Infrared Studies of Galactic Center X-ray Sources

MISSING IMAGE

Material Information

Title:
Infrared Studies of Galactic Center X-ray Sources
Physical Description:
1 online resource (208 p.)
Language:
english
Creator:
Dewitt,Curtis Noel
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Astronomy
Committee Chair:
Eikenberry, Stephen S
Committee Co-Chair:
Bandyopadhyay, Reba M
Committee Members:
Sarajedini, Ata
Packham, Christophe C
Muller, Guido

Subjects

Subjects / Keywords:
Astronomy -- Dissertations, Academic -- UF
Genre:
Astronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
In this dissertation I use a variety of approaches to discover the nature of a subset of the nearly 10,000 X-ray point sources in the 2 X 0.8 degree region around the Galactic Center.I produced a JHKs source catalog of the 17' X 17' region around Sgr A*, an area containing 4339 of these X-ray sources, with the ISPI camera on the CTIO 4-m telescope. I cross-correlated the Chandra and ISPI catalogs to find potential near-infrared (NIR) counterparts to the X-ray sources. The extreme NIR source crowding in the field means that it is not possible to establish the authenticity of the matches with astrometry and photometry alone. I found 2137 IR/X-ray astrometrically matched sources; statistically I calculated that my catalog contains 289 +/- 13 true matches to soft X-ray sources and 154 +/- 39 matches to hard X-ray sources. However, the fraction of matches to hard sources that are spurious is 90%, compared to 40% for soft source matches, making the hard source NIR matches particularly challenging for spectroscopic follow-up. I statistically investigated the parameter space of matched sources and identified a set of 98 NIR matches to hard X-ray sources with reddenings consistent with the GC distance which have a 45% probability of being true counterparts. I created two additional photometric catalogs of the GC region to investigate the variability of X-ray counterparts over a time baseline of several years. I found 48 variable NIR sources matched to X-ray sources, with 2 spectroscopically confirmed to be true counterparts (1 in previous literature and one in this study). I took spectra of 46 of my best candidates for counterparts to X-ray sources toward the GC, and spectroscopically confirmed 4 sources as the authentic physical counterpart on the basis of emission lines in the H and K band spectra. These sources include a Be high mass X-ray binary located 16 pc in projection away from Sgr A*; a hard X-ray symbiotic binary located 22 pc in projection from Sgr A*; an O-type supergiant at an distance of 3.7 kpc; and an O star at the Galactic Center distance. I also identified 3 foreground X-ray source counterparts within a distance of 1 kpc which do not show obvious emission features in their spectra. However, on the basis of the low surface density of unreddened sources along the line-of-sight to the Galactic Center and our previous statistical analysis (Dewitt et al. 2010), these can be securely identified as the true counterparts to their coincident X-ray point sources. Lastly, I used the results of my matching simulations to infer the presence of 7 +/- 2 true counterparts within a set of late type giants that I observed without detectable emission features. I conclude from this work that the probable excess in red giant X-ray counterparts without emission lines needs to be confirmed both with larger samples of spectroscopically surveyed counterparts and more advanced statistical simulations of the match authenticity. Also, the nature of the compact object in two of my counterpart discoveries, the Be HMXB and the symbiotic binary, can be strongly constrained with X-ray spectral fitting. Lastly, I conclude that spectroscopic surveys for new X-ray source counterparts in the GC may be able to increase their efficiency by specifically targeting photometric variables and very close astrometric matches of IR/X-ray sources.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Curtis Noel Dewitt.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Eikenberry, Stephen S.
Local:
Co-adviser: Bandyopadhyay, Reba M.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043234:00001


This item is only available as the following downloads:


Full Text

PAGE 2

2

PAGE 3

Thanksgoouttomydissertationadvisorsfortheirterricsupportandtheopportunitytoworkonsuchacooltopic.Thanksgoouttomyparentsfortheirpatience,loveandsupport.Thanksalsogoouttomyfriends,withoutwhomthissagawouldnothavebeennearlyasenjoyable. 3

PAGE 4

page ACKNOWLEDGMENTS .................................. 3 LISTOFTABLES ...................................... 7 LISTOFFIGURES ..................................... 9 ABSTRACT ......................................... 13 CHAPTER 1X-RAYBINARIESANDTHECENTEROFTHEGALAXY ............ 15 1.1Introduction ................................... 15 1.2HistoryofX-rayAstronomy .......................... 16 1.3X-rayEmissionandAbsorptionMechanisms ................ 18 1.3.1BlackBodyRadiation .......................... 18 1.3.2ThermalBremsstrahlung ........................ 19 1.3.3RecombinationEmission ........................ 19 1.3.4InverseComptonRadiation ...................... 20 1.3.5X-rayAbsorptionfromtheInterstellarMedium ............ 20 1.4X-rayEmittingObjects ............................. 21 1.4.1X-rayBinariesandCataclysmicVariables .............. 21 1.4.2HighMassX-rayBinaries(HMXBs) .................. 22 1.4.3LowMassX-rayBinaries(LMXBs) .................. 24 1.4.4CataclysmicVariables ......................... 25 1.4.5SymbioticStars ............................. 26 1.4.6Wolf-Rayet(WR)Stars,OStarsandWR/OBinaries ........ 27 1.5TheGalacticCenter .............................. 27 1.6TheGalacticCenterinX-rays ......................... 29 1.7ScopeofthisDissertation ........................... 32 2X-RAYANDNEAR-INFRAREDIMAGINGOFTHEGALACTICCENTER ... 39 2.1ChandraX-rayDataoftheGalacticCenter ................. 40 2.2Near-Infrared(NIR)Photometry ........................ 40 2.2.1DatafromtheInfraredSidePortImager(ISPI) ............ 40 2.2.2DatafromtheUnitedKingdomInfraredDeepSkySurvey(UKIDSS) 41 2.2.3DatafromtheVisibleandInfraredSurveyTelescopeforAstronomy(VISTA) ................................. 42 2.2.42MicronAllSkySurvey(2MASS)AllSkyDataRelease ...... 42 2.3DataReduction ................................. 43 2.3.1CreatingtheISPIcatalog ....................... 43 2.3.2CorrectingaSpatiallyVaryingZeropointintheISPIPhotometry .. 45 2.3.3MergingtheISPICatalogPointingsandFilters ........... 48 4

PAGE 5

...................... 49 2.5ConstructingtheVISTACatalog ........................ 52 3ASTROMETRICMATCHINGOFTHEX-RAYANDNIRSOURCECATALOGS 60 3.1RealandSpuriousInfrared(IR)MatchestotheChandraX-rayCatalog 62 3.1.1DenitionsoftheCatalogProperties ................. 63 3.1.2EstimatingtheNumberofSpuriousNIRMatchestotheChandraCatalog ................................. 65 3.1.3TheFractionalExcessofX-raySourceswithNIRMatches ..... 70 3.1.4CharacteristicsoftheX-raySourceswithoutNIRMatches ..... 71 3.1.5ProbableMatchestoSoftX-raySources ............... 74 3.1.6ProbableMatchestoHardX-raySources .............. 74 3.2TotalNumberofRealCounterpartsintheISPICatalog ........... 76 3.2.1SoftSources .............................. 76 3.2.2HardSources .............................. 77 3.3ColorMagnitudeDiagram(CMD)PositionsofProbableRealMatches .. 78 3.3.1CMDofSoftSourceCounterparts .................. 79 3.3.2CMDofHardSourceCounterparts .................. 80 3.4ConclusionsfromAstrometricMatching ................... 83 4PHOTOMETRICVARIABILITYOFNIRCOUNTERPARTSTOX-RAYSOURCES 93 4.1PhotometricVariabilityinHardX-rayEmittingSystems ........... 94 4.1.1PhotometricVariabilityinHighMassX-rayBinaries ......... 94 4.1.2PhotometricVariabilityinCollidingWindBinaries .......... 95 4.1.3PhotometricVariabilityfromSymbioticStars ............. 95 4.1.4LowMassX-rayBinaryVariability ................... 95 4.1.5SourcesofStellarVariabilityUnrelatedtoX-rayActivity ....... 96 4.2Cross-matchingtheISPIandUKIDSSCatalogs ............... 97 4.2.1CutsonMagnitudeDuetoSaturationLimitsandEddingtonBias 99 4.2.2TheEffectofMismatchedKandKsFilterswithISPIandUKIDSS 99 4.2.3DeningtheVariabilityofCandidates ................. 100 4.2.4VariabilityofKnownLongPeriodVariablesintheGC ........ 101 4.3VariabilityoftheNIRCounterpartstoX-raySources ............ 102 4.4AddinganAdditionalEpochfromtheVISTACatalog ............ 105 4.5WhataretheX-rayMatchedVariables? ................... 108 4.5.1SpectrallyTypedVariableSourcesfromSpectroscopicFollow-up 109 4.5.2AretheX-raySourcesTrulyRelatedtotheVariables? ....... 110 5SPECTROSCOPICFOLLOWUPOFCANDIDATECOUNTERPARTSTOGCX-RAYSOURCES .................................. 130 5.1SpectroscopicTargetSelection ........................ 131 5.2SpectroscopicObservationsandReduction ................. 133 5

