Mon.Not.R.Astron.Soc.000,000–000(0000) Printed5February2008 (MNLATEXstylefilev2.2) The Ultra Luminous X-ray sources in the High Velocity System of NGC1275 O. Gonza´lez-Mart´in1,2⋆, A.C. Fabian2 and J.S. Sanders2 1InstitutodeAstrofi´sicadeAndaluci´a(CSIC),Apdo3004,18080Granada,Spain 2InstituteofAstronomy,MadingleyRoad,CambridgeCB30HA 6 0 5February2008 0 2 n ABSTRACT a We report the results of a study of X-ray point sources coincident with the High Velocity J System(HVS) projectedin frontofNGC1275.A verydeepX-rayimageofthe coreofthe 9 PerseusclustermadewiththeChandraObservatoryhasbeenused.We findapopulationof Ultra-LuminousX-raysources(ULX;7sourceswithL (0.5−7.0keV)>7×1039ergs−1). X 1 AswiththeULXpopulationsintheAntennaeandCartwheelgalaxies,thoseintheHVSare v associatedwith a regionofveryactivestar formation.Severalsourceshavepossibleoptical 0 counterparts found on HST images, although the X-ray brightest one does not. Absorbed 8 power-lawmodelsfittheX-rayspectra,withmosthavingaphotonindexbetween2and3. 1 1 Keywords: galaxies:clusters:individual:Perseus-ULX-galaxies:individual:NGC1275 0 6 0 / h 1 INTRODUCTION different fromthesampleof stellar-massBHCand areconsistent p withbeingIMBH. - The study of Ultra-Luminous X-ray sources (ULX) has been o Here we report on the discovery of a population of 8 point greatlyexpandedbythehighspatialresolutionandspectralgrasp r X-raysourcestotheNofthenucleusofNGC1275, whichisthe t oftheChandraandXMM-Newtonobservatories,respectively.ULX s centralgalaxyinthePerseuscluster.Allexceed1039ergs−1 inX- a sources(Fabbiano&White2003;Miller&Colbert2004)have2– : 10 keV X-ray luminosities exceeding 1039ergs−1 and are found rayluminosity,and7areformallyULX,iftheyareatthedistance v ofthecluster.Thespatialregionwheretheyliecoincideswiththe some distance from the centres of galaxies; they are not active i HighVelocitySystemofNGC1275.Weassumethattheyarepart X galactic nuclei. Their luminosity exceeds that for a 10M⊙ black ofthatsystem.Weseenootherpoint sources(apartfromthenu- r hole accreting at the Eddington limit which radiates isotropically a andsohavecreatedmuchinterestinthepossibilitythattheycontain cleus)overthebodyofNGC1275(Fig.1). even higher mass black holes, such as InterMediate Black Holes NGC1275 is embedded in a complex multiphase environ- (IMBH) of ∼103M⊙ (Makishima et al 2000; Miller, Fabian & ment. Optical imaging and spectroscopy first established the ex- Miller2004).Alternativelytheymayappearsoluminousbecause istence of two distinct emission-line system toward NGC1275: a of beaming (Reynoldset al1999; Kinget al2001, Zezas&Fab- low-velocity component associated withthe galaxy itselfat 5200 biano2002)orduetosuperEddingtonaccretion(Begelman2002). kms−1 andahigh-velocitycomponent at8200kms−1 projected ULXaremostcommoninstarburstgalaxiesandinveryactive nearbyonthesky(Minkowski1955,1957).Thislattercomponent star-forming regions, such as in the Antennae and the Cartwheel is associated with a small gas-rich galaxy falling into the cluster galaxy, where populations of tens of them are found (Zezas et al along our line of sight (Haschick, Crane & van der Hulst 1982). 