al. 2000)and SGR1806–20(Israel et al. 2005;Kosugi et al. sourceswithinandinthevicinityofthe99%X-rayerrorcircle 2005).FortheAXP1E1841–045 andSGR1900+14,wepro- (seeFig.3).Noneoftheseobjectsshowedsignificantvariability poseheretwo possible candidatesbased onthedetectionofIR withrespecttootherobservationsintheliterature,whichcould variability.ForalltheremainingAXPsandSGRswe stillmiss helpidentifyingthemasreliablemagnetarcounterparts. acandidateorconfirmedIRcounterpart(seee.g.Wachteretal. Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 9 Duncan,R.C.,&Thompson,C.1992,ApJ,392,L9 Durant,M.,2005,ApJ,632,563 Durant,M.&vanKerkwijk,2005,ApJ,628,L135 Durant,M.&vanKerkwijk,2006,ApJ,648,534 Durant,M.&vanKerkwijk,2007,ApJinpress(arXiv:0711.3985) Frail,D.A.,Kulkarni,S.R.,Bloom,J.S.1999,Nature,398,127 Gelfand,J.D.&Gaensler,B.M.2007,ApJ,667,1111 Gotthelf,E.V.,Halpern,J.P.,Buxton,M.&Bailyn,C.,2004,ApJ,605,368 Gotthelf,E.V.&Halpern,J.P.,Ap&SS,308,79 Halpern,J.,Gotthelf,E.V.,Becker,R.H,etal.,2005,ApJ,632,L29 Heyl,J.S.&Hernquist,L.2005,MNRAS,362,777 Hulleman,F.,vanKerkwijk,M.H.&Kulkarni,S.R.2000,Nature,408,68 Hulleman,F.,Tennant,A.F.,vanKerkwijk,M.H.,etal.2001,ApJ,563,L49 Hulleman,F.,vanKerkwijk,M.H.&Kulkarni,S.R.2004,A&A,416,1037 Israel,G.L.,Covino,S.,Stella,L.,etal.2002,ApJ,580,L143 Israel,G.L.,Covino,S.,Perna,R.,etal.2003,ApJ,589,L93 Israel,G.L.,Stella,L.,Covino,S.,etal.2004a,IAUSymposium218.Editedby F.Camilo&B.M.Gaensler;p.247(astro-ph/0310482) Israel,G.L.,Rea,N.,Mangano,V.,etal.2004b,ApJL,603,L97 Israel,G.L.,Covino,S.,Mignani,R.,etal.2005,ApJ,438,L1 Fig.6. SGR1900+14. Zoom-inof the area of the K band im- Kaplan,D.L.,Kulkarni,S.R.,Frail,D.A.&vanKerkwijk,M.H.2002,ApJ, s age of 2006 March 31st around the radio position of Frail et 566,378 Kosugi,G.,Ogasawara,R.&Terada,H.,2005,ApJ,623,L125 al. (1999). The error circle (radius 0.81 at 99% confidence ′′ Kuiper,L.,Hermsen,W.&Me´ndez,M.2004,ApJ,613,1173 level) is shown. Two large ghosts are visible atop the position Kuiper,L.,Hermsen,W.,denHartog,P.&Collmar,W.2006,ApJ,645,556 ofSGR1900+14,hidingstarEofKaplanetal.(2002). Leggett,S.K.,Hawarden,T.G.,Currie,J.M.,etal.2003,MNRAS,345,144 Mereghetti,S.,Mignani,R.,Covino,S.,etal.2001,MNRAS,321,143 Muno,M.P.,etal.2006,ApJ,636,L41 2004; Durant & van Kerkwijk 2005, 2007; Muno et al. 2006; Perna,R.,Hernquist,L.&Narayan,R.2000,ApJ,541,344 Perna,R.&Hernquist,L.2000,ApJ,544,L57 Gelfand&Gaensler2007). Persson,S.E.,Murphy,D.C.,Krzeminski,W.,etal.1998,AJ,116,2475 Ours and others results on optical and IR observations of Rea,N.,Testa,V.,Israel,G.L.,etal.2004,A&A,425,L5 AXPs and SGRs, show that up to now we are far from having Rea,N.,Oosterbroek,T,Zane,S.,etal.2005,MNRAS,361,710 anoverallpictureoftheopticalandIRbehavioroftheseneutron Rea,N.,Turolla,R.,Zane,S.,etal.2007a,ApJ,661,L65 stars. Further observations, possibly simultaneously on a wide Rea,N.,Israel,G.L.,Oosterbroek,T.,etal.2007b,Ap&SS,308,505 Safi–Harb,S.&West,J.2005,AdSpR,35,1172 energyrange,areneededtorefinecurrenttheoreticalmodelsand Skrutskie,M.,etal.2006,AJ,131,1163 shedlightonAXPs’andSGRs’opticalandIRemissionmecha- Stetson,P.B.1992,ASPConf.Ser. 25:AstronomicalDataAnalysisSoftware nisms. andSystemsI,25,297 Tam,C.R.,Kaspi,V.M.,vanKerkwijk,M.H.,Durant,M.2004,ApJ,617,L53 Acknowledgements. BasedonobservationscollectedattheEuropeanSouthern Tam,C.R.,Gavriil,F.P.,Dib,R.,Kaspi,V.M.,Woods,P.M.,Bassa,C.,ApJin Observatory, Paranal, Chile, under program ID 071.D-0503, 072.D-0297, press(arXiv:0707.2093) 073.D-0381,076.D-0383,277.D-5005,attheCanada-France-HawaiiTelescope Thompson,C.,&Duncan,R.C.1995,MNRAS,275,255 (CFHT) which is operated by the National Research Council of Canada, Wachter,S.,Patel,S.,Kouveliotou,C.,etal.2004,ApJ,615,887 the Institut National des Sciences de l’Univers of the Centre National de la Wang,Z.&Chakrabarty,D.2002,ApJ,579,L33 Recherche Scientifique of France, and the University of Hawaii, and at the Wang,Z.,Chakrabarty,D.&Kaplan,D.L.2006,Nature,440,772 GeminiObservatory, whichisoperated bytheAssociation ofUniversities for Woods,P.&Thompson,C.2006,CompactstellarX-raysources.EditedbyW. Research inAstronomyInc.,underacooperative agreement withtheNSFon Lewin&M.vanderKlis.CambridgeUniversityPress(astro-ph/0406133) behalf of the Gemini partnership: the National Science Foundation (United States), the Science and Technology Facilities Council (United Kingdom), the National Research Council (Canada), CONICYT (Chile), the Australian ResearchCouncil(Australia),CNPq(Brazil)andCONICET(Argentina).This publication makesuseofdataproductsfromtheTwoMicronAllSkySurvey, which is a joint project of the University of Massachusetts and the Infrared Processing andAnalysis Center/California Institute ofTechnology, funded by the National Aeronautics and Space Administration and the National Science Foundation.WethanktheYepun(UT4)andNACOteamfortheirconstanthelp ontheobservationoptimizationduringthe3yearscoveredbythisprogram,and therefereeforhis/herverycarefulreadingofthepaperandconstructivereport. VTandGLIacknowledgesupportfromMIURfunds.NRissupportedbyNWO VeniFellowship.SZacknowledgesSTFC(ex-PPARC)forsupportthroughan AdvancedFellowship. References Baring,M.G.&Harding,A.K.1998,ApJ,507,L55 Beloborodov,A.&Thompson,C.2007,ApJ,657,967 Bertin,E.&Arnouts,S.1996,A&AS,117,393 Burgay,M.,Rea,N.,Israel,G.L.,etal.2006,MNRAS,372,410 Burgay,M.,Rea,N.,Israel,G.L..etal.2007,Astronomers’Telegram#903 Camilo,F.,Ransom,S.,Halpern,J.,etal.2006,Nature,442,892 Camilo,F.,Ransom,S.M.,Halpern,J.P.,Reynolds,J.2007a,ApJ,666,L93 Camilo, F., Ransom, S. M., Penalver, J., et al. 2007b, ApJ in press (arXiv:0705.4095) Camilo,F.,Cognard,I.,Ransom,S.,etal.2007c,ApJ,663,497 Campana,S.,Rea,N.,Israel,G.L.,etal.2007,A&A,463,1047 Chatterjee,P.,Hernquist,L.&Narayan,R.,2000,ApJ,534,

ebook img

Adaptive optics near-infrared observations of magnetars PDF

0.35 MB·English
by  V. Testa
#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 Adaptive optics near-infrared observations of magnetars