PAGE 6

........................ 133 5.2.2LUCIFER1DataReduction ...................... 135 5.2.3ObservationswiththeOhioStateInfraRedImager/Spectrometer(OSIRIS) ................................. 137 5.2.4OSIRISDataReduction ........................ 137 5.2.5ObservationswiththeFolded-portInfraRedEchellette(FIRE) ... 139 5.3DiscoveriesofTrueNIRCounterpartstoX-raySources .......... 139 5.3.1XID3275:aCandidateGalacticCenterBeHMXB ......... 140 5.3.2XID6592:aCandidateGalacticCenterSymbioticBinary ..... 144 5.3.3XID1944:anOsupergiantwithd<4kpc ............... 151 5.3.4XID947:aCandidateGalacticCenterOstar ............ 152 5.3.5ForegroundCounterparts:XID3178,XID3390andXID7214 ... 154 5.3.5.1XID3178 ........................... 154 5.3.5.2XID3390 ........................... 155 5.3.5.3XID7214 ........................... 156 5.4NIRCounterpartswithoutEmissionLines .................. 157 5.4.1SpectralTypingoftheSourceswithoutEmissionLines ....... 158 5.4.2CounterpartstoReddenedSoftX-raySources ........... 160 5.5StatisticsofSourceswithoutEmissionLines ................. 162 5.5.1ProbableCounterpartswithinGroup1 ................ 163 5.5.2ProbableCounterpartswithinGroup2 ................ 164 5.5.3ProbableCounterpartswithinGroup3 ................ 164 5.6WhataretheMissingX-rayCounterparts? .................. 165 6CONCLUSIONS ................................... 194 6.1StatisticsoftheNIRAstrometricMatchestoX-raySources ........ 194 6.2Variability .................................... 195 6.3SpectroscopicSurveys ............................. 196 6.4FutureWork ................................... 198 REFERENCES ....................................... 201 BIOGRAPHICALSKETCH ................................ 208 6

PAGE 7

Table page 1-1Relativecompactnessfordifferentobjects ..................... 35 2-1CharacteristicsofdifferentimagingdataoftheGalacticCenter(GC) ...... 54 2-2InfraredSidePortImager(ISPI)pointsourcecatalogcharacteristics ...... 55 2-3UnitedKingdomInfraredDeepSkySurvey(UKIDSS)pointsourcecatalogcharacteristics .................................... 56 2-4VisibleandInfraredSurveyTelescopeforAstronomy(VISTA)pointsourcecatalogcharacteristics ................................ 57 3-1NumberofX-raysourceswith0,1,2,or>3astrometricmatches ........ 85 3-2VariabledenitionsforcalculatingnumbersofrealcounterpartstoX-raysources,withunspeciedinfrared(IR)properties ...................... 86 3-3VariabledenitionsforcalculatingnumbersofrealcounterpartstoX-raysourceswithagivenIRcounterpartproperty ........................ 88 3-4RelativematchingfractionforX-raysources .................... 89 3-5MatchingstatisticsforsoftX-raysourcesmatchedtoasingleIRsource .... 89 3-6MatchingstatisticsforhardX-raysourcesmatchedtoasingleIRsource .... 90 3-7RangesofMKspossibleforrealsourcesincolormagnitudediagram(CMD)binsfromFigure 3-7 ................................. 90 4-1VariabilityinknowncounterpartstoGCX-raysources .............. 122 4-2VariablesourcecounterpartstosoftX-raysourceswith3variationsacrosstheISPIandUKIDSSobservationepochs ..................... 124 4-3VariablesourcecounterpartstohardX-raysourceswith3variationsacrosstheISPIandUKIDSSobservationepochs ..................... 126 4-4Tableofcandidatevariableswithdetected3variabilityinVISTA ........ 127 4-5SpectroscopicallyidentiedvariablesmatchedtoX-raysources ......... 128 5-1TableofspectroscopicobservationswiththeLargeBinocularTelescopeNearInfraredSpectroscopicUtilitywithCameraandIntegralFieldUnitforExtragalacticResearch1(LUCIFER1) ............................... 169 5-2TableofspectroscopicobservationswiththeOhioStateInfraRedImager/Spectrometer(OSIRIS) ....................................... 170 7

PAGE 8

......................................... 171 5-4NearinfraredphotometryofthecounterpartstoX-raysources,takenwiththeISPIinstrument .................................... 172 5-5Chandrasourcepositionsanduxes ........................ 173 5-6LineidenticationsandparametersforthecounterpartspectrumofXID3275 176 5-7LineidenticationsandparametersforthecounterpartspectrumofXID6592 178 5-82tforXID6592counterpart ........................... 179 5-9LineidenticationsandparametersforthecounterpartspectrumofXID1944 181 5-10ColorsandmagnitudesforFdwarfstars ...................... 184 5-112tforXID7214counterpart ........................... 186 5-12ResultsofspectralttingofNIRcounterpartstoX-raysources ......... 188 5-13VisualspectraltypingofNIRcounterpartstoX-raysources ........... 189 5-14ProposedclassicationfromvisualinspectionforreddenedNIRcounterpartstosoftX-raysources ................................. 191 5-15SpectraltypesandestimatedX-rayparametersforsourcesobservedfromGroup1 ........................................ 193 8

PAGE 9

Figure page 1-1TheX-rayskyasseenbytheHighEnergyAstronomyObservatory1(HEAO1) 33 1-2Grazingincidencemirrors .............................. 33 1-3TheoreticalX-rayextinctiontowardtheGalacticCenter(GC) .......... 34 1-4TypesofX-raybinaries ................................ 34 1-5DiagramofalowmassX-raybinary(LMXB) .................... 36 1-6OrbitsofstarsclosetoSagittarius(Sgr)A 36 1-7Near-infrared(NIR)colorimageoftheGC ..................... 37 1-8Therelationshipbetweensupermassiveblackholemassandbulgevelocitydispersion ....................................... 37 1-9ChandraimageoftheGC .............................. 38 1-10Fluxcolordistributionofsourcesinthe Munoetal. ( 2009 )catalog ....... 38 2-1InfraredSidePortImager(ISPI)imagesoftheGalacticCenter .......... 53 2-2ThedifferenceinKsmagnitudeversusmeanKsinanoverlappingregionoftheISPIGCeld ................................... 53 2-3Magnitudeversuserror,calculatedfromthestandarddeviationofKsinFigure 2-2 ........................................... 55 2-410020Hbandimageofanareawithintheoverlappingregion ......... 55 2-5ImageofthesameregionshowninFigure 2-4 withthevalueofeachpointrepresentingthemedianoffset ........................... 55 2-6100100HbandISPIimageofamatchedareabetweentheISPIandtheUnitedKingdomInfraredDeepSkySurvey(UKIDSS)datasets ......... 56 2-7100100imageofthelocalmedianoffsetbetweenthemagnitudeofstarsinISPIandinUKIDSSwithinFigure 2-6 ....................... 57 2-8ThedifferenceinKsmagnitudeversusmeanKsinanoverlappingregionoftheISPIGCeld ................................... 58 2-9Magnitudeversuserror,calculatedfromthestandarddeviationofKsinFigure 2-8 ........................................... 58 2-10ISPIphotometricerrorandcompleteness ..................... 59 9