2002; Gao et al 2003; Wolter & Trinchieri 2004). In some cases A merger scenario has been proposed (Minkowski 1955, 1957). variability rules out the possibility that they are just clusters of However, interactionof thelow and/or high-velocitysystemwith lower-luminosityobjects.TheoriginofIMBHisunclear.Theymay athirdgas-richgalaxyorsystemofgalaxies(Holtzmanetal.1992; formasaresultofbinaryinteractionsindensestellarenvironments Conselice, Gallagher & Wyse 2001), or influences from the sur- (PortegiesZwart&McMillan2002).AcomparisonofIMBHULX roundingdenseintraclustermedium(ICM)(Sarazin1988;Boroson candidateswithanumber of wellknown stellar-massblackholes 1990;Cauletetal.1992)havebeendiscussed. candidates(BHC;Milleretal2004)demonstratesthattheULXare DeepChandraobservationshaveclarifiedthepositionofthe moreluminousbuthavecoolerthermaldiskcomponentsthanstan- HighVelocitySystem(HVS).ThedepthoftheobservedX-rayab- dardstellar-massBHC.Therefore,ULXinthissampleareclearly sorption(e.g.Fig.1)isnorinfilledbyemissionfromhotgaspro- jectedalongtheline-of-sightsotheHVSmustliewellinfrontof NGC1275. Gillmon, Sanders & Fabian (2004) have estimated a lowerlimitonthedistanceoftheHVSfromthenucleusof57kpc. ⋆ E-mail:[email protected] Thelow-andhigh-velocitysystemarethereforenotyetdirectlyin- (cid:13)c 0000RAS 2 O. Gonza´lez-Marti´n,A.C. FabianandJ.S.Sanders Figure1.NGC1275inthe0.3–7.0keVband.Pixelsare0.49arcsecinsizeandtheN-Sheightoftheimageis97arcsec.Northistothetopandeastistothe leftinthisimage.ThepointsourceswhichweidentifyasULXarelocatedinthehigh-velocitysystemisseeninabsorptiontothenorthofthebrightnucleus. Noneareseeninthesouthlobe teracting.TheHVSishowever stronglyinteractingwiththeICM 29gain ctiN0003).Thedatasetsanalysedeachusedanaimpointon of the Perseus cluster, which has triggered strong star formation. theACIS-S3CCD.Thedatasetswerefilteredusingthelightcurve InthispaperwedescribethedetailedanalysisoftheX-rayspatial in the 2.5 to 7 keV band on ACIS-S1 CCD, which is a back- andspectralpropertiesofthediscretesourcesinthehighvelocity illuminatedCCDliketheACIS-S3.TheCIAOLC CLEANtoolwas system. used to remove periods 20 per cent away from the median count Thepaperisorganizedasfollows:inSect.2and3wepresent rateforallthelightcurves.Thisprocedurewasnotusedfordatasets reductionandresultsfromtheimaginganalysisandspectralanaly- 03209and04289whichdidnotincludetheS1CCD,howeverno sis,respectively;inSect.4wediscusstheresults;andSect.5sum- flareswereseenintheseobservationsontheS3CCD.Eachofthe marizesourfindings. observationswasreprojectedtomatchthecoordinatesystemofthe Throughout this paper we use a redshift of 0.018 and H0= 04952observation. 70kms−1Mpc−1.Thisgivesaluminositydistancetothecluster The 900 ks X-ray image covering the energy range of80Mpc;1arcseccorrespondstoaphysicaldistanceof370pc. 0.3−0.8keVisshowninFig2.ThebrightNGC1275nucleusis clearlyseenatRA3h19m48sandDec.+41o30’42”(J2000)andthe high-velocity systemisseen inabsorption tothenorth of thenu- cleus. 