Astronomy&Astrophysicsmanuscriptno.AXP˙obs˙astropacca c ESO2008 (cid:13) February2,2008 Adaptive optics near-infrared observations of magnetars V.Testa1,N.Rea2,3,R.P.Mignani4,G.L.Israel1,R.Perna5,S.Chaty6,L.Stella1,S.Covino7, R.Turolla8,4,S.Zane4,G.LoCurto9,S.Campana7,G.Marconi9,S.Mereghetti10 1 INAF-OsservatorioAstronomicodiRoma,ViaFrascati33,00040MontePorzioCatone,Italy 2 UniversityofAmsterdam,AstronomicalInstitute“AntonPannekoek”,Kruislaan403,1098SJ,TheNetherlands 8 3 SRONNetherlandsInstituteforSpaceResearch,Sorbonnelaan,2,3584CA,Utrecht,TheNetherlands 0 4 MSSL,UniversityCollegeLondon,HolmburySt.Mary,DorkingSurreyRH56NT,UK 0 5 JILAandDepartmentofAstrophysicalandPlanetarySciences,UniversityofColorado,440UCB,Boulder,CO,80309,USA 2 6 LaboratoireAIM,CEA/DSM–CNRS–Universite´ ParisDiderot,DAPNIA/Serviced’Astrophysique,Baˆt.709,CEA–Saclay,FR- 91191Gif-sur-Yvette,France n 7 INAF-OsservatorioAstronomicodiBrera,viaBianchi46,23807Merate(Lc),Italy a J 8 UniversityofPadua,PhysicsDepartment,viaMarzolo8,35131,Padova,Italy 9 ESO-EuropeanSouthernObservatory,AlonsodeCordova3107,Vitacura,Santiago,Chile 4 10 INAF-IstitutodiAstrofisicaSpazialeeFisicaCosmica‘G.Occhialini’,viaBassini15,20133,Milano,Italy 2 Received/Accepted ] h ABSTRACT p - o Context.Wereportonnear-infrared(IR)observationsofthethreeanomalousX-raypulsarsXTEJ1810–197,1RXSJ1708–4009and r 1E1841–045,andthesoftgamma-rayrepeaterSGR1900+14,takenwiththeESO-VLT,theGemini,andtheCFHTtelescopes. st Aims.This work is aimed at identifying and/or confirming the IR counterparts of these magnetars, as well as at measuring their a possibleIRvariability. [ Methods.InordertoperformphotometryofobjectsasfaintasK 20,wehaveuseddatatakenwiththelargesttelescopes,equipped s∼ withthemostadvancedIRdetectorsandinmostofthecaseswithAdaptiveOpticsdevices.Thelatterarecriticaltoachievethesharp 2 spatialaccuracyrequiredtopinpointfaintobjectsincrowdedfields. v Results.We confirm with high confidence the identification of the IR counterpart to XTEJ1810–197, and its IR variability. For 1 1E1841–045 andSGR1900+14 weproposetwocandidateIRcounterpartsbasedonthedetectionofIRvariability.For1RXSJ1708– 7 4009 weshowthatnoneofthepotentialcounterpartswithinthesourceX-rayerrorcirclecanbeyetconvincinglyassociatedwith 1 thisAXP. 4 Conclusions.TheIRvariabilityoftheAXPXTEJ1810–197 doesnotfollowthesamemonotonicdecreaseofitspost-outburstX-ray . 2 emission. Instead, the IR variability appears more similar to the one observed in radio band, although simultaneous IR and radio 1 observations are crucial to draw any conclusion in this respect. For 1E1841–045 and SGR1900+14 follow-up observations are 7 neededtoconfirmourproposedcandidateswithhigherconfidence. 0 Keywords.stars:pulsars:individual:XTEJ1810–197,1RXSJ1708–4009,1E1841–045,SGR1900+14 : v i X 1. Introduction power-law(Γ 2–4;fora reviewsee e.g.Woods&Thompson r ∼ a 2006).Their X-ray energyoutputis much larger than the rota- In recent years, the study of isolated neutron stars has become tionalenergylossandimpliesthatthesesourcesarenotrotation- oneofthemostchallengingresearchareasinhigh-energyastro- allypowered,atvariancewithmostyoungisolatedneutronstars. physics,largelyasaresultofthediscoveryofseveralnewclasses Rather,AXPsandSGRsarebelievedtobepoweredbytheneu- ofsourcesbesidestheclassical,well-studiedradiopulsars.The tronstarultra-strongmagneticfield(Duncan&Thompson1992; most extreme objects are the so-called “magnetars”. This class Thompson & Duncan 1995). In the ”magnetar” model, crustal comprises the Anomalous X–ray Pulsars (AXPs) and the Soft deformations,drivenbymagneticstressesimpartedtothecrust Gamma–ray Repeaters (SGRs), observationallyvery similar in bythestronginternaltoroidalmagneticfield,areresponsiblefor manyrespects.TheyallareslowX–raypulsarswithperiodsin theobservedactivity,X/γ-rayburstsandgiantflares. a narrow range (P= 2–12s), relatively large period derivatives (P˙ = 10 13 10 10ss 1), spin-down ages of 103 104yr, and Until not long ago, the persistent emission of AXPs and − − − magnetic fie−lds, as inferred from the classical ma−gnetic dipole SGRs was detected only in the soft X-ray range. However, in spin-down formula1, of 1014 1015G, larger than the electron the last few yearsthe availability of more sensitive optical and quantumcritical field (B −4.4 1013G) abovewhich quan- IRtelescopes,aswellasofγ-raysatellites,openednewwindows cr tum effects become crucial.≃AXPs×and SGRs are strong X-ray inthestudyofAXPsandSGRs,unveilingtheirmulti-bandprop- emitters, with X-ray luminosities of about 1034 1036ergs 1. erties.Inparticular,observationsfromESO,Gemini,CFHTand − Their0.1–10keVpersistentemissionhasrelativel−ysoftspectra Keck (see Israel et al. 2004a for a review) led to the discovery usuallymodeledbyanabsorbedblackbody(kT 0.2–0.6)plusa of faint (Ks 20), and in some cases variable, IR counterparts ∼ to five out o∼f seven confirmed AXPs (one of which was also Sendoffprintrequeststo:V.Testa:[email protected] detected in the opticalband by Hulleman et al. (2000)),and to 1 B 3 1019√PP˙ G one(outoffour)confirmedSGR(Israeletal.2005,Kosugietal. ∼ × 2 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars Telescope/Instrument Target Date(YY.MM.DD) MJD Filter ExposureTime(s) FWHM(′′) airmass VLT/NACO XTEJ1810–197 2003.10.08 52920 H 2520 0.189 1.17 2003.10.08 52920 Ks 2520 0.243 1.49 2004.03.14 53078 H 1080 0.108 1.47 2004.03.14 53078 Ks 1200 0.094 1.20 2004.03.14 53078 J 1080 0.081 1.13 2004.09.09 53257 J 1800 0.170 1.01 2004.09.09 53257 H 3600 0.140 1.08 2004.09.13 53261 Ks 3600 0.099 1.01 1RXSJ1708–4009 2003.05.20 52779 Ks 2400 0.083 1.04 2003.06.19 52809 H 2400 0.122 1.40 2003.06.20 52810 J 2400 0.122 1.04 2004.03.25 53089 L’ 2100 0.119 1.04 SGR1900+14 2006.04.01 53826 Ks 1560 0.125 1.43 2006.07.21 53937 Ks 2040 0.111 1.21 Gemini/NIRI XTEJ1810–197 2003.09.18 52900 Ks 285 0.397 1.62 2003.09.18 52900 H 285 0.397 1.99 2003.09.18 52900 J 285 0.397 1.78 2004.06.10 53166 Ks 510 0.631 1.32 2004.06.10 53166 H 510 0.526 1.30 2004.06.10 53166 J 510 0.655 1.32 2004.07.28 53214 Ks 510 0.468 1.39 2004.07.28 53214 H 510 0.526 1.53 CFHT/AOBIR 1E1841–045 2002.08.17 52503 K’ 1260 0.137 1.11 2002.08.17 52503 H 1260 0.161 1.31 Table1.Observationssummary. Fig.1.XTEJ1810–197:Leftpanel:VLT/NACOK-bandimage(March2004).Theradioposition(Camiloetal.2007c)ismarked togetherwithitscandidatecounterpart(objectA).Twelvenearbyobjects,someofwhichusedasareferenceforrelativephotometry, arealsomarkedinthefigure.Rightpanel:K vs.H-K colour-magnitudediagramofthesourcesdetectedintheNACOfieldofview S S (filledcircles).ObjectA ismarkedwitharectangleindicatingtherangeincolourandmagnitudecoveredthroughouttheepochs, whilesomeofthenearbyobjectsmarkedinthemapandlistedinTab.2areplottedinredandlabeled.Magnitudesandcoloursfor allthesourcesbutobject’A’havebeenaveragedoveralltheobservations. 