PAGE 10

.................................. 85 3-2Colormagnitudediagram(CMD)oftheISPIcatalog ............... 86 3-3Color-colordiagramoftheISPIcatalog ...................... 87 3-4DistributionsofpositionalerrorsizesofChandraX-raysources ......... 87 3-5HistogramsoftheNIRcolorsofmatchestosoftandhardX-raysources .... 88 3-6HKs,KscolormagnitudediagramofallISPIsourceswiththepositionsofprobabletruecounterpartstosoftX-raysources ................. 91 3-7HKs,KscolormagnitudediagramofallISPIsourceswiththepositionsofprobabletruecounterpartstohardX-raysources ................. 92 4-1JmagnitudehistogramsfortheISPI,UKIDSSandcross-matchedISPI/UKIDSScatalogs ........................................ 112 4-2HmagnitudehistogramsfortheISPI,UKIDSSandcross-matchedISPI/UKIDSScatalogs ........................................ 112 4-3K=KsmagnitudehistogramsfortheISPI,UKIDSSandcross-matchedISPI/UKIDSScatalogs ........................................ 113 4-4JISPIversusJUKIDSSJISPIforallcross-matchedISPIandUKIDSSsources .. 113 4-5HISPIversusHUKIDSSHISPIforallcross-matchedISPIandUKIDSSsources 114 4-6Ks,ISPIversusKUKIDSSKs,ISPIforallcross-matchedISPIandUKIDSSsources 114 4-7DifferenceinJltermagnitudes ........................... 115 4-8DifferenceinHltermagnitudes .......................... 115 4-9DifferenceinK/Ksltermagnitudes ......................... 116 4-10Standarddeviationofthescatterinthemagnitudedifferencesforcross-matchedNIRsourcesinISPIandUKIDSS .......................... 116 4-11JISPImagnitudeversusJUKIDSSJISPIforNIRsourcesintheISPIGCeldwithvariablesourcesmarked ............................... 117 4-12HISPImagnitudeversusHUKIDSSHISPIforNIRsourcesintheISPIGCeldwithvariablesourcesmarked ............................ 117 4-13Ks,ISPImagnitudeversusKUKIDSSKs,ISPIforNIRsourcesintheISPIGCeldwithvariablesourcesmarked ............................ 118 10

PAGE 11

........................ 118 4-15HISPImagnitudeversusHUKIDSSHISPIforNIRsourcesintheISPIGCeldwithLPVsmarked .................................. 119 4-16Ks,ISPImagnitudeversusKUKIDSSKs,ISPIforNIRsourcesintheISPIGCeldwithLPVsmarked .................................. 119 4-17JISPImagnitudeversusJUKIDSSJISPIforNIRsourcesintheISPIGCeldwithX-raycounterpartsmarked ............................. 120 4-18HISPImagnitudeversusHUKIDSSHISPIforNIRsourcesintheISPIGCeldwithX-raycounterpartsmarked ........................... 120 4-19Ks,ISPImagnitudeversusKUKIDSSKs,ISPIforNIRsourcesintheISPIGCeldwithX-raycounterpartsmarked ........................... 121 4-20ISPIKsandUKIDSSKimagesoftheregionaroundthecounterparttoXID2983 .......................................... 121 4-21ISPIKsandUKIDSSKimagesoftheregionaroundthecounterparttoXID3687 .......................................... 123 4-22MagnitudedifferenceKVISTA-KISPIversusKISPIforthecross-matchedVISTA/ISPIcatalogs ........................................ 125 4-231errorsfromthecross-matchedKsmagnitudesofVISTAtoISPI ....... 125 4-24ISPIcolormagnitudediagramshowingthepositionsofvariablecandidatesmatchedtosoftX-raysourcesandhardX-raysources .............. 128 4-25ISPIcolormagnitudediagramshowingthepositionsoflongperiodvariables 129 5-1SpectraofLBV1806-10takenonsuccessivenights ................ 172 5-230"30"ndingchartsforthecounterparttoXID3275fromISPIdata .... 174 5-3Hbandspectrumoftheinfrared(IR)counterparttoX-raysourceXID3275 .. 175 5-4KbandspectrumoftheIRcounterparttoX-raysourceXID3275 ........ 175 5-5TheHKscolorofthecounterparttoXID3275comparedtothecolorsofallNIRsourceswithin20"ofitsposition ........................ 177 5-630"30"ndingchartsforthecounterparttoXID6592fromISPIdata .... 177 5-7KbandspectrumofthecounterparttoXID6592 ................. 178 5-8HandKsbandlightcurveoftheNIRcounterpartofXID6592 .......... 179 11

PAGE 12

................... 180 5-1030"30"ndingchartsforthecounterparttoXID1944fromISPIdata .... 180 5-11KbandOSIRISspectrumofthecounterparttoXID1944 ............ 181 5-1230"30"ndingchartsforthecounterparttoXID947fromISPIdata .... 182 5-13TheKbandspectrumofthecounterparttoXID947 ............... 183 5-14TheHKscolorofthecounterparttoXID947,comparedtothecolorsofallNIRsourceswithin20"ofthetargetposition ................... 183 5-15BrackettequivalentwidthsmeasuredforGandFdwarfs ............ 184 5-16KbandspectraofthecounterpartstoXID3178andXID3390 ......... 185 5-17KbandspectrumofthecounterparttoXID7214 ................. 186 5-18KbandspectrumoftheNIRcounterparttoXID7204 .............. 187 5-19KbandspectrumoftheNIRcounterparttoXID2212 .............. 187 5-20KbandspectrumoftheNIRcounterparttoXID861 ............... 190 5-21KbandspectrumoftheNIRcounterparttoXID6055 .............. 190 5-22KbandspectrumoftheNIRcounterparttoXID6010 .............. 191 5-23KbandspectrumoftheNIRcounterparttoXID6019 .............. 192 5-24PreliminaryNIRspectrumforthecounterparttoXID13 ............. 192 12

PAGE 13

13

PAGE 14

DeWittetal. 2010 ),thesecanbesecurelyidentiedasthetruecounterpartstotheircoincidentX-raypointsources.Lastly,Iusedtheresultsofmymatchingsimulationstoinferthepresenceof72truecounterpartswithinasetoflatetypegiantsthatIobservedwithoutdetectableemissionfeatures.IconcludefromthisworkthattheprobableexcessinredgiantX-raycounterpartswithoutemissionlinesneedstobeconrmedbothwithlargersamplesofspectroscopicallysurveyedcounterpartsandmoreadvancedstatisticalsimulationsofthematchauthenticity.Also,thenatureofthecompactobjectintwoofmycounterpartdiscoveries,theBeHMXBandthesymbioticbinary,canbestronglyconstrainedwithX-rayspectraltting.Lastly,IconcludethatspectroscopicsurveysfornewX-raysourcecounterpartsintheGCmaybeabletoincreasetheirefciencybyspecicallytargetingphotometricvariablesandverycloseastrometricmatchesofIR/X-raysources. 14

PAGE 15

Munoetal. 2009 ).ThemajorityofthesourcesarelikelytobeamixedpopulationofaccretingbinariesattheGalacticCenterdistance,buttheirlargenumbersandhard,faintX-rayemissionaredifculttoreconcilewithknownpopulationsofX-raysourcesintheGalaxy.Inthisdissertation,Iusenear-infrared(NIR)photometryandspectroscopytoidentifyasampleofthesesources.TheGalacticCenter(GC)isobservationallychallengingbecausethelineofsightgoesthroughtheGalacticPlane,wheremoleculargasanddustisconcentrated.TypicalvisualextinctionsareAV>30mag( Baganoffetal. 2003 ),whichprecludesthepossibilityforobservationsintheultra-violet,optical,andsoftX-raybands.Near-infraredlightexperiencesdramaticallylessextinctionfromtheGCdistance,withtypicalvaluesofAK2-3mag( Schodeletal. 2010 ).NIRobservingisavailablefromtheground,delivershighspatialresolution(<100)andcontainsmanydiagnosticspectralfeaturesofstarsandactivebinaries( Bandyopadhyayetal. 1999 ; Hansonetal. 2005 ; Rayneretal. 2009 ).TherewardsforidentifyingtheGCX-rayemittingpopulationofstarscanbeverygreat.Theprojected20.8regionaroundtheGalacticCenterincludes1%ofthestellarmassoftheGalaxy( Pfahletal. 2002 )injust0.004%oftheareaonthesky.ThisregionhasbeenobservedwithhighsensitivityatX-raywavelengthswiththeChandraX-rayObservatory,reachingLX1030ergss1forsourcesattheGCdistance( Munoetal. 2009 ).TheChandradatasetisthereforebroadenoughinscopetocoverextremelyrareobjectsandsensitiveenoughtorevealX-raysourcesthatareeitherpersistentlyfaintorthequiescentstatesofmorepowerfulbuttransientlyactivesources. 15