2 IMAGINGANALYSIS The CIAO CELLDETECT source detection routine was then TheChandradatasetsincludedinthisanalysisarelistedinTable1. used on the reprocessed level 2 event data to produce a prelimi- Thetotalexposuretime,afterremovingperiodscontainingflares, narylistofpointsources.Thecellsizerangesbetween4pixelsto is890ks.Topreparethedataforanalysis,allofthedatasetswere 8pixels.Thisalgorithmstronglydependsonthelocalbackground reprocessed tousethe latestappropriate gainfile(acisD2000-01- and the detection cell in not adjustable to the size of the source. (cid:13)c 0000RAS,MNRAS000,000–000 X-raysourcesin theHVS ofNGC1275 3 Figure2.ThecentralregionofNGC1275inthe0.3–0.8keVband.Pixelsare0.49arcsecindimensionandtheentireimageis1.77×0.89arcmin2.Northis tothetopandeastistotheleftinthisimage.Thehigh-velocitysystemisseeninabsorptiontothenorthofthebrightnucleuswhichisatRA3h19m48s,Dec. +41o30’42” Obs.ID Sequence Observationdate Exposure(ks) Nominalroll(deg) PointingRA PointingDec 3209 800209 2002-08-08 95.8 101.2 3:19:46.86 +41:31:51.3 4289 800209 2002-08-10 95.4 101.2 3:19:46.86 +41:31:51.3 6139 800397 2004-10-04 51.6 125.9 3:19:45.54 +41:31:33.9 4946 800397 2004-10-06 22.7 127.2 3:19:45.44 +41:31:33.2 4948 800398 2004-10-09 107.5 128.9 3:19:44.75 +41:31:40.1 4947 800397 2004-10-11 28.7 130.6 3:19:45.17 +41:31:31.3 4949 800398 2004-10-12 28.8 130.9 3:19:44.57 +41:31:38.7 4950 800399 2004-10-12 73.4 131.1 3:19:43.97 +41:31:46.1 4952 800400 2004-10-14 143.2 132.6 3:19:43.22 +41:31:52.2 4951 800399 2004-10-17 91.4 135.2 3:19:43.57 +41:31:42.6 4953 800400 2004-10-18 29.3 136.2 3:19:42.83 +41:31:48.5 6145 800397 2004-10-19 83.1 137.7 3:19:44.66 +41:31:26.7 6146 800398 2004-10-20 39.2 138.7 3:19:43.92 +41:31:32.7 Table1.Chandraobservations includedinthisanalysis.Theexposuregivenisthetimeremainingafterfilteringthelightcurveforflares.Allobservations weretakenwiththeaimpointontheACIS-S3CCD.AllpositionsareinJ2000coordinates. AstheX-raydiffuseemissionoftheNGC1275isverystrong,the Allthepoint-likesourcesarelistedinTable2,showingtheir source list may well include false detections in high background positionsandcountrates. levelregions.Thereforeproblematicsourcesembeddedinsuchre- We have used archival HST observations of NGC1275 in gionshavebeenexcludedinouranalysis.Moreover,asmentioned order to search for optical counterparts. The galaxy was imaged above,weonlyincludedsourcesassociatedwiththeHVS. withtheWFPC2cameraonHST usingtheF814W(∼I,on2001 We have detected 8 bright sources close to the nucleus of November6withanexposuretimeof1200s)andF702W(∼R,on NGC1275,locatedinthenortherninnerradiolobeof3C84.Allof 1994March31withanexposuretimeof140s)broad-bandfilters. thesesourceareembeddedinthesameregionastheHVS(seeFig. Several coincidences between X-raysources and optical knots of 1).Therearenosourcesassociatedwiththesouthernlobe(Fig.1), emission(F814W)canbeseeninFig.4(right),showingthesame thusweassumethesesourcesareassociatedwiththeHVS. regionsasFig.4(left). Fig.4(left)showsthesmoothed ACIS-S3imageinthe0.3– TheHST imageshowsmanyhighlyabsorbedfeatures.When 7.0keVband,includingnumberedlabelsofallthedetectedsources wecompareindetail,sourcesN7andN8arelocatedinstarforming (top),centredonsourcelabelledN3(centre)andcentredonsource regions, while N2 and N6 have a point-like counterpart. Sources N5(bottom). N1, N3, N4 and N5have no optical identification. Therefore, we (cid:13)c 0000RAS,MNRAS000,000–000 4 O. Gonza´lez-Marti´n,A.C. FabianandJ.S.Sanders N Position(J2000) CountRate N NH G c 2/d.o.f. (countsks−1) (1021cm−2) 1.... 