2005). Furthermore, deep INTEGRAL observations in the 20– ofthesesourcesisconcerned.Theextrapolationofthecanonical 200keV band revealed that most of these highly magnetized blackbodyplus power-law,used to model the soft X-ray emis- sourcesarealsohardX-rayemitters(Kuiperetal.2004,2006). sion of these sources, largelyover-predictstheir opticalandIR Therecentdiscoveryofpulsedradioemissionfromthetransient emission. Recently, new attempts have been made at modeling magnetars XTEJ1810–197 (Halpernetal. 2005, Camiloetal. the multi-bandspectra of magnetarsto overcomethis problem. 2006)and1E1547.0 5408(Camiloetal.2007a),theonlycases In particular, Reaetal. (2007a) showed that AXP spectra can − sofar(Burgayetal.2006,2007),isintriguingsinceradioemis- be fitted by a resonant cyclotron scattering model or two log- sion was believed to be quenched in magnetic fields above the parabolicfunctions,whichdonotover-predicttheirdimoptical quantumcriticallimit(Baring&Harding1998). and IR magnitudes. However, despite solving this issue, these newspectralmodelsarestillfarfromgivinganyphysicalinter- Despite the more complete observational picture we have pretationfortheopticalandIRemission. nowforAXPsandSGRs,thephysicalprocessesatthebasisof theiremissioninthedifferentbandsarenotfullyunderstoodas Some possibilities are that either the magnetars opti- yet.ThisisparticularlytrueasfarastheopticalandIRemission cal/IR emission arise from the star surface/magnetosphere Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 3 (Beloborodov& Thompson2006),or fromthe reprocessingof the X-ray emission via a fossil disk around the neutron star (Chatterjee et al. 2000, Perna et al. 2000). Indeed, the Spitzer mid-IR detection of the AXP 4U0142+614(Wangetal. 2006) hasbeeninterpretedasthefirstevidenceofapassivefossildisk aroundanAXP,believedtoemitthroughreprocessingoftheX- rayradiationfromthemagnetar. In this paper we report on Gemini and ESO-VLT observa- tionsofthetransientAXPXTEJ1810–197,onESO-VLTobser- vations of the AXP 1RXSJ1708–4009 and of SGR1900+14, andCFHTobservationsoftheAXP1E1841–045( 2).There- § sultsarepresentedanddiscussedinthecontextofdifferentemis- sionscenarios,andcomparedwiththeopticalandIRproperties ofothermagnetars(see 4). § 2. Observationsanddatareduction 2.1.Observations Observationswereperformedwiththe8mVeryLargeTelescope (VLT)andthe3.5mCanada-France-HawaiiTelescope(CFHT), bothequippedwithadaptiveoptics(AO)andnear-IRcameras.In addition,weusedarchivalobservationsperformedwiththe8m Gemini telescope, also equipped with an IR camera but with- outAO, thatwere offeredstartingfrom2004only.A summary of the telescopesand instrumentsused, as well as the observa- tionslogs,isreportedinTab.1.Inallcasesimageswereobtained throughthestandardobservingtechniquecommonlyappliedto IR arrays, i.e. by stacking repeated exposuresalong each node of a pre-defined dithering pattern, with each node jittered by a few arcsec with respectto the previousone. Thisallowsto ob- Fig.2.XTEJ1810–197:Toppanel:X-rayflux(showninunitsof tain a set of dithered exposures that are used to produce sky- 10 11ergcm 2s 1)historyinthe0.5–10k˙eVenergyrange(from − − − backgroundimages. Gotthelf & Halpern 2007). Lower panels: from top to bottom: VLT/YepunobservationsofXTEJ1810–197,1RXSJ1708– K ,H,andJbandrelativemagnitudevariationforXTEJ1810– s 4009 andSGR1900+14 werecarriedoutindifferentobserving 197 counterpart(red star) and for two nearby, bright compari- runs (see Tab. 1) using NAos COnica (NACO)2 with the S27 son objects. Magnitude variations are computed relative to the camera (0′.′027 pixel size, 28′′ 28′′ field of view). Gemini- September2004observation.Tobetterfitintheplot,magnitude × North observations of XTEJ1810–197 were carried out using differencesforobjects#2and#8havebeenshiftedby+1.1and- the Near-InfraRed Imager (NIRI)3 with the f/6 camera (0′.′117 1.1mag,respectively.BottompanelshowstheH-Kscolourvari- pixel scale, 120′′ 120′′ field of view). 1E1841–045 was ob- ation.Magnitudeerrorsat(1σ)areshown.Theseincludeinter- served at the CFH×T using the AdaptiveOptics Bonnet(AOB)4 nalerrors,aperturecorrectionerrorsandzeropointerrors. InfraRedcamera(0.035pixelsize,36 36 fieldofview). ′′ ′′ ′′ × 2.2.Datareductionandcalibration in the VLT-NACO images the PSF was found to be variable acrossthefieldofview,whileintheCFHT-AOBimagesitwas TheVLTdatawerepre-reducedwiththeESONACOpipeline5, found to feature an unusual profile with a clearly visible sec- whichisbasedonthepackageeclipse,whilethenativeversion ondarydiffractionring.Botheffectsaredifficultto accountfor, oftheeclipse6packagewasusedfortheGeminiandCFHTdata. even by using more sophisticated and position-dependentPSF Inallcases,theimagepre-reductionproducedafinalco-added, models.Ontheotherhand,intheGemini-NIRIimagesthePSF sky-subtractedandflat-fieldedimageforeachband. was found to be more stable, although worse sampled due to Dependingonthecase,fluxeswerecomputedeitherthrough thelargerpixelsize.Forthesereasons,wedecidedasageneral PSF or aperture photometry using the IRAF7 version of the strategytouse aperturephotometryforourAO images,i.e. the daophotpackage(Stetson1992).ForAOobservationswefound NACOandAOBones,andtousePSFphotometryfortheNIRI someinstrument-dependenteffectswhichhamperedthePSFfit- images.Fortheformer,aperturephotometrywasperformedus- ting and discouraged the use of PSF photometry.In particular, ing an aperture or diameter twice the measured PSF. Aperture andPSFphotometrieshavebeencomparedbyapplyingaperture 2 www.eso.org/instruments/naco correctionsusingtheIRAFdaogrowalgorithm,whichcalculates 3 www.gemini.edu/sciops/instruments/niri the aperturecorrectionextrapolatingto anideally infinite aper- 4 www.cfht.hawaii.edu/Instruments ture. 5 www.eso.org/instruments/naco/ 6 www.eso.org/eclipse For the VLT images the photometric calibration was per- 7 IRAF is distributed by the National Optical Astronomy formed by observing standard stars from the Perssonetal. Observatories, which are operated by the Association of Universities (1998) catalog, usually observed under similar conditions(and forResearchinAstronomy,Inc.,undercooperativeagreementwiththe ataboutthesameepoch)asourtargets.ForalltheNACOruns NationalScienceFoundation. thezeropointuncertaintyhasbeenfoundtobeverysmall,i.e.in 4 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars SourceId. R.A.(2000) Dec.