PAGE 16

E=12.4Ato0.01AandT=k E=1.2106Kto1.2108K,wherehisPlanck'sconstantandkistheBoltzmannconstant.Earth'satmosphereisopaquetoX-raysandtherstX-rayobservationsofspaceawaitedasoundingrocketightin1948( Seward&Charles 1995 ).Soundingrocketstraveltoheightsof100km,wheretheyareabletotakemeasurementsabovetheatmosphereforseveralminutesbeforefallingbacktoEarth.ThesunwasrevealedastherstastronomicalX-raysourcebutitsbrightnessisonlyduetoitsproximity;itsX-rayluminosityisbetween1026.8and1027.9ergss1( Judgeetal. 2003 ),whichamountsto107ofitstotal3.841033ergss1luminosityoverallwavelengths.IfeverystarhadsuchlowX-rayux,theearlyX-raydetectorswouldneedagaininsensitivityofmorethan105justtodetectthecloseststars( Seward&Charles 1995 ).Therefore,itwasasurprisewhenasourcewasdetectedinthedirectionofScorpiuswitha>4keVX-rayuxthatwasnearlyasbrightastheSun's.Thissource,namedScorpiusX-1,istherstknownX-raybinaryandthersthintofthepowerfulhighenergyprocessesthatX-rayastronomyisknownfortoday( Seward&Charles 1995 ).ScoX-1hasbeendeterminedtobeanaccretingneutronstar2.8kiloparsecsaway,generating1038ergss1inX-rays. 16

PAGE 17

1-1 .Figure 1-1 iscenteredontheGalacticCenterwiththeplaneoftheGalaxylyinginthemajoraxis.ThebrightestX-raysources,denotedbythelargeblackdots,aremostlyX-raybinaries(discussedinx1.4),andappeartobeconcentratedintheplaneoftheGalaxy.Thefainter,evenlydistributedsourcesontheskyareActiveGalacticNuclei(AGN)andmassiveclustersofgalaxieswithhotinterclusterplasma.TherehavebeendozensofrocketandsatelliteborneX-rayexperimentsfollowingsoundingrocketmissionsofthe1950-1960s.IwillsummarizethesemissionsbyhighlightingthefourthatmarkedlyadvancedX-rayimaging.Uhuru(1971)wastherstorbitingX-raysatellitededicatedtoastronomy.Overits2.5yearlifetimeitdiscovered339newX-raysources,includingAGNs,galaxyclusters,supernovaremnantsandX-raybinaries.However,itsimagingresolutionwasbetween0.55.LikepreviousX-raydetectorsinspace,Uhuruusedcollimatorsratherthanimagingoptics.Thesecylindricalshieldsrestrictedtheangularareathatreachedthedetector,butwithatradeoffincollectingareaandsensitivity.TheEinstein(1978)ObservatorywastherstX-raytelescopetouseimagingoptics,meaningitcouldcollectphotonsfromarelativelylargeareaandfocusthemintoanimageonthedetector.ThiswasatechnologicalchallengebecauseX-rayspassthroughmirrorsthatareusedface-onasinopticaltelescopes.Einstein'smirrorsreectedX-raysthroughgrazinganglesbyreplacingthecentrallyplacedlled-aperturemirroroftraditionaltelescopeswithahollowannulusofsteeplyslantedmirrorsintheshapeoftheoutersegmentsofaparaboloid.AnillustrationoftheChandramirrors,similartoEinstein's,isshowninFigure 1-2 .Theincreaseineffectivecollectingareaandthesharperimaging(FWHM500)madeEinsteinnearly1000timesmoresensitivethanUhuru( Seward&Charles 1995 ). 17

PAGE 18

Weisskopf 2010 ).Itscollectingareaandimagingsharpness(FWHM100)representedanother3ordersofmagnitudegaininsensitivityoverEinsteinandROSAT.Critically,forthestudydescribedinthisthesis,thesubarcsecondpositioningallowedustomatchtheX-raysourcepositionstosourcesinthedenselypackedNIReld(seeChapter3).DuringnormalimagingmodeswiththeACISinstrument,ChandrasimultaneouslyobtainslowresolutionX-rayspectroscopyaswellasarrivaltimesoftheincidentphotonsto3saccuracy. Carroll&Ostlie 1996 ).Accretiondisksalsoproduceblackbodyradiation.Inthiscasetheorbitalspeedsandenergyproducedbyviscousdissipationincreasewithsmallerseparationsfrom 18

PAGE 19

Pringle 1981 ).Toachieve1068KtemperaturesrequiredforX-rayemission,thediskneedstoextenddeepintothepotentialwellofacompactobjectsuchasaneutronstarorablackhole( Remillard&McClintock 2006 ).Twoothercontextswhereblackbodyradiationmattersisintheatmospheresofyoung,isolatedneutronstarsandinthermonuclearburstsonthesurfacesofneutronstars.Youngneutronstarshavebirthtemperaturesof>108K.Neutronstarburstsareaprocesswhereheliumandhydrogenthathaveaccretedontothesurfaceofaneutronstarundergorunawaynuclearburning( Campanaetal. 2001 ). Shu 1991 ). 19

PAGE 20

Carlstrometal. 2002 ).Inthehighenergycontext,inverseComptonscatteringoccursinthehotcoronaethatformabovetheaccretiondisksofX-raybinariesandactivegalacticnuclei.Inthesecircumstances,UV/opticalphotonsproducedbytheaccretiondiskcanbescattereduptohardX-rayenergies.TheresultantspectrumhasapowerlawshapeofFE/E)]TJ/F1 11.95 Tf 5.32 -4.33 TD[(,whereFEisthephotonuxatagivenenergyEand)]TJ/F1 11.95 Tf 10.09 0 TD[(iscalledthephotonindexandsetstheslopeofthespectrum.ThevalueofthephotonindexresultingfrominverseComptonscatteringis1-2. Wilmsetal. 2000 ).TheX-rayabsorptionlawconsistsofthesuperpositionofallphotoelectricabsorptioncross-sections,multipliedbytheabundanceoftheparticularelementresponsibleforthetransition.Anadditionalconsiderationisthefractionoftheseelementsthatgointodustgrains,calledthedepletionfactor.Metalsingrainsare 20

PAGE 21

Wilmsetal. 2000 ).Theshapeoftheextinctioncurve(shownintheupperpanelofFigure 1-3 )resemblesapower-law,withlowerenergiesmoreattenuatedthanhigherenergies,andjaggedfeaturesduetoprominentindividualtransitions.Theextinctioncurveisusuallyreferencedtothecolumndepthofhydrogen,eventhoughHgasisnotasignicantabsorberofX-rays.ThemagnitudeofvisualextinctionhasbeenmeasuredtocorrelatewithX-rayextinctionwithaconversionratioofNH Predehl&Schmitt 1995 ). 21

PAGE 22

Rc2( Shapiro&Teukolsky 1986 ).Table 1-1 liststherelativecompactnessesofthesun,anaveragewhitedwarf,aneutronstarandastellarmassblackhole.AlsolistedisthemaximumX-rayluminosityattainedwitheachobjectastheprimaryinaspeciedaccretingsystem.Clearly,therearelargejumpsintheenergyavailabletobeliberatedfromaccretionfordifferenttypesofprimaries.ThisfactisreectedinthemaximumX-rayluminositiesobservedinsystemscontainingtheseprimarytypes.TheclassicationofX-raybinariesisoftendoneonthebasisofthecompactobjecttypeandthemassofthesecondarydonorstar.Figure 1-4 illustratestheclassicationofX-raybinaries,splitrstintocompactobjecttypes,andthenintodonorstarmasses. Liuetal. 2006 );60%oftheentriesareBeHMXBsand32%aresupergiantHMXBs.Theremaining8%haveX-raypropertieswhichsuggestaHMXBdesignationbutforwhichthesecondarystarhasnotyetbeentyped.BeHMXBscontainasecondaryOorBstarrotatingnearitsbreakupspeed( Reig 2011 ).Thesestarsformequatorialgasdiskswhichareoftenseenwithemissionlinesintheoptical/NIRduetothehotradiationfromtheO/Bstarand,accordingly,areclassiedwithOeorBespectraltypes( Steele&Clark 2001 ).BeHMXBswithblackholeprimariesmayexist,buttodate,nonehavebeenconrmed( Reig 2011 ).TheneutronstarsinBeHMXBsareusuallyoneccentricorbitsduetoasymmetrickicksduringtheirsupernovaphases.ThiscausestheNSstoundergoclosepassagestotheBestarsecondarieswheretheycanaccreteanddisrupttheBeequatorialgasdisks. 22