03:19:48.736+41:30:47.25 0.34±0.05 1.... 2.5(a) 3.20+0.23 112.90/101 −0.37 2.... 03:19:48.166+41:30:46.64 1.69±0.06 2.... 2.72+1.43 1.78+0.30 101.50/109 −0.87 −0.24 3.... 03:19:48.090+41:31:01.88 2.60±0.08 3.... 2.49+0.40 2.08+0.09 153.86/142 −0.40 −0.09 4.... 03:19:47.994+41:30:52.30 1.42±0.09 4.... 2.05+0.91 2.29+0.44 156.24/152 −0.96 −0.28 5.... 03:19:47.925+41:30:47.50 1.19±0.09 5.... 2.64+1.23 2.92+1.44 124.09/139 −0.93 −0.36 6.... 03:19:47.602+41:30:47.01 0.74±0.06 6.... 3.74+1.57 3.51+0.48 102.58/92 7.... 03:19:47.422+41:30:51.93 0.95±0.08 7.... 4.03+−11..7389 3.20+−10..3696 133.69/135 8.... 03:19:47.214+41:30:47.62 1.28±0.08 8.... 2.66−+11..4050 2.13−+00..4582 150.81/138 −0.91 −0.25 Table2.PositionsofsourcesdetectedneartheNGC1275centreanddis- Table4.Spectralfits.(a)ThecolumndensityofsourceN1hasbeenfixed played in Fig. 4(column 2) andcount rate in the energy range between duetothelowcountrate. 0.5–7.0keV(column3). N Fobs(0.5–7.0keV) Fcorr(0.5–7.0keV) logLX N FX/FF814W FX/FF702W MF814W MF702W ergcm−2s−1 ergcm−2s−1 0.5–7.0keV 1.... 2.09×10−15 4.34×10−15 39.51 1.... >26.5 >16.2 >22.6 >22.1 2.... 7.59×10−15 9.97×10−15 39.86 2.... 25.4 23.6 20.4 20.3 3.... 1.64×10−14 2.28×10−14 40.22 3.... >18800 ... >26.8 ... 4.... 7.76×10−15 1.10×10−14 39.91 4.... >1081 >800 >24.5 >24.2 5.... 5.36×10−15 1.07×10−14 39.90 5.... >123 >60.6 >22.6 >21.9 6.... 3.02×10−15 9.23×10−15 39.84 6.... 26.2 28.4 21.7 21.7 7.... 4.28×10−15 1.18×10−14 39.95 7.... 76.7 51.4 22.3 21.9 8.... 7.67×10−15 1.16×10−14 39.93 8.... 134 90.1 22.2 21.7 Table 5. Fluxes (observed and k-corrected) and luminosities assuming a Table 3. Optical analysis. X-ray to optical ratios (columns 2and3) and cosmologicalmodelwithH0=70kms−1Mpc−1andz=0.018. magnitude determinations (columns 4 and 5) for the filters F814W and F702W,respectively,withtheX-rayfluxbetween1.0–7.0keV. eitherasource-freecircularannulusorseveralcirclessurrounding eachsource,inordertotakeintoaccountthespatialvariationsof havefoundapossiblecorrelationbetweencompactX-raysources thediffuseemissionandtominimizeeffectsrelatedtothespatial and regions of vigorous star formation. The implications are dis- variationoftheCCDresponse. cussedlater. For each source, we extracted spectra from each of the InordertoinvestigatetheemissionmechanismoftheseULX, datasets.Thesespectraweresummedtoformatotalspectrumfor the X-ray to optical flux ratios have been computed between the eachsource.Responseandancillaryresponsefileswerecreatedfor F702WandF814WHSTbroad-bandsand1.0–7.0keVX-rayband. eachsourceineachoftheobservationsusingtheCIAOMKACIS- Preliminaryprocessingoftherawimagesincludingcorrectionsfor RMFandMKWARFtools.Theresponsesforaparticularsourcewere theflatfieldingwasdoneremotelyattheSpaceTelescopeScience summedtogether,weightingaccordingtothenumberofcountsin