(2000) 52900 52920 53078 53166 53214 53257 53261 Ks A 272.46283 -19.73110 20.81 0.08 20.82 0.09 21.18 0.05 21.21 0.10 21.04 0.09 1 272.46390 -19.73016 13.87±0.07 13.85±0.04 14.26±0.04 14.01±0.04 14.01±0.08 2 272.46391 -19.73061 17.23±0.07 17.25±0.04 17.23±0.04 17.36±0.04 17.30±0.08 3 272.46338 -19.73070 19.09±0.07 19.21±0.05 19.47±0.04 19.15±0.05 19.41±0.08 4 272.46335 -19.73110 19.43±0.07 19.37±0.05 19.46±0.04 19.42±0.04 19.47±0.08 5 272.46399 -19.73098 19.52±0.07 19.77±0.06 19.67±0.04 19.62±0.04 19.70±0.08 6 272.46384 -19.73145 18.61±0.07 18.61±0.04 18.60±0.04 18.66±0.04 18.67±0.08 7 272.46327 -19.72952 17.69±0.07 17.70±0.04 17.73±0.04 17.75±0.04 17.75±0.08 8 272.46535 -19.72897 15.48±0.07 15.40±0.04 15.45±0.04 15.51±0.04 15.53±0.08 9 272.46264 -19.72882 15.25±0.07 15.15±0.04 15.23±0.04 15.27±0.04 15.23±0.08 10 272.46082 -19.73013 14.43±0.07 14.19±0.04 14.55±0.04 14.44±0.04 14.29±0.08 11 272.46430 -19.73252 14.83±0.07 14.74±0.04 14.91±0.04 14.87±0.04 14.90±0.08 12 272.46502 -19.73141 16.59±0.07 16.50±0.04 16.51±0.04 16.63±0.04 16.54±0.08 ± ± ± ± ± H A 21.67 0.12 21.81 0.16 22.55 0.07 21.48 0.12 21.73 0.11 22.29 0.18 1 14.27±0.05 14.20±0.04 14.39±0.04 14.24±0.09 14.24±0.08 14.24±0.08 2 17.66±0.05 17.68±0.04 17.66±0.04 17.69±0.09 17.71±0.08 17.67±0.08 3 20.00±0.06 20.11±0.05 20.07±0.04 19.90±0.09 19.78±0.09 20.07±0.08 4 22.24±0.14 22.05±0.11 22.06±0.09 21.66±0.23 21.87±0.10 22.64±0.17 5 22.28±0.28 22.74±0.42 ±- 21.79±0.21 21.68±0.10 22.80±0.11 6 21.17±0.09 21.38±0.07 21.38 0.05 21.11±0.10 21.15±0.09 21.34±0.09 7 18.16±0.06 18.19±0.04 18.14±0.04 18.14±0.09 18.15±0.08 18.20±0.08 8 18.95±0.06 19.00±0.05 18.90±0.04 18.90±0.09 18.93±0.08 18.95±0.08 9 18.42±0.06 18.42±0.04 18.34±0.04 18.44±0.09 18.41±0.08 18.44±0.08 10 14.62±0.05 14.51±0.04 14.65±0.05 14.61±0.09 14.61±0.08 -± 11 15.15±0.05 15.12±0.04 15.19±0.04 15.13±0.09 15.14±0.08 15.12 0.08 12 16.91±0.05 16.89±0.04 16.83±0.04 16.93±0.09 16.91±0.08 16.90±0.08 ± ± ± ± ± ± J A 22.92 0.22 23.89 0.23 23.45 0.12 1 ± 15.37±0.06 15.04±0.04 2 18.85−0.05 18.83±0.06 18.88±0.04 3 21.06±0.06 21.35±0.07 21.35±0.04 7 19.34±0.05 19.39±0.06 19.36±0.04 9 18.06±0.04 18.00±0.06 18.01±0.04 10 ± 15.54±0.06 ± 11 15.98−0.04 16.01±0.06 15.97− 0.04 12 19.04±0.05 18.94±0.06 18.99±0.04 ± ± ± Table 2.XTEJ1810–197:Fromtoptobottom:K ,HandJmagnitudesforthecandidateIRcounterpartofXTEJ1810–197 and s some of the twelve nearby objects, for the 6 differentobservingruns(see Tab.1). Numbersatop the magnitudecolumnsgive the observationsepochsinModifiedJulianDays(MJD).Magnitudeerrorsaregivenat1σandincludezeropointerrors,eitherabsolute orrelative,thatare,forthebrightestobjects,thedominanterrorsource. the0.02-0.05magnituderange.Theresidualuncertaintyintro- totheSeptember2004one.Therelativephotometricuncertainty, ducedbytheextinctiontermisalmostwipedoutbythecloseness basedonabout100starsincommonamongthevariousimages intimeandairmassoftargetfieldsandstandards.Sinceitturns is of the order of 0.05 magnitudes for the NACO-to-NACO outtobe<0.01magnitudesithasbeenneglected. registration,andsli≈ghtlyworsefortheGemini-to-NACOone( ∼ 0.08magnitudes). ≈ The Gemini images of XTEJ1810–197 have been at first FortheCFHT photometry,forwhichnoNACO imagesare instancecalibratedwiththeavailablestandardstars,buttheun- available, the only optionwas to use the 2MASS catalog,after certainty in the zero points and the zero point differences usu- carefully accounting for all the caveats mentioned above. For ally found between standards taken on the same nights sug- theL observationof1RXSJ1708–4009 wehaveusedstandard gested us to discard the primarystandard calibration.A photo- ′ starsfromtheUKIRTIRstandardcatalog(Leggettetal.2003), metric calibration was also tried using stars from the 2MASS yieldingaphotometricaccuracyof 0.1magnitudes. (Skrutskieetal. 2006) catalog identified in the target fields. ∼ However, such an on-the-frame calibration suffered from two In all cases, astrometric calibration was performed using 2MASS stars as reference, yielding an average uncertainty of main problems. Firstly, the numberof suitable 2MASS stars is 0.2, after accounting for the intrinsic astrometric accuracy often quite small and the brighter ones were found to be gen- ′′ ∼ of 2MASS ( 200 mas) and the r.m.s. of the astrometric fits. erallysaturatedinourimages.Furthermore,the2MASScatalog ∼ However,duetothesmallinstrumentpixelscales,thelatterhas wasbuiltfromimageswitha2 pixelsizesothat,inmanycases, ′′ beenfoundtobemuchsmallerthanthe2MASSastrometricac- the measured magnitudescorrespondto blended objectswhich curacyandhenceithasbeenneglected. are, instead, resolved in our higher resolution images. Thus, given the accuracy of the NACO zeropoints and the very low r.m.s. of the photometric solution, we chose to re-calibrate the 3. Dataanalysisandresults Geminiphotometryof the XTEJ1810–197 field on the NACO onebyusing,assecondaryphotometriccalibrators,asetofwell- 3.1.XTEJ1810–197 suited field stars in common between the NACO and the NIRI images. As a reference, we chose the September 2004 NACO Fig.1(left)showstheXTEJ1810–197 field(K band)observed s observationsbecause of their overallbetter quality. In order to withNACOonOctober2003(seealsoIsraeletal.2004b).The ensurefullconsistencyamongobservationsperformedatdiffer- XTEJ1810–197 candidate IR counterpart, labeled A, is coin- ent epochs (Tab. 1), also the NACO October 2003 and March cident with the radio position (RA 18h 09m 51s.087, DEC - 2004photometryoftheXTEJ1810–197 fieldwasre-calibrated 19 43 51.93; error = 0.01; Camilo et al. 2006). For both ◦ ′ ′′ ′′ Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 5 the NACO and NIRI observations, photometry was computed Id RA(◦) DEC(◦) J H Ks andcalibratedasdescribedintheprevioussection.Foreachin- 12 225577..1199555110 --4400..1144789165 2202..9233±00..0197 1189..6804±00..0069 1178..5831±00..0045 strument, the resulting J, H and Ks band catalogues were then 3 257.19560 -40.14799 22.11±0.25 20.01±0.12 18.85±0.05 matched and compared with the multi-band ones. The multi- 4 257.19503 -40.14792 22.03±0.17 20.19±0.09 19.63±0.07 5 257.19498 -40.14800 ± 20.50±0.10 19.95±0.08 bandmagnitudesoftheXTEJ1810–197 candidatecounterpart, 6 257.19495 -40.14800 − ± 20.03±0.09 all calibrated to the NACO system, are reported in Tab.2 to- 7 257.19494 -40.14793 − 20.90−0.11 20.25±0.08 gether with those of a number of field objects, chosen among 8 257.19563 -40.14790 − ± 20.83±0.10 9 