PAGE 23

Reig 2011 ).AsX-raysources,BeHMXBsareseentohavehardpowerlawspectra,with)-346(=0.52andluminositiesthatrangebetween1033-1038dependingonwhethertheyareobservedduringquiescenceoroutburst( Munoetal. 2003b ).SomeBeHMXBshavesmalleccentricitiesandperiodsabove100days.Thesesourcesarerelativelyfainter(Lx<1035ergss1)andhavelessdramaticoutbursts( Reig 2011 ).Thecircumstellarmaterialand/orstellarwindsinthesesystemsarecapturedviaBondi-Hoyleaccretion.Bondi-Hoyleaccretionisaresultofthegravitationalfocusingthatoccurswhenastarmovesthroughagaseousmedium.Asgastrajectoriesarechanged,theresultingdensitiesareincreasedtothepointwherethegascanexperienceshocksagainstotheraccretinggas,whichmakesthematerialfallintowardthecentralmass.Dependingontheresidualangularmomentuminthegas,thematerialcaneitherbeaccreteddirectlyontothestar,orcollapseintoanaccretiondiskwherethematteriscarriedinwardbyviscousdissipationprocessesasinotheraccretionsystems.SupergiantHighMassX-rayBinaries(sgHMXBs)containabluesupergiantstarorWolf-Rayetstarinorbitsoftensofdays.BluesupergiantsandWolf-Rayetshavemassesof20-50Mandproducemassivewindswithmasslossratesofupto106Myr1( Reig 2011 ).Thesesystemshavebeenseentocontainbothblackholesandneutronstars( Reig 2011 ).Althoughtheyusuallyaccreteviawindcapture,inthreecasestheorbitalseparationissmallenoughthatdisk-fedaccretionoccursdirectly( Reig 2011 ).sgHMXBshaveasimilarluminosityrangeandspectrumtotheBeHMXBs,butordinarilytheyarepersistentsources.ThemajorexceptionisforthenewlyidentiedclassofSuperFastX-rayTransients(SFXTs)discoveredwiththeINTEGRALsatellite.Thesesourcesexhibitaresof1036-1037ergss1,thatlastfordurationsofafewhoursbeforereturningtoalowquiescentstate( Reig 2011 ).Thesesourceshavebeen 23

PAGE 24

Grimmetal. ( 2003 )ndthatthesurfacedensityofHMXBscorrelateswiththestarformationrateingalaxiesasderivedfromothermethods.ForareasofheavyextinctionsuchastheGalacticCenter,commonlyusedstarformationindicatorsareoftenunavailable.InthesecircumstancesHMXBscouldbeusedtomoreeffectivelyprobethestarformationrates.BinarysynthesissimulationsofHMXBssuggestthattheremaybeasmanyas10000HMXBsintheGalaxyatthecurrentepoch( Meurs&vandenHeuvel 1989 ). Ritter&Kolb 2003 ),contains71LMXBs.LMXBsaccreteviaRocheLobeoverowofthesecondarystar'senvelope.TheRochelobeistheequipotentialsurfacebetweentwogravitatingbodies.WhenmatterspillsacrosstheRochelobeintothedomaindominatedbythecompactprimary,ittypicallyattensintoanaccretiondisk( Paczynski 1971 ).AdiagramofaLMXBisshowninFigure 1-5 .LMXBscreateX-raysbyavarietyofmeans,includinginneraccretiondiskblackbodyemission,accretionshocksonthesurfaceoftheneutronstar,shocksinthejetemissionanddiskwinds,neutronstarthermonuclearburstsandapersistenthardinverseComptoncomponent( Remillard&McClintock 2006 ).ThemaximumluminosityofLMXBswithblackholesorneutronstarsisbetween1037and1039.However,morethan30%ofLMXBsaretransient,andspendmostoftheirtimeinaquiescentstate( Tanaka&Shibazaki 1996 ).Thequiescentstatesfor 24

PAGE 25

Asaietal. 1998 ).BothtypesofLMXBaredominatedbya)]TJ/F2 11.95 Tf 11.01 0 TD[(2powerlawphotonindexspectrumduringperiodsofquiescence.ForBHtransients,theperiodsbetweenoutburstscanlastbetween1-60years( Remillard&McClintock 2006 ).Thesesystemsareextremelyusefulaslaboratoriesofextremephysics.AlongwiththesgHMXBCygX-1,thedynamicallydeterminedmassmeasurementsofLMXBsystemcomponentshaveproventheexistenceofblackholes( Remillard&McClintock 2006 ).ForNSLMXBs,nuclearburstluminositiesandtemperatureshavepermittedthemeasurementofneutronstarradiiandmasses.Spectralttingand/orobservationsoftheX-rayvariabilitymayeventuallyleadtothemeasurementofblackholespins( Remillard&McClintock 2006 ).Asastellarpopulationindicator,thenumbersofLMXBstracetheamountofaccumulatedmassinastellarpopulation( Gilfanov 2004 ).InregionsofdeepextinctionandstellarcrowdingsuchastheGC,faint,lowmassstarsaredifculttoobserve.LMXBscouldbeakeytraceroftheoldstellarpopulationwhichconstitutesthemajorityofthestellarmass.LMXBsareveryrare.Binarysynthesiscalculationstypicallypredict1000totalintheGalaxy,withmostlyingunseeninquiescence( Pfahletal. 2002 ; Munoetal. 2003a ). 1-1 )thesesystemscannotreachthesamebrightX-rayluminositiesasLMXBs.CataclysmicVariables(CVs)containanumberofsubtypes,suchasclassicalnovae,dwarfnovae,andpolars.Thenovaetypevariablesaretheresultofthermonucleardetonationsofaccretedhydrogenonthesurfaceofthewhitedwarf.Thesedetonations 25

PAGE 26

Munoetal. 2003b ).TheexceptionisforthemagneticCVs,includingpolarsandintermediatepolars.PolarsarenamedbecauseofthehighWDmagneticelds(106-108Gauss)thatcausecircularpolarizationintheiremission.Inaccretionontosuchmagnetizedobjects,theaccretedmatterisentrainedtoowontothepolarregionsoftheWD,wherethemagneticeldsarestrongest.ThismeansthatthedensityoftheaccretedmaterialishigherthanforunmagnetizedWDs,andtherebygeneratesmoreenergeticshockswhenthematterhitstheWDsurface.Theshock-heatedplasmacreatedduringaccretionontomagneticWDscanreachtemperaturesof25keV( Munoetal. 2009 ),makingmagneticCVssourcesofhardX-rayemission.Thenumberofcataclysmicvariablesofalltypeshasbeenestimatedtobe106intheGalaxy( Howelletal. 2001 ). Belczynskietal. 2000 ).Nearly80%ofsymbioticshaveanormalMgiantsecondary,whiletheother20%containasemi-regularvariableorMiravariablestar.OnlyasmallnumberofsymbioticstarshavebeenseentoemithardX-rays.Inonecase,thisisprobablyduetothecompactobjectbeinganeutronstar( Pauletal. 2005 )andinothercases,internalshockswithintheaccretionowmaybeabletogenerate 26