Institutethroughthestandardpipeline.Foreachframe,cosmicrays eachobservation. wereremovedbyimagecombination,usingtheIMCOMBINErou- ThespectrawerefittedusingXSPECv.11.3.2.Inordertouse tine in IRAF. After cosmic ray removal, the frames were added the c 2 statistic,wegroupedthedatatoincludeatleast20counts usingtaskWMOSAICinSTSDASpackage. Photometricmeasure- perspectralbin,beforebackground subtraction.Inspectralfitting mentsweremadewithPHOTtask,withintheNOAOpackage.Fi- weexcludedanyeventswithenergiesabove7.0keVorbelow0.5 nally the fluxes and magnitudes have been determined using the keV. photometriczero-pointinformationintheheaderofthecalibrated Table4summarizesourspectralresultsintermsoftheabsorb- imagefiles. ingcolumndensityandphotonindex. TheseresultsareshowninTable3,includingtheX-raytoopti- Thesourceshavebeenmodelledwithanabsorbedpowerlaw calfluxratiosfromtheF814WandF702Wbroad-bandfilters,and slopewithphotonindexbetweenG =[1.78-5.56]andanequivalent the magnitude determinations from the same filters. In the cases columndensityofNH=[2.05−4.03]×1021cm−2.Inallthecases whereanopticalcounterparthasnotbeenfound(N1,N3,N4and thesinglecomponentpowerlawgivesatisfactoryfits.Thecolumn N5),themagnitudesandfluxesarejustalowerlimit. densityofsourceN1hasbeenfixedduetothelowcountrate.The fittedN values areconsistent withtheintrinsicabsorption mea- H sured e.g. in theoptical band; the value of A =0.54 corresponds V 3 SPECTRALANALYSIS to NH∼1.1×1021cm−2, assuming AV=NH×5.3×10−22 for RV=3.1(Bohlinetal.1978).Thisvalueshouldbealowerlimit WeextractedspectraforallthedetectedsourcesclosetotheHVS, tothefittedN valuetobeconsistent,asisseeninTable4. H usingextractionregionsdefinedtoincludeasmanyofthesource As an example of our spectral fits, the brightest source, N3, photonsaspossible,butatthesametimeminimizingcontamination hasbeenfittedwithapower-lawwithspectralindexof2.08±0.09 fromnearbysourcesandbackground.Thebackgroundregionwas andabsorptionofNH=2.5±0.4×1021 cm−2(seeFig.3). (cid:13)c 0000RAS,MNRAS000,000–000 X-raysourcesin theHVS ofNGC1275 5 Figure4. Top-left:Broadband(0.5–7.0keV)X-raysmoothedimagewithsourceslabelled.Top-right:HST/WFPC2F814Wbroadbandimage.Centre-left: Broadband(0.5–7.0keV)X-raysmoothedimagecentredinsource3.Centre-right:HST/WFPC2F814Wbroadbandimagecentredinsource3.Bottom-left: Broadband(0.5–7.0keV)X-raysmoothedimagecentredinsource5.Bottom-left:HST/WFPC2F814Wbroadbandimagecentredinsource5. InTable5welistthe0.5-7keVfluxand(absorptioncorrected) The brightest point source has a luminosity of luminositiesoftheindividual sourcesbased onthebest-fitpower LX(0.5−7.0keV)=1.67×1040ergs−1, and is one of the lawmodel. brightestindividualsourcesfoundinagalaxy.AULXsourcemore luminousthantheentireX-rayluminosityofanormalgalaxyhas The lower limit of the luminosity of point sources been found intheCartwheel system withaluminosityof at least in the image, if at the distance of NGC1275, is L ∼2−4×1040ergs−1 (Gao et al. 2003; Wolter & Trinchieri LX(0.5−7.0keV)=3.2×1039ergs−1, which is already 20X04).Theyexplainthisluminositywithahigh-massX-raybinary well above the Eddington limit for a neutron star binary source (HMXB). The high X-ray luminosity suggests either a (L ∼3×1038ergs−1) and is also above the limit of canonical X singleextremelybrightsource,oraverydensecollectionofseveral ULX,i.e.≥1039ergs−1. (cid:13)c 0000RAS,MNRAS000,000–000 6 O. Gonza´lez-Marti´n,A.C. FabianandJ.S.Sanders thePerseusgalaxyclusterhastriggeredstarclusterformationinthe HVC,thentheclustersshouldhaveageslessthan∼108yr. Ourinterpretationofthespatialcorrespondencewithstarclus- tersisthattheregionsareespeciallyactive,indicatingareallinkbe- tweenULXandstar-formingregions,andmeaningtheyareyoung objects. However the optical limitson sources N3 and 4 rule out anyassociation withmassive clustersinthose cases(thelimit on theabsolutemagnitudeisabout−8). In M31 and the Milky Way (Grimm, Gilfanov & Sunyaev 2003), XRB have luminosities consistent with the Eddington limit of a ∼2M⊙ accreting object. They produce luminosities ∼3×1038ergs−1,aboutoneorderofmagnitudebelowthelimit- ingluminosityinoursample(3.2×1039ergs−1).Itispossiblethat ourULXconsistofatleast15(or130,inthecaseofthebrightest sourcefound)‘normal’XRBclusteredtogether,perhapsinayoung Figure3.ACIS-SspectrumofsourceN3.Thesolidlinecorrespondswith starcluster.Howeverinotherobjectsweknowthatvariabilityre- a power law model with a spectral index of G =2.08 and absorption of NH=2.5×1021cm−2.Thefitresidualsarepresentedinthelowerpanel. quires the presence of intrinsically luminous X-ray sources (e.g. M82;Griffithsetal.2000,Kaaretetal.2001).Alternativepossibil- itiesarethatblackholesources,withmassesintherangeofgalactic highL sources,whichwouldbeevenmorepeculiar.Evidenceof blackholebinaries,aremildlybeamed(Reynoldsetal.1999and X timevariabilitymightsuggestthatisasinglehighL source. Kingetal.2001).Spectralandtimingfeatureshoweverruleoutthis X Time variability analysis has been performed. The observa- possibilityinsomeULX(e.g.Strohmayer&Mushotzky2003).We tionsspanabouttwoyears.Twodatafileswereobservedon2002 notethatcompactsupernovaremnantssometimeshaveULXlumi- August8and10,andtheotherelevendatafileswereobservedfrom nosities(e.g.Fabian&Terlevich1998), butnorecentsupernovae 2004 October 4 to 2004 October 20, giving an almost daily cov- havebeen reported for NGC1275 (SN1968A wastotheS of the erage. The exposure times are between 22 and 143 ks. The data HVS;Capetti2002). characteristicsallowsusdetermineshortvariationin16days(sec- Finally,werecalltheIMBHmodelwhichhasspectralsupport ondperiod)andlong-termvariabilityof2years.Becauseofthelow from some sources (Miller et al 2004; the level of absorption in countratesofthesourcesinNGC1275(seeTable2),itisveryhard NGC1275 is too high for any soft excess to be observed). They tosearchforshort-termvariability.Weextractedlight-curves,us- mayformindensestarclusters. ingDMEXTRACTCIAOtaskforthetwobrightestsources(N3and OuropticalstudieshaveclearlyshownthattheULXhavevery N4) (net count rate greater than 0.98 count s−1) binned with bin high X-Ray to optical flux ratios. X-ray selected AGN from the sizesof500,1000,2500and5000s.Inbothcasesthepointswere Rosatallskysurveytendtohavelog(FX/Fopt)∼1.ThustheULX consistentwiththerespectivemeanvaluesandvariabilityhasnot do not have the optical properties expected if their were simple beenfound.Furthermore,themeanvaluesbetween2002and2004 extensionsofAGN(IMBH,aslowluminositylimit).However,low are the same, including errors bars. Therefore, evidence of time massX-raybinariesintheMilkyWayhaveFX/Fopt∼100−10000 variabilityhasnotbeenfoundduringthewholesetofobservations. (Mushotzky2004). Theresultsfoundinoursystemindicatethatwehaveamixed groupofobjects(seeTable3).Atleast4outof8sources(N3,N4, N5and N8) havehigh X-rayto optical fluxratios. Atleast 3 out 4 DISCUSSION of8(N1,N2andN6)havelowerX-raytoopticalratios,possibly Chandra has revealed significant populations of ULX in the in- becausetheylieinstarclusters. teractingsystemsoftheAntennae(NGC4038/9;Zezas,Fabbiano Our data are consistent with no significant variability, simi- &Murray2002) andtheCartwheelringgalaxy(Gaoetal.2003; lartotheresultobtainedonNGC3256byLiraetal.(2002).Time Wolter&Trinchieri2004),wheredramaticeventshavestimulated variability is frequently observed in ULX (e.g. IC342, Sugiho et massivestarformation.Wehavereportedhereonanotherexample al.2001 orM51X-1,Liuetal.2002), arguingthatmostof them (Fig.4left)intheHVSofNGC1275whichisinteractingwiththe aresingle compact objects, rather than asum of numerous lower ICMofthePerseuscluster. luminosityobjectsinthesameobject.WhilemostULXvary,many The sources are spatially associated with the distribution of showlowamplitudevariabilityonlongtimescales(e.g.theAnten- absorbing clouds seen in soft X-ray (Fig. 2) and optical (Fig. 3) naegalaxies,Zezasetal.2002),whichisverydifferenttogalactic images. Two sources (N7 and N8) are directly linked with dust blackholes.PortegiesZwart,Dewi&Maccarone(2004)findthata knotsandanothertwo(N2andN6)haveanopticalpoint-likecoun- persistentbrightULXrequiresadonerstarexceeding15M⊙.The terpart (Fig. 4 bottom). Similar correspondence have been found searchforcharacteristicfrequenciesisoneofthemostproductive in the Cartwheel galaxy withthe outer ring (Wolter & Trinchieri wayofdeterminingthenatureoftheULX. 2004) andintheAntennaegalaxieswith39X-raysourceswithin theWFPC2field(Zezasetal.2002).Theopticalbrightnessofthe counterparts in the HVC are too high to be individual stars and 5 CONCLUSIONS somaybeassociatedwithyoung starclusters.Followingthedis- cussionofyoungstarclustersinNCG1275givenbyRicheretal We have described the detailed analysis of the spatial and spec- (1993),anobjectofmagnitude22correspondstoaclustermassof tral properties of thediscrete X-ray sources detected with adeep about 106M⊙ if itsage isabout 107yr.The HVCsystem travels Chandra ACIS-S observation around NGC1275. Our results are atleast30kpcin107yrsoifastronginteractionwiththecoreof summarizedbelow: (cid:13)c 0000RAS,MNRAS000,000–000 X-raysourcesin theHVS ofNGC1275 7 (i) We have detected a total of 8 sources to the north of MinkowskiR.,1957,inVandeHulstH.C.,ed.,Proc.IAUSymp.4,Radio NGC1275nucleus. Astron..CambridgeUniv.Press,Cambridge,p.107 (ii) ThesourcesarespatiallycoincidentwiththeHighVelocity MushotzkyR.,2004,Prog.ofTheo.Phys.Supplement,155,27 System and thus probably associated with it. They are therefore PortegiesZwartS.F.,McMillanS.L.W.,2002,ApJ,576,899 PortegiesZwartS.F.,DewiJ.,MaccaroneT.,2004,MNRAS,355,413 ULX. ReynoldsC.S.,HenzS.,FabianA.C.,BegelmanM.C.,1999,ApJ,521,99 (iii) FourofthesourceshaveanopticalcounterpartintheIand Richer,H.B.,Crabtree,D.R.,Fabian,A.C.,Lin,D.N.C.,1993,AJ,105,877 Rbands(fromHST images); twoofwhicharepoint-likesources Sarazin C. L., 1988, X-Ray Emissions from Clusters of Galaxies, Cam- andtheothertwoareassociatedwithstar-formingregions. bridgeUniv.Press,Cambridge. (iv) Inallthecasesasinglecomponent power lawgivessatis- SugihoM.,KotokuJ.,MakishimaK.,KubotaA.,MizunoT.,FukazawaY., factoryfits,withspectralindexofG =[1.78-3.51]andanequivalent TashiroM.,2001,ApJ,561,L73 columndensityofNH=[2.05−4.03]×1021cm−2. StrohmayerT.E.,MushotzkyR.F.,2003,ApJ,586,L61 (v) The minimum luminosity is WolterA.,TrinchieriG.,2004,A&A,426,787 LX(0.5−7.0keV)=3.2×1039ergs−1 (source N1), which is ZezasA.,FabbianoG.,,2002,ApJ,577,726 alreadyabovethelimitofcanonicalULX. ZezasA.,FabbianoG.,RotsA.H.,MurrayS.S.,2002,ApJSupplement, 142,239 (vi) No variability was detected in the two brightest sources found. Our results add to the growing evidence that some episodes of rapid star formation lead to the production of ULX. Young, massive, starclustersmaybeinvolvedinsome, but notall ofthe sources. ACKNOWLEDGEMENTS OGMacknowledgesthefinancialsupportbytheMinisteriodeEd- ucacionyCienciathroughtheprogramAYA2003-00128andgrant FPIBES-2004-5044.ACFthankstheRoyalSocietyforsupport. REFERENCES BegelmanM.C.,2002,ApJ,568,L97 BohlinR.C.,SavageB.D.,DrakeJ.F.,1978,ApJ,224,132 BorosonT.A.,1990,ApJ,360,465 Capetti,A.,2002,ApJ,574,L25 CauletA,WoodgateBE.,BrownL.W.,GullT,R.,HintzenP,Lowenthal J.D.,OliversenR.J.,ZieglerM.M.,1992,ApJ,388,301 ConseliceC.J.,GallagherJ.S.,WyseR.F.,2001,AJ,122,2281 FabbianoG.,WhiteN.E.,2003,astro-ph/0307077 FabbianoG.,ZezasA.,MurrayS.,2001,ApJ,554,1035 FabianA.C.,TerlevichR.,1996,MNRAS,280,L5 GaoY.,WangQ.D.,AppletonP.N.,LucasR.A.,2003,ApJ,596,L171 GebhardtK.,RichR.M.,HoL.C.,ApJ,inpress,astro-ph/0508251 GillmonK.,SandersJ.S.,FabianA.C.,2004,MNRAS,348,159 GriffithsR.E.,PtakA.,FeigelsonE.D.,GarmireG.,TownsleyL.,Brandt W.N.,SambrunaR.,BregmanJ.N.,2000,Science,290,1325 GrimmH.,GilfanovM.,SunyaevR.,2003,MNRAS,339,793 HaschickA.D.,CraneP.C.,vanderHulstJ.,1982,ApJ,262,81 HoltzmanJ.A.,FaberS.M.,ShayaE.J.,LauerT.R.,GrothJ.,HunterD. A.,BaumW.A.,EwaldS.P.,HesterJ.J.,LightR.M.,LyndsC.R., O’NeilE.J.,WestphalJ.A.,1992,AJ,103,691 KaaretP.,PrestwichA.H.,ZezasA.,MurrayS.S.,KimD.-W.,KilgardR. E.,SchlegelE.M.,Ward,M.J.,2001,MNRAS,321,L29 KingA.R.,DaviesM.B.,WardM.J.,FabbianoG.,Elvis,M.,2001,ApJ, 552,L109 LiraP.,WardM.,ZezasA.,Alonso-HerreroA.,UenoS.,2002,MNRAS, 330,259 LiuJ.,BregmanJ.N.,IrwinJ.,SeitzerP.,2002,ApJ,581,L93 MakishimaK.,KubotaA.,MizunoT.,OhnishiT.,Tashiro,M.,ArugaY., AsaiK.,DotaniT.,MitsudaK.,UedaY.,UnoS.,YamaokaK.,Ebisawa K.,KohmuraY.,OkadaK.,2000,ApJ,535,632 MillerJ.M.,FabianA.C.,MillerM.C.,2004,ApJ,614,L117 MillerJ.M.,ColbertE.,2004,IJMod.Phys.,13,1 MinkowskiR.,1955,CarnegieYearbook,54,25 (cid:13)c 0000RAS,MNRAS000,000–000