257.19502 -40.14786 − 21.55−0.18 20.92±0.11 those used as a referencefor the photometrycross-calibrations 10 257.19531 -40.14795 − ± 20.94±0.11 (seeprevioussection).Wenotethatthecandidatecounterpartis 11 257.19500 -40.14804 − − 20.69±0.10 not detected in the J and Ks-band Gemini observations of the 1132 225577..1199552468 --4400..1144786112 −− −− 2211..2132±±00..1149 June 2004 run which were performedin non-photometriccon- 14 257.19545 -40.14774 − − 21.34±0.21 ditions. Given the short integration time, the overall poor data 15 257.19523 -40.14807 − − 21.52±0.17 16 257.19509 -40.14787 − − 21.58±0.12 quality,andthelackofstandardobservations,wecouldonlyde- 17 257.19510 -40.14799 − − 21.59±0.16 rive tentative estimates on the limiting magnitudes (3σ upper 18 257.19538 -40.14816 − 21.34−0.15 21.66±0.14 limits of J>21.9 and K >20) which are of no use to constrain 19 257.19550 -40.14813 − ± 21.74±0.16 s 20 257.19530 -40.14792 − − 21.88±0.19 thefluxevolutionofoursource. − − ± The time-averaged (Ks, H-Ks) colour-magnitude diagram Table3.1RXSJ1708–4009:J,H,andKsmagnitudesforallthe (CMD) of the field is shown in Fig.1 (right). A well-defined objectsdetectedwithin1′.′1ofthesourceX-rayposition,ordered main sequence group of stars, probably tracking a young star bydecreasingKs magnitude.Noobjecthasbeendetectedinthe cluster, is recognizable in the left part of the diagram, while L′band.Magnitudeerrorsaregivenat1σ,andincludezeropoint a second red clump of stars, most likely red giants or highly uncertainties. absorbed main sequence stars, is present on the right. The XTEJ1810–197 counterpart (green star) appears much fainter than the nearby comparison stars (red dots), and its colour in- deedsuggeststhatitmightbeapeculiarobject. The variability of the XTEJ1810–197 IR counterpart is shown in Fig.2, together with the X-ray variability of the X- ray source (see 4 for details). While in the X-rays the source § features a clear, almost monotonic flux decay, in the IR there is no clear trend in the flux evolution, and hence no correla- tion with the X-rays. Indeed, the flux seems to vary in an er- ratic way in all the threeIR bands.Inparticular,we foundthat the overallvariabilityis apparentlylargerin the H andJ bands (1.07 0.24and0.97 0.32magnitudes,respectively)thaninthe ± ± K one (0.40 0.13 magnitudes). We note that our detailed re- s ± analysis of the October 2003 and March 2004 observations of XTEJ1810–197 yields a K band variability slightly smaller s (0.36 0.10) than that originally reported by Rea et al. (2004) ± usingthesameobservations(0.50 0.10),althoughitisstillcon- ± sistent within the errors.We attribute this (nonsignificant)dif- ference to the more accurate relative photometry between the two epochs, performed using a larger set of secondary photo- Fig.3.1RXSJ1708–4009:NACOKs-bandimageofthesource metric calibrators than the one used in Rea et al. (2004). The fieldwiththe1′.′1radius(99%confidencelevel)X-rayerrorcir- last panelshows the (H-K ) colourvariabilityof the source.In cleoverlaied.Objectsdetectedat 3σarelabeled. s ≥ general,we note that the sourceseems to becomeredderwhen itsfluxdecreases,althoughnoclear trendcanbe recognizedin oursparsephotometry.Weneedtowarn,however,thatanycon- increase the overall uncertaintieson the measured magnitudes. clusion on the source flux and colourvariability must be taken Toquantifytheserandomerrorswehavecross-checkedthepho- with due care. First of all, the apparent lack of correlation be- tometryofanumberofteststarswithbrightnesscomparableto tween the variability in the H and K bands is only suggested theoneofourtarget.Wefoundthatallteststarsshowascatterof s by the July 2004 Gemini Ks band observation. If we take this ∼0.30magnitudesinJandof∼0.10magnitudesintheHandKs observationoutandweconsideronlytheco-evalfluxmeasure- bandsfortheNACO-to-NACOregistration,whilethescatterin- ments,theK bandlightcurvebecomesmoreconsistentwiththe creasesto 0.20magnitudesfortheGemini-to-NACOregistra- s ∼ H(aswellaswiththeJ)bandone.Moregenerally,althoughspe- tion.Hence,wetakethesevaluesasrepresentativeoftherandom cial care was devotedto check the cross-instrumentcalibration errorsofourrelativephotometry.Byaddingthemtotheformal (see 2.2),itispossiblethatourrelativephotometryisaffected errorswe findthatonly the H bandvariability(1.07 0.26)can byra§ndomerrors.Thesemaybeinduced,e.g.byfluctuationsin beconsideredformallysignificant,anditisconsisten±twiththat the atmosphericconditions, sky background,seeing, quality of measured in the J band (0.97 0.43)and, marginally,also with ± theprimarycalibrationframes(darkandflatfields),andglitches theKs bandone(0.40 0.16). ± in the detector performance.Of course, while these effects are marginal for the relatively bright stars that we have used as a 3.2.1RXSJ1708–4009 referencetocomputeourcross-instrumentphotometriccalibra- tion, they are indeed significant for much fainter stars, such as Fig.3 shows the NACO K -band image of the field around the s the XTEJ1810–197 candidate counterpart (K 21), and can Chandrapositionof1RXSJ1708–4009(RA17h08m46s.9,DEC s ∼ 6 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars Fig.4.1RXSJ1708–4009:Colour-magnitudediagramsfortheobjectsinthefieldofview.Sourcesdetectedwithinorclosetothe X-rayerrorcircle(seeFig.3)areshowninredandlabeled. -40 08 52.4).ThesizeoftheX-rayerrorcirclehasbeenesti- ROSATX-rayposition,liesmorethantwoerrorradiiawayfrom ◦ ′ ′′ matedconsideringtheuncertaintyontheChandraposition,that the updated Chandra position. The CMD shows a well-defined is 0.7 (Israeletal. 2003), combined with the overall accuracy mainsequenceextendinguptoK 12,withthecandidatesoc- ′′ ′ ∼ ofourastrometriccalibration(see 2.2),yieldinga finaluncer- cupyingthe fainter regionandstretchingovera broadrangeof § tainty of 1.1 at 99% confidence level. The magnitudes of all colours. ′′ theobjectsdetectedwithintheX-rayerrorcirclearereportedin Bycomparingourphotometricresultswiththosereportedin Tab.3.Nootherobjectisdetecteddowntothe3σlimitingmag- literaturefor1E1841–045 (Durant2005),itisevidentthatour nitudesof22.98,22.20,22.26intheJ,H,K bands,respectively. source #9 (source B in Durant (2005)), shows a >4σ variabil- s IntheL bandnoobjectisdetectedwithintheX-rayerrorcircle ity (∆Ks =1.30 0.24).As a comparison,the average ∆Ks and ′ ± down to 17.8 (the deepest limit ever obtained for an AXPs in ∆HbetweenourphotometryandthatofDurant(2005)for9ob- thisband).Fig.4showsthe(K ,J-K )and(H,H-K )CMDsof jectsinthefieldofviewis 0.2magnitudes.Wethententatively s s s ∼ alltheobjectsdetectedinthefield.ThetwoCMDsshowafairly proposethisobjectastheIRcounterparttothisAXP. scattered sequence.Despite previousclaims (Israelet al. 2003; Safi-Harb&West2005;Durant&vanKerkwijk2006),noneof 3.4.SGR1900+14 these objects show significant (>3σ) IR variability to justify a safe identification as the IR counterpart of 1RXSJ1708–4009 Fig.6 shows the NACO K band image around the s (seealso 4). SGR1900+14 radio position (RA 19h 07m 14s.33, DEC § +09 19 20.1, with a 1 σ uncertainty of 0.15 in each coor- ◦ ′ ′′ ′′ dinate; Frail et al. 1999). Our astrometry yields an overall 3σ 3.3.1E1841–045 positionuncertaintyof0.81,whichcompareswellwiththeone ′′ An updated determination of the 1E1841–045 position of0.79(99%confidencelevel)quotedbyKaplanetal.(2002). ′′ (Wachteretal.2004)wasobtainedwithChandraandgivesRA We notethat the PSFs of field objectsare slightly elongatedin 18h 41m 19s.336, DEC -04 56 10.83, with a 3σ uncertainty the NACO images, and one of the objects detected within the ◦ ′ ′′ radius of 0.9. However, the latter uncertaintyis based only on radioerrorcircle (our#3and objectC of Kaplanet al. (2002)) ′′ five 2MASS reference stars used for the boresight correction looksapparentlyextended.However,its PSF is consistentwith (Wachter et al. 2004). We decided to use a conservative nom- that measured for most field objects. In this case, photometric inal Chandra positional uncertainty of 0.7 at 90% confidence calibrationwasperformedusingasreferencethemagnitudesof ′′ level. Fig.5 shows a revised finding chart of the field around a few objects detected aroundthe targetposition (Kaplanetal. the updated X-ray position. The size of the X–ray error circle, 2002, Frailetal. 1999) (see Fig.6). The three new objects 1.2(99%c.l.)accountsfortheoverallaccuracyofourastromet- detected in our NACO images are also marked (# 7, 8 and 9). ′′ riccalibration(see 2.2).Themagnitudesofallobjectsdetected In particular, object #7 falls very close to the position of #3, § withinorclosetotheX-rayerrorcirclearereportedinTab.4.No whichpromptedusto usePSF photometryratherthanaperture other objecthas beendetected down to the 3σ limiting magni- photometryasdone,instead,fortheotherNACO observations. tudesof21.54and21.00forHandK,respectively.Allobjects The magnitudes of all the detected objects are listed in Tab. ′ havebeendetectedinboththeHandK bands,exceptfor#9and 5.No other objectis detectedclose or within the radio position ′ #10.Fig.5showsthepositionoftheseobjectsinthe(K,H-K) error circle down to a 3σ limiting magnitude of K = 22.60 ′ ′ s colour-magnitude diagram compared to other objects detected and 22.25 for the March 31 and for the July 20 observations, inthefield.Wenotethatobject#19ofMereghettietal.(2001), respectively.Wenotethatamongournewlydetectedobjects,#7 originally considered a potential candidate on the basis of the (K 19.7)istheonlynewsourcewithinthe99%errorcircle, s ∼ Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 7 Fig.5. Left panel: 1E1841–045: CFHT K-band image of field, with the 1.2 radius (99% confidence level) X-ray error circle ′ ′′ overlaied.Objectsdetectedat 3σare labeled.Object#19of Mereghettietal. (2001)is labeledasa reference.Rightpanel:K , ′ ≥ H-K CMDfor1E1841–045 field.ObjectsdetectedwithinorclosetotheX-rayerrorcircle(left)areshowninredandlabeled. ′ anditdisplayssomeevidenceofvariability(∆K =0.47 0.11) with respectto the X/IR ratio thatan AXP orSGR is expected s ± between the two epochs. Unfortunately,its proximity to object tohave,andiii)withtheexceptionoftheirIRvariability,theIR #3 prevents a better determination of the actual magnitude counterparttotheseneutronstarsare,atourcurrentknowledge, variation, even through PSF photometry. As we did in 3.1, consistentwiththeIRemissionofverylowmassstars,whichif § we evaluated the effects of random errors on our relative layinginthesourceerrorcirclewillthenresultinthesameX/IR photometry by measuring the magnitude scatter of a number ratio.Thisiswhyaslongassignificantvariabilityisnotdetected, oftest starsofbrightnesscomparableto object#7andselected ifmorethanonefaintIRobjectispresentinthepositionalcircle, close to the field center in order not to introduce biases in our anyproposedIRconterpartneedsafurtherconfirmation. PSFphotometryduetotheposition-dependentNACOPSF. We foundthatfortheteststarsthescatteris 0.1magnitudes.Even ≈ 4.1.XTEJ1810–197 IRvariability adding the random error we found that the measured variation is still significant, although only at the 3σ level. Based on IR observations with the VLT–NACO camera of XTEJ1810– ∼ variability,wethenproposeobject#7asatentativeIRcandidate 197 performed after the outburst, unveiled its IR counter- counterparttoSGR1900+14. part (Israeletal. 2004b). Follow-up observations performed six months later detected IR variability from the candidate counterpart(Reaetal.2004),withafluxdecreasebyaboutafac- 4. Discussion torof2.SimultaneousX–rayobservationsoverthesameperiod Beforediscussingourresults,wewouldliketoclarifyherehow revealeda similar X-ray flux decrease between the two epochs wedefineareliableidentificationofanIRcounterparttoanAXP (Rea et al. 2004; Gotthelf et al. 2004). This suggested that the orSGR. Allofthe currentwelldefinedcounterpartshavebeen correlatedIR/X-rayvariabilityisacharacteristicofthissource, establishedeitherthroughaveryaccuratepositionalcoincidence asexpectediftheIRemissionweredominatedbythereprocess- (noothercandidatesinthepositionalerrorcircle)orbythede- ingoftheX-rayemissionfromafossildiskaroundthemagnetar tectionofsignificantIRvariability.Insomecases,whatareusu- (Perna et al. 2000;Perna & Hernquist2000).However,our re- allydefined(notveryclearly)as”strange”colourswithrespect centIRobservationsshowthatsuchcorrelationisnotasobvious totheotherstarsinthefieldofviewhasbeenusedasamethod as thoughtbefore (see also Camilo et al. 2007b).While the IR toclaimthedetectionofa counterpart.However,weprefernot flux is indeed variable, it does not follow the same monotonic toadoptthis”colour”criterionbecausewefinditrathermislead- decrease of the X-ray flux observed in the post-outburstphase ing.Infact,i)almostallcounterpartsreliablyidentifiedthrough (see Fig.2 and Gotthelf & Halpern 2007). Instead, the IR vari- IR variability detection or positional coincidence have colours ability appears more similar to the one observed in radio band not much at variance with some of the their field stars, and ii) (Camiloetal.2006,2007c).Unfortunately,theIRobservations ourcurrentlackofknowledgeofthetheexactphysicsbehindthe are too sparse for any firm conclusion to be drawn regarding IR emission of AXPs and SGRs does not allow any prediction a possible connection between the radio and the IR variabil- on theirIR spectra.Anothercriterion,oftenused to proposeor ity. Furthermore, we note that the onset of the radio emission strengthenanidentificationistheX/IRfluxratio.Webelievethis (January2005;Camiloetal.2006)occurredafterourIRobser- methodratherroughandnotconstrainingatallforthefollowing vations. reasons: i) AXPs and SGRs are variable both in the X-ray and Inthefossildiskscenario,theXTEJ1810–197 IRvariabil- IRbands,andwedonothaveyetaclearideaontheconnection ity can be hardly explainedif the IR emission were dominated betweenthesetwobands,ii)therearenotheoreticalpredictions by reprocessing of the X-ray radiation from the pulsar. In this 8 Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars Id. RA(◦) DEC(◦) H K′ Id. RA(◦) DEC(◦) K(2006.03.31) K(2006.07.20) 1 280.330811 -4.936699 20.70 0.31 19.53 0.15 1 286.81000 9.32198 19.36 0.09 19.39 0.06 2 280.330444 -4.936720 20.62±0.29 20.95±0.49 2 286.80949 9.32190 18.14±0.07 18.05±0.05 3 280.330597 -4.936695 19.58±0.10 18.19±0.03 3 286.80957 9.32212 17.31±0.07 17.25±0.05 4 280.330658 -4.936555 18.48±0.03 18.56±0.05 4 286.80948 9.32216 18.42±0.07 18.48±0.05 5 280.330475 -4.936538 19.01±0.06 19.86±0.18 5 286.80989 9.32189 20.10±0.11 19.96±0.08 6 280.330627 -4.936442 19.42±0.08 19.24±0.11 6 286.80974 9.32220 20.73±0.26 20.85±0.22 7 280.330933 -4.936414 17.88±0.02 17.07±0.02 7 286.80963 9.32215 19.21±0.08 19.68±0.08 8 280.330353 -4.936280 20.69±0.29 18.31±0.04 8 286.80997 9.32206 20.86±0.19 20.55±0.13 9 280.330760 -4.936186 >21±.54 20.47±0.28 9 286.81016 9.32210 20.59±0.20 20.46±0.15 10 280.330670 -4.936334 >21.54 19.93±0.16 ± ± ± Table 5. SGR1900+14: K magnitudes of the March and July Table4.1E1841–045:HandK′ magnitudesforalltheobjects 2006 observations,for all objects within 1′.′6 from the position detectedwithin1′.′1ofthesourceX–rayposition.Magnitudeer- (see text for details). Errorson the magnitudesare givenat 1σ rorsaregivenat1σconfidencelevel. confidencelevel. Note that the magnitudes of both the previous candidates caseinfacttheIRfluxshoulddecreaseastheX-rayfluxdrops (our#1and#3;Israeletal.2003,Safi-Harb&West2005,Durant (see Rea et al. 2004 for details), and no increase in any of the & van Kerkwijk 2006) would make 1RXSJ1708–4009 much IR bandswould be expected. If, on the other hand,the IR flux brighterintheIRbandthananyotherAXP(seee.g.Israeletal. were dominated by the disk emissivity resulting from viscous 2004a). Furthermore,all the other fainter candidates we report dissipation (which could be the case if the disc inner radius is here,layingwithinthesourcepositionalerrors(seeFig.3),have fartherout),thenfluxvariabilitywouldbetheresultofavaria- brightness in better agreementwith the IR emission properties tioninthediskmassinflowrate.Giventhenon-correlatedIRand of AXPs than both objects #1 and #3. Durant&vanKerkwijk X-rayvariabilityweobserve,thislatterscenariocouldbemore (2006)proposedsource#3asa possiblecandidateonthebasis viable.However,whileavariabilityintheIRemissioniseasily ofIRvariability,thestrangecoloursandtheresultingX/IRratio predictable as resulting from a mass inflow rate variation, this of this object. We believe that given the large number of faint shouldtakeplacesimilarlyinalltheIRbands.Asnotedin 3.1, IR sources laying within the 1RXSJ1708–4009 positional er- § weseeahintforanIR-banddependentvariability,althoughno rorcircle,andtheratherbrightmagnitudesofsource#3,neither significantconclusioncanbedrawnonthisaspectyet. the 2.5σIRvariability,noritscolours(seeFig.4)oritsX/IR Within themagnetarsscenario,a fewdifferentmodelshave ratio∼can reliablytightthis objectto the AXP (see also above). beenproposedfortheproductionoftheIRradiation(e.g.Heyl We would like to stress that we can not exclude any of these & Hernquist2005;Beloborodov& Thompson2007),but none candidates,butthatafirmidentificationofanIRcounterpartto of them makes (yet) specific predictions for the IR variability 1RXSJ1708–4009 isfarfrombeingasettledissue. thatwouldpermitadirectcomparisonwithourdata. This AXP has been recently observed to have a vari- IR variability has been observed for other magnetars, able X-ray emission (Rea et al. 2005; 2007b; Campana et i.e. 1E1048.1–5937 (Israel et al. 2002; Tam et al. 2007), al. 2007), which, if correlated somehow with the IR as for 4U0142+614(Hulleman et al. 2004), SGR1806–20 (Israel et 1E2259+586(Tam et al. 2004), would imply a variable IR al. 2005), and 1E2259+586 (Tam et al. 2004). Unfortunately, counterpart. In particular, the X-ray variability observed for withthecurrentdatanofirmconclusioncanbedrawnyetabout 1RXSJ1708–4009 in X-ray observations sparse over several theIRvariabilityinconnectionwithvariabilitiesinotherobserv- years was of the order of 50%, which would roughlyimply (if ingbands,exceptfor1E2259+586 andSGR1806–20 (Tamet X-rayandIR are directlycorrelated)anIR variabilityof 0.5 ∼ al.2004;Israeletal2005),whereacorrelationwiththesource magnitudes,notdetectedfromanyoftheobjectsneartheposi- burstingactivityandX-rayfluxenhancementhasbeendetected. tion of 1RXSJ1708–4009. However, the lack of simultaneous IRandX-rayobservationsofthissourcepreventusfromdraw- inganyfirmconclusionsaboutpossiblecorrelatedvariability. 4.2.Othersources:1E1841–045,SGR1900+14 and 1RXSJ1708–4009 5. Conclusions NoIRcounterparthadbeenclaimedsofarfor1E1841–045 and SGR1900+14, likely because of the crowding of the field in TheIRcounterparttoXTEJ1810–197 isconfirmedtobevari- whichthesemagnetarshappenstolay.Bycomparingourphoto- able in time (as previosuly proposed by Rea et al. (2004)). metricresultswiththosereportedinliteraturefor1E1841–045, However,despitepreviousclaims,itsvariabilitymightbemore it is evident that there is one object, our source #9 (source B similartoitsradiobehaviorthantoitsX–rayvariability,which in Durant (2005)), with magnitudes similar to the other AXP appears to decay smoothly after the outburst. Simultaneous IR counterparts,andshowinga>4σvariability.Furthermore,com- andradioobservationsofthisAXPareneededtoreliablyassess paringourtwoobservationsofSGR1900+14 wefounda 3σ thispossibility. ∼ variability in source #7 (not detected by Kaplan et al. (2002) To date, IR counterparts to AXPs and SGRs have been becauseitwastoofaintfortheirlimitingmagnitudes).Wethere- confirmed only for XTEJ1810–197 (Israel et al. 2004b), foreconsider these objectsas possible IR counterpartsto these 1E1048.1–5937(Wang&Chakrabarty2002;Israeletal.2002), twoneutronstars. 1E2259+586(Hulleman et al. 2001), 4U0142+614(Hulleman The new deep observation of 1RXSJ1708–4009 show 20 et al. 2000)and SGR1806–20(Israel et al. 2005;Kosugi et al. sourceswithinandinthevicinityofthe99%X-rayerrorcircle 2005).FortheAXP1E1841–045 andSGR1900+14,wepro- (seeFig.3).Noneoftheseobjectsshowedsignificantvariability poseheretwo possible candidatesbased onthedetectionofIR withrespecttootherobservationsintheliterature,whichcould variability.ForalltheremainingAXPsandSGRswe stillmiss helpidentifyingthemasreliablemagnetarcounterparts. acandidateorconfirmedIRcounterpart(seee.g.Wachteretal. Testaetal.:Adaptiveopticsnear-IRobservationsofmagnetars 9 Duncan,R.C.,&Thompson,C.1992,ApJ,392,L9 Durant,M.,2005,ApJ,632,563 Durant,M.&vanKerkwijk,2005,ApJ,628,L135 Durant,M.&vanKerkwijk,2006,ApJ,648,534 Durant,M.&vanKerkwijk,2007,ApJinpress(arXiv:0711.3985) Frail,D.A.,Kulkarni,S.R.,Bloom,J.S.1999,Nature,398,127 Gelfand,J.D.&Gaensler,B.M.2007,ApJ,667,1111 Gotthelf,E.V.,Halpern,J.P.,Buxton,M.&Bailyn,C.,2004,ApJ,605,368 Gotthelf,E.V.&Halpern,J.P.,Ap&SS,308,79 Halpern,J.,Gotthelf,E.V.,Becker,R.H,etal.,2005,ApJ,632,L29 Heyl,J.S.&Hernquist,L.2005,MNRAS,362,777 Hulleman,F.,vanKerkwijk,M.H.&Kulkarni,S.R.2000,Nature,408,68 Hulleman,F.,Tennant,A.F.,vanKerkwijk,M.H.,etal.2001,ApJ,563,L49 Hulleman,F.,vanKerkwijk,M.H.&Kulkarni,S.R.2004,A&A,416,1037 Israel,G.L.,Covino,S.,Stella,L.,etal.2002,ApJ,580,L143 Israel,G.L.,Covino,S.,Perna,R.,etal.2003,ApJ,589,L93 Israel,G.L.,Stella,L.,Covino,S.,etal.2004a,IAUSymposium218.Editedby F.Camilo&B.M.Gaensler;p.247(astro-ph/0310482) Israel,G.L.,Rea,N.,Mangano,V.,etal.2004b,ApJL,603,L97 Israel,G.L.,Covino,S.,Mignani,R.,etal.2005,ApJ,438,L1 Fig.6. SGR1900+14. Zoom-inof the area of the K band im- Kaplan,D.L.,Kulkarni,S.R.,Frail,D.A.&vanKerkwijk,M.H.2002,ApJ, s age of 2006 March 31st around the radio position of Frail et 566,378 Kosugi,G.,Ogasawara,R.&Terada,H.,2005,ApJ,623,L125 al. (1999). The error circle (radius 0.81 at 99% confidence ′′ Kuiper,L.,Hermsen,W.&Me´ndez,M.2004,ApJ,613,1173 level) is shown. Two large ghosts are visible atop the position Kuiper,L.,Hermsen,W.,denHartog,P.&Collmar,W.2006,ApJ,645,556 ofSGR1900+14,hidingstarEofKaplanetal.(2002). Leggett,S.K.,Hawarden,T.G.,Currie,J.M.,etal.2003,MNRAS,345,144 Mereghetti,S.,Mignani,R.,Covino,S.,etal.2001,MNRAS,321,143 Muno,M.P.,etal.2006,ApJ,636,L41 2004; Durant & van Kerkwijk 2005, 2007; Muno et al. 2006; Perna,R.,Hernquist,L.&Narayan,R.2000,ApJ,541,344 Perna,R.&Hernquist,L.2000,ApJ,544,L57 Gelfand&Gaensler2007). Persson,S.E.,Murphy,D.C.,Krzeminski,W.,etal.1998,AJ,116,2475 Ours and others results on optical and IR observations of Rea,N.,Testa,V.,Israel,G.L.,etal.2004,A&A,425,L5 AXPs and SGRs, show that up to now we are far from having Rea,N.,Oosterbroek,T,Zane,S.,etal.2005,MNRAS,361,710 anoverallpictureoftheopticalandIRbehavioroftheseneutron Rea,N.,Turolla,R.,Zane,S.,etal.2007a,ApJ,661,L65 stars. Further observations, possibly simultaneously on a wide Rea,N.,Israel,G.L.,Oosterbroek,T.,etal.2007b,Ap&SS,308,505 Safi–Harb,S.&West,J.2005,AdSpR,35,1172 energyrange,areneededtorefinecurrenttheoreticalmodelsand Skrutskie,M.,etal.2006,AJ,131,1163 shedlightonAXPs’andSGRs’opticalandIRemissionmecha- Stetson,P.B.1992,ASPConf.Ser. 25:AstronomicalDataAnalysisSoftware nisms. andSystemsI,25,297 Tam,C.R.,Kaspi,V.M.,vanKerkwijk,M.H.,Durant,M.2004,ApJ,617,L53 Acknowledgements. BasedonobservationscollectedattheEuropeanSouthern Tam,C.R.,Gavriil,F.P.,Dib,R.,Kaspi,V.M.,Woods,P.M.,Bassa,C.,ApJin Observatory, Paranal, Chile, under program ID 071.D-0503, 072.D-0297, press(arXiv:0707.2093) 073.D-0381,076.D-0383,277.D-5005,attheCanada-France-HawaiiTelescope Thompson,C.,&Duncan,R.C.1995,MNRAS,275,255 (CFHT) which is operated by the National Research Council of Canada, Wachter,S.,Patel,S.,Kouveliotou,C.,etal.2004,ApJ,615,887 the Institut National des Sciences de l’Univers of the Centre National de la Wang,Z.&Chakrabarty,D.2002,ApJ,579,L33 Recherche Scientifique of France, and the University of Hawaii, and at the Wang,Z.,Chakrabarty,D.&Kaplan,D.L.2006,Nature,440,772 GeminiObservatory, whichisoperated bytheAssociation ofUniversities for Woods,P.&Thompson,C.2006,CompactstellarX-raysources.EditedbyW. Research inAstronomyInc.,underacooperative agreement withtheNSFon Lewin&M.vanderKlis.CambridgeUniversityPress(astro-ph/0406133) behalf of the Gemini partnership: the National Science Foundation (United States), the Science and Technology Facilities Council (United Kingdom), the National Research Council (Canada), CONICYT (Chile), the Australian ResearchCouncil(Australia),CNPq(Brazil)andCONICET(Argentina).This publication makesuseofdataproductsfromtheTwoMicronAllSkySurvey, which is a joint project of the University of Massachusetts and the Infrared Processing andAnalysis Center/California Institute ofTechnology, funded by the National Aeronautics and Space Administration and the National Science Foundation.WethanktheYepun(UT4)andNACOteamfortheirconstanthelp ontheobservationoptimizationduringthe3yearscoveredbythisprogram,and therefereeforhis/herverycarefulreadingofthepaperandconstructivereport. VTandGLIacknowledgesupportfromMIURfunds.NRissupportedbyNWO VeniFellowship.SZacknowledgesSTFC(ex-PPARC)forsupportthroughan AdvancedFellowship. References Baring,M.G.&Harding,A.K.1998,ApJ,507,L55 Beloborodov,A.&Thompson,C.2007,ApJ,657,967 Bertin,E.&Arnouts,S.1996,A&AS,117,393 Burgay,M.,Rea,N.,Israel,G.L.,etal.2006,MNRAS,372,410 Burgay,M.,Rea,N.,Israel,G.L..etal.2007,Astronomers’Telegram#903 Camilo,F.,Ransom,S.,Halpern,J.,etal.2006,Nature,442,892 Camilo,F.,Ransom,S.M.,Halpern,J.P.,Reynolds,J.2007a,ApJ,666,L93 Camilo, F., Ransom, S. M., Penalver, J., et al. 2007b, ApJ in press (arXiv:0705.4095) Camilo,F.,Cognard,I.,Ransom,S.,etal.2007c,ApJ,663,497 Campana,S.,Rea,N.,Israel,G.L.,etal.2007,A&A,463,1047 Chatterjee,P.,Hernquist,L.&Narayan,R.,2000,ApJ,534,

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