PAGE 27

Luna&Sokoloski 2007 ; Lunaetal. 2010 ).Frombinarysynthesissimulations,thenumberofsymbioticstarsintheGalaxyisestimatedtobebetween8000and15000( Luetal. 2007 ).However,thenumberofconrmedsymbioticstarsatthetimeofthe Belczynskietal. ( 2000 )catalogwasjust188. Mauerhanetal. 2010a ),andtheLxnormallyscaleslinearlywiththebolometricluminositybyafactorofLx Oskinovaetal. 2009 ).FortypicalOI/III/Vstarluminosities,thistranslatestoLx=1031to1033ergss1( Martins&Plez 2006 ).Sometimesthesewindemittingsystemsexistinbinarieswithasecondwindemittingstar.Inthesecases,thecollidingwindsattainhigherrelativespeedsthanforself-shockedwinds,increasingtheplasmatemperatureasT/v2rel( Draine&McKee 1993 ).Theplasmatemperatureinwindcollidingbinariesreaches6keV( Munoetal. 2003b ),makingthemeffectivehardX-rayemittersintheChandraenergyrange. Goslingetal. 2009 ).Therefore,itwaswithrelativelyunaffectedcentimeterradiowavesthatthepreciselocationoftheobjectatthedynamicalcenteroftheMilkyWaywaseventuallydiscoveredin1974( Balick&Brown 1974 ).ThisobjectwascalledSagittariusAtodistinguishitfromtheradioemissioncomplexSgrAinwhichitresided.SgrAistheradioemittingplasmasurroundingasuper-massiveblackholewithmass3.6106Mmeasuredbytheorbital 27

PAGE 28

1-6 )( Ghezetal. 2005 ; Gillessenetal. 2009 ).SgrAiscenteredwithintheGalacticBulge,aspheroidaldistributionofstarsdistinctfromtheGalacticDisk,withanageofabout10Gyr( Zoccalietal. 2006 ).However,studiesoftheinnerbulgendamixofold,intermediateageandyoungstellarpopulations( Blumetal. 2003 ).At200pcfromSgrAistheCentralMolecularZone(CMZ),adiskshapedstructureseeninmillimeterCOemissionandrepresentingwarmmoleculargassurroundingthecentralarea( Morris&Serabyn 1996 ).TheCMZaccretesgasfromtheGalacticdiskviaaninstabilitycausedbytheGalacticBar( Binneyetal. 1991 ).TheGCisthusfedasupplyofmoleculargasbutwhatexactlyhappensindetailtothegasafteritarrivesinthecentralregionisunknown.Near-infraredphotometryandspectroscopyofGCstarsshowsaclusterofmassivehotstarsaroundSgrA( Krabbeetal. 1991 ).Thesestarsonlylivefor10Myr,anditisunclearhowtheycouldforminthepresenceofsuchenormousshearingpressureandgastemperatures,orarriveatthislocationfromsomewhereelse( Morris&Serabyn 1996 ).Figure 1-7 showsaJHKscolorimageoftheGCfrom2MASS,withthecentralstarclusterandthelocationofSgrAatthecenteroftheimage.ThemysteryoftheGCstarformationisdeepenedbythediscoveryofdozensofhighmasssupergiantsandWolf-Rayetstarsthatareseeminglyunassociatedwithanyknowncluster( Mauerhanetal. 2010a ).SuchmassivestarsarenormallyonlyformedinthemostmassiveclustersinthespiralarmsoftheMilkyWay.StudiesoftheCentralStarClusterintheGCndatop-heavydistributionofstarformation,whereunusuallylargenumbersofmassivestarsareproducedcomparedtolowermassstars.( Manessetal. 2007 ; Bartkoetal. 2010 ).ThestarformationrateintheGCisinevitablylinkedtotheavailablegasreservoirthathascollectedintheCMZ.ThesamereservoiralsocontrolsSgrA'saccretionrate.Inprinciple,thestarformationrateintheGCcanbecalculatedforthepastseveralGyr, 28

PAGE 29

Gebhardtetal. 2000 ).Indeed,SgrAistheclosestblackhole/bulgesystemthatfollowstheM-sigmarelationship,andthereforeaffordsthemostdetailedviewoftheprocessesatwork. Worralletal. 1982 ).ThespectraofthisemissionindicatesaplasmatemperatureofnearlyT=108K( Yamasakietal. 1997 ).Ifitisreallydiffuseasitappearsratherthanbeingcomprisedoffaintdiscretesources,thenitsexistenceisatheoreticalchallengebecauseitsinferredtemperatureisabovetheescapespeedfortheGalacticplane.Thehotplasmawouldthereforeneedreplenishment;e.g.intheformofasupernovarate10greaterthanwhatisinferredfortheinnerGalaxy( Revnivtsevetal. 2009 ).WhenChandrarstobservedtheGalacticCenterregionhowever,theenormousresolutionandsensitivitygainallowedthediscoveryofnumerousfaint,hardX-rayemittingpointsources. Wangetal. ( 2002 )uncovered1000suchpointsourcesina20.8eldwhilesearchingforthesourceoftheHe-likeFeemissionemanatingfromtheGC.GravitationallycompactpointsourcescaneffectivelyholdontotheT=108Kplasma,andbyresolving50-80%oftheGalacticRidgeemissionintothispopulationofindividualsources,Chandrahaspartiallysolvedalong-standingmysteryoftheinnerGalaxy( Wangetal. 2002 ; Revnivtsevetal. 2009 ).Itisanopenquestion 29

PAGE 30

Revnivtsevetal. 2009 ).However,initsplace,wemustask,whatarethe1000X-raypointsourcesdiscoveredby Wangetal. ( 2002 )?Toinvestigatethenatureofthesepointsources, Munoetal. ( 2003b )re-observedthecentral170170aroundSgrAforanadditional590ks,increasingthetotalobservingtimeto626ksandthenumberofdetectedX-raypointsourcesjustwithinthissmallerregionto2357.Morethan2000ofthesourceswereundetectedbelow1.5keV,indicatingalargeamountofextinction.Theline-of-sighttotheGalacticCenterhasanaveragecolumndensityofNH=61022cm1( Baganoffetal. 2003 ),whichcausestheuxbelow1.5keVtoexperiencemorethan=2ofattenuation( Tan&Draine 2004 ).ThissuggeststhatthehardX-raysourceswithattenuatedemissionbelow1.5keVlieatorbeyondtheGCdistance.InordertosorttheX-raysourcesintoprobableforegroundandbackgroundsources, Munoetal. ( 2003b )usedthesofthardnessratio,denedasHR0=hs h+s,wherehisthephotoncountsofasourcebetween2.0
PAGE 31

Munoetal. ( 2009 )plotthesurfacedensityofhardX-raysourcesagainsttheNIRderivedstellardensityandndthatbothhavesharpmaximumdensitiesatthesameGalacticlatitudeandlongitude.TheimplicationisthatthehardX-raysourcedistributionfollowsthatoftheGalacticCenterstellarpopulation.ThesoftforegroundX-raysourcescouldintheorybelongtoanyofthesourcetypesenumeratedinx1.4basedontheirX-raySEDs;howevertheirlowX-rayluminositysuggeststhattheymaybemostlikelytobeCVs,RSCVnsorcoronallyactivesinglestars.ForthehardX-raysources,onlyLMXBs,HMXBs,wind-collidingmassivebinaries,symbioticstarsandmagneticCVshavebeenshowntoproduceenoughhardX-raystobedetectablethroughthelargeextinctiontotheGC.Themajorityofthesourcesaretoofaintforanadequatecharacterizationoftheirspectrawhichwouldallowustodisentanglethedifferentemissionmechanismsandtheextinction( Munoetal. 2009 ).InFigure 1-10 ,hardX-raysourcesareplottedbytheirhardcoloragainsttheirphotoncountsinthehardband.Theplotshowstheluminositiesofthesourcesforfree-freeplasmamodelsandpowerlawemission.Clearly,sincethevastmajorityofsourcesarefainterthan1033ergss1,wemustlooktointrinsicallyfaintorquiescentXRBsaspossiblecounterparts.WithinthisnewlyidentiedGCpopulation,therehavebeen13brightsourcesidentiedonthebasisoftheirX-rayemissionalone,including7LMXBsdiscoveredbytheiroutburstcharacteristics( Munoetal. 2005 ),and8likelyintermediatepolarsidentiedonthebasisoftheirperiodicX-rayvariability( Munoetal. 2003a ).However,thevastmajorityofX-raysourcesinthecatalogaretoofainttoperformsimilaridenticationswithonlytheX-rayemission. 31

PAGE 32

32

PAGE 33

TheX-rayskyasseenbytheHighEnergyAstronomyObservatory1(HEAO1)andlabeledwithwellknownX-raysources( Woodetal. 1984 ). Figure1-2. GrazingincidencemirrorsforChandraX-rayObservatory. 33

PAGE 34

TheoreticalX-rayextinctiontowardtheGalacticCenter(GC),assumingavalueofAK=3.5mag( Tan&Draine 2004 ). Figure1-4. TypesofX-raybinaries( Reig 2011 ). 34

PAGE 35

Relativecompactnessfordifferentobjectsandtheirtypicalmaximumluminositieswheninactivelyaccretingbinaries( Shapiro&Teukolsky 1986 ; Munoetal. 2003b ). StellarObjectRelativeCompactnessTypicalmaximumLX(ergss1)SystemType Sun1061032AlgolTypebinaryWhiteDwarf1041033CataclysmicVariableNeutronStar1011038NSLMXB/HMXBBlackHole11039BHLMXB(stellarmass)

PAGE 36

DiagramofalowmassX-raybinary(LMXB).(courtesyR.Hynes). Figure1-6. OrbitsofstarsclosetoSagittarius(Sgr)Aobservedover7yearswithKeckadaptiveoptics,whichdemonstratethepresenceofasupermassiveblackhole( Ghezetal. 2005 ). 36

PAGE 37

Near-Infrared(NIR)ColorimageoftheGCovera0.60.7region,takenwiththe2MASSsurvey.SgrAislocatedinthebrightcentralclusteroftheimage,butisoverwhelmedbythelightofthesurroundingcentralclusterstars. Figure1-8. (RightPanel)Therelationshipbetweensupermassiveblackholemassandbulgevelocitydispersionforgalaxiesasmeasuredbyreverberationmapping( Gebhardtetal. 2000 ).Theleftpanelshowstheblackholemassversusbulgeluminosity. 37

PAGE 38

Munoetal. 2009 ). Figure1-10. Fluxcolordistributionofsourcesinthe Munoetal. ( 2009 )catalog. 38

PAGE 39

Munoetal. ( 2009 ).Wetargetedthecentral170170regionaroundSgrAbecausethisregionhasmoreaccumulatedexposuretimeandagreaternumberofdetectedpointsourceswithChandrathananyotherregioninthe20.8catalog.WeproducedthreecatalogsofNIRphotometryinthiswork,utilizingJHKsimagingwithISPIattheCTIO4mtelescope,JHKimagingfromtheUKIDSSGalacticPlaneSurvey( Lucasetal. 2008 )andJHKsimagingfromtheVISTAEarlySciencedata( Emersonetal. 2006 ).WeusedtheISPIdatatocreateacatalogofNIRcounterpartstothe Munoetal. ( 2009 )X-raycatalog,whiletheISPI,UKIDSSandVISTAcatalogswereusedtogethertondphotometricvariabilityinthecandidatecounterparts.Wealsousedthe2MASSallskydatareleasecatalog( Skrutskieetal. 2006 )tocalibratetheastrometryandphotometriczeropointsfortheISPI,UKIDSSandVISTAdatasets.ThoughmanyotherphotometricobservationshavebeenmadeoftheGalacticCenter,thesethreedatasetscomparefavorablytootherrecentworkintermsofthecoverageareaandthesensitivity(expressedinTable 2-1 asthe50%completenessmagnitudelimit,i.e.themagnitudewhere1 2ofthestarsaredetected).Thecompletenesslimitisafunctionofthestellarsurfacedensityandsharpnessoftheimaging( Olsenetal. 2003 ).OurthreedatasetseachhaveatmosphericseeingvaluesofFWHM=000.9.InFigure 2-1 welistthecompletenesslimits,ltersandareaofseveralrepresentativeGCNIRobservations.Therstfourdatasets(2MASS,VLT/ISAAC,Magellan/PANIC,andIRSF/SIRIUS),haveallbeenusedforlocatingNIRcounterpartstotheGalacticCenter( Bandyopadhyayetal. 2005 ; Laycocketal. 2005 ; Miklesetal. 2006 ; Mauerhan 39

PAGE 40

, 2009 ).Thenextthreedatasetsareusedinthiswork.ThelasttworepresentthecurrentstateoftheartwithspacebasedNIRphotometryandadaptiveopticsimagingoftheGalacticCenter.Thesedatasetswillbeinvaluableinotherways,butforourpurposeofdiscoveringtheX-raystellarpopulationsoverawideeld,theHubble/NICMOSPaemissionlinesurveywillhavelittleleverageforndingstellarcolors,andthewidesteldAOobservationswithVLT/NAOSarestilllessthan1arcmin2insurveyarea( Schodeletal. 2010 ).OurdatasetsareexceededindepthbytheVLT/ISAACobservationsof Bandyopadhyayetal. ( 2005 )andtheMAGELLAN/PANICobservationsof Laycocketal. ( 2005 ),buttheseobservationscoveredsmallerareasoftheGC.Therefore,forthetimebeing,ourphotometricdataoftheGCiscompetitivewithwhatisavailableforourtargeteldsize. ( 2009 )compiled88ChandraobservationsoftheGalacticCentertotaling2Ms,takenbetweenApril2000andAugust2007andnd9017pointsourcesovera20.8eld.The170170eldcenteredonSgrAhasbeenthetargetof1MsofexposureswithChandratodateandhassomeofthebestsensitivity(41031ergss1forGCsourcesat8.3kpcdistance)andthebestpositionalaccuracyintheX-raycatalog.76%oftheX-raysourcesinthisregionhavepositionalaccuraciesof0.700orbetter. 2.2.1DatafromtheInfraredSidePortImager(ISPI)WeusedtheISPIcameraontheCTIO4mBlancoTelescopetoobservethe170170centralregionoftheGC,coveringtheregionofhighestsourcedensityandsensitivityintheChandraX-raydata.ISPIhasa10.50squareeldofviewanda0.300pixel1platescale( vanderBlieketal. 2004 ).Theobservationscoveredthe170170regioninJ,HandKsltersonAugust10,2005,attainingthefulleldwithatotalof4pointings.Theindividualexposuretimeswereshort,3.2sforHandKsand5sforJ,so 40

PAGE 41

2-1 showsa3-colorcompositeimageofthefull170170eldanddemonstratestheextremestellarcrowdinginthisregion.PanelBofFigure 2-1 displaysthepointingpatternofourobservations. Lawrenceetal. 2007 ).OneofthesubsurveysistheGalacticPlaneSurvey(GPS),whichincludesJHKcoverageoftheregionbetween5
PAGE 42

Minnitietal. 2010 ).TheGalacticCenterregionwassurveyedaspartoftheearlysciencedemonstrationprogramoftheVVVsurveyon15August,2010withaditherpatternof12exposuresperlterof6s,4sand4sexposurelengthsforJHKs,respectively.Theseeingwas0.900fortheimageintheregionwhichoverlappedwithourISPIdata.TheVVVsurveyisdesignedtoreach5limitsofJ=20.2H=18.2Ks=18.1( Arnaboldietal. 2010 ). Skrutskieetal. 2006 ).Theastrometricprecisionof2MASSis0.1500RMS.WeusedthiscatalogforcalculatingtheISPIimagedistortionanddeningtheastrometricplatescaleoftheimages.Thedepthofthe2MASSsurveyissignicantlyshallowerthantheISPIdata,with10limitsofJ=15.8mag,H=15.1magandKs=14.3magandapointspreadfunction(PSF)fullwidthhalfmaximum(FWHM)whichwastypicallybetween2.500and300.TheseverecrowdingintheGCmakestheeffectivemagnitudelimitsevenbrighterthanthesenominalvalues.ThecombinationofthehighcrowdingandlargerPSFof2MASSmeantthatmanystarsthatcouldhavebeenusedasphotometricreferencesforourISPIeldwerebadlyblendedin2MASS. 42

PAGE 43

2.3.1CreatingtheISPIcatalogWereducedtheISPIimageswiththeFATBOYpipeline(FloridaAnalysisToolBornOfYearningforhighqualitydata),animagingandspectroscopypackagedevelopedattheUniversityofFlorida( Warneretal. 2008 ).FATBOYusesPythonroutinessimilartothestandardimagereductiontasksfoundinIRAF.Theprocessingstepsincludepixellinearization,darksubtraction,at-eldcorrection,cosmic-rayremoval,sky-backgroundsubtractionandimagestacking.TherawISPIdatacontainedimagedistortionswithanamplitudeof100.5.WemadeadistortionmapforeachlterusingtheIRAFtask,msctpeak.Thisprogramoverlaysasetofknownsourcecoordinates(from2MASS,inthiscase)ontothedistortion-affectedimage.Theuserselectsstarsfromtheimagethatmatchtheoverlaidcoordinatesandtheprogramtsa2-Dpolynomialtothesetofmatchedpositions.msctpeaksavesthetcoefcientsintotheWorldCoordinateSystem(WCS)sectionoftheimageheader.Inourimplementation,weselecteda2-D6thorderpolynomial,thehighestorderavailableintheprogram.TheIRAFfunctionmscimageisdesignedtobethenextstepafterusingmsctpeak.Itinterpolatestheimagedataontoanimagearraywithauniformplatescale.Atthispoint,thedistortion-correctedimagescanbealignedandstackedwithinIRAF.However,theFATBOYpipelinehasasuperiormethodforimagestackingtothestandardIRAFroutines.ThereforeweconvertedthepolynomialtcoefcientstotheleformatusedbyFATBOYandcompletedtheimagestacking.Atthisstagewehad5imagesineachlterwhichtogethercoveredthefull170170eld.TheresidualRMSerrorofthedistortioncorrectedISPIplatesolutionwithrespectto2MASSwas0.200,comparedtotheuncorrectedISPIimagewhichhadanRMSresidualerrorof0.700andtheinternalastrometricprecisionof2MASSof0.1500( Skrutskieetal. 2006 ). 43

PAGE 44

Stetson 1987 ).ThefunctionofDAOPHOTistodetectsourcesinanimage,manipulatetheimagetocleantheregionaroundselectedPSFstarsofcontaminatingsources,andtamodeltothePSFstars.ALLSTARusesthePSFmodelderivedbyDAOPHOTtosimultaneouslytallthesourceswithinanimage.Weiterativelybuilt2-DGaussianmodelPSFswithalook-uptableand3rdorderpositionalvariationsforeachimage.Thelook-uptablereferstoadditionalPSFfeaturesnottbytheanalyticcomponentofthemodelPSF.Thesefeatures,suchasnon-Gaussianwingsanddiffractionspikes,systematicallyappearintheresidualsofthet,andDAOPHOTsavesthemintothelook-uptable,anarraywhichisusedinconjunctionwiththeanalyticPSFmodel.Themeanreduced-squaredvalueofourPSFmodelforourPSFstarswas0.010.02forallimagesandlters.Weusedonlyonepassforsourcedetection.ItiscommontoperformtwoormorepasseswithDAOPHOTforsourcedetection.Normally,onerunstheseconddetectionpassontheresidualimageafterALLSTARhastandsubtractedalloftherstpasssources.However,inourcase,wefoundthatasecondpassdetectedaunusuallylargenumberoffalsedetectionsinthetresidualsaroundbrightandmoderatelybrightstars.ALLSTARshouldrejectspuriousstarsduringthesecondttingiteration,butwefoundthathundredsofspuriousdetectionswerenotbeingrejectedandthuscorruptingthephotometryoftherstpassstars.Therefore,weonlyusedonepassbutmadesuretocarefullyaccountforcompletenessusingarticialstartests(x2.3.3). 44

PAGE 45

2-1B )intheJ,HandKsbands.Wematchedthecoordinates 45

PAGE 46

Miller 2007 ),andrecordingonlytheclosestmatchedsources.Multiplematchesoccurredforfewerthan0.5%ofsources.Figure 2-2 showsthedifferenceinKsmagnitudesbetweenpointings,Ks,1Ks,2,versustheKs,1magnitudefora20100regionofoverlappingcoverageintheISPIGCdata.Evenatbrightmagnitudesthescatterisnearly0.1mag.Wemeasuredthephotometricerrorbycalculatingthe3-clippedstandarddeviationofthemagnitudedifferencesshowninFigure 2-2 ,anddividingtheresultbyp 2-3 .Thisplotshowsthattheerrorlevelsoutatabout0.08mag;thisislargecomparedtootherPSF-tJHKsphotometryofthiseldintheliterature,whichtypicallyreportbesterrorsof0.01-0.03mag(e.g. Mauerhanetal. ( 2010a )).Thereasonforthislargeerrorooristhespatialvariationsinthezeropoint.InFigure 2-4 weshowaregionofISPIeldwithoverlappingcoveragebetweentwoimagesintheHband.In 2-5 ,weshowthesameregion,butforeachpixelweshowmedianvalueofthemagnitudedifferencesH2H1forthenearest150sourcestothepixelcoordinate.Thisquantiesthelocalvariationsinthezeropointbetweenthetwoimagesforeachskyposition.However,thisquanticationisnotusefulatthisstagebecausethemagnitudedifferencesmayarisefromeitheroneorbothoftheimages,whichpreventsthederivationofauniquecorrectiontothemagnitudesforeachpositionandimage.Thezeropointingproblemwasobservedinallltersandoverlappingregions,withnoconsistentspatialpatternwithrespecttothedetector.Tocorrectthespatiallyvariantzeropointproblem,werequiredacalibratedsetofNIRphotometrywhichcompletelycoveredtheISPIGCeldandwaswellmatchedtotheseeingconditionsoftheISPI 46

PAGE 47

2-7 weshowthezeropointvariationswederivedfortheISPIHbandimageshowninFigure 2-6 .Figure 2-7 showsthevalueofthemedianphotometricdifferenceinthevicinityofeachpixelpositionforthe150nearestISPI/UKIDSSmatchedstars.WecorrectedthephotometryofeachstarineachpointingandlterbysubtractingthemedianoftheregionalphotometricdifferencesbetweenISPIandUKIDSS.ReturningtothemagnitudedifferencesfoundintheISPIoverlapregions,wefoundamarkedreductioninthepermagnitudephotometricscatter.InFigure 2-8 weshowthedifferenceinthenewlycorrectedKs,1andKs,2magnitudesversusKs,1.Figure 2-9 showstheupdatedmagnitudeversuserrorderivedfromthestandarddeviationofthemagnitudedifferencesfromFigure 2-8 .Forthisparticularoverlapregion,thenoiseoorimprovedfrom0.08magto0.03mag.SimilarimprovementsweremadeforalloverlapregionsinJ,HandKs. 47

PAGE 48

48

PAGE 49

2-1 andmakesupjust2%ofthetotal170170areaoftheISPIGCeld.Foreach0.2magbin,wecalculatedthephotometricerrorsbyonceagainusingthephotometryoftheoverlappingregions,andcalculatingthe3-clippedstandarddeviation,dividedbyp 2-10 .WecalculatedthecompletenessfractionofourcatalogsbyinjectingarticialstarsofvariousmagnitudesintotheimageswiththeDAOPHOTtooladdstar.Thestarswereoverlaidontheoriginalimagesinasparserectangulargridwherethemagnitudesandxandyvaluesofthegridpositionwererandomlyassigned.Wemadesurethatthemagnituderangecoveredtheintervalwherebetween1%and95%ofstarswererecovered.Werestrictedthenumberofinjectedstarstofewerthan10%ofthenumberofstarsalreadycontainedintheimageinordertopreventtheintroductionofdifferentcrowdingcharacteristics.TheDAOPHOTandALLSTARparameterswerekeptthesameasintheoriginalreduction.Wecountedanarticialstarasrecoveredifitwasfoundwithin0.500ofitsinjectedpositionandweexcludedanyarticialstarsthatfellwithin0.500ofarealstarintheISPIimages.Figure 2-10 showsthecompletenessfractionversusmagnitudeforarticialstarsfortheJ,H,andKslters.ThesevaluesshouldreecttypicalregionsintheISPIeld,butthelocalcompletenesslimitwilllikelyvarywiththesourcedensitychangesacrosstheeld.InTable 2-2 weshowthenumberofsourcesinthenalmergedISPIcatalog,the5limitingmagnitudepredictedbytheISPIexposurecalculatorfromPoissonstatistics,the5magnitudelimit(wherethemagnitudeerroris0.2mag)calculatedfromthephotometryofstarsobservedmultipletimes,andthemean50%completenessmagnitudeforeachofthelters. Irwinetal. ( 2004 ).FiveUKIDSSpointingsperlter 49

PAGE 50

Lucasetal. 2008 ).WematchedourUKIDSSPSF-ttingphotometrytothepipeline-producedUKIDSScatalogandfounda0.04magstandarddeviationofthemagnitudedifferenceforbrightunsaturatedstarsbetweenourPSF-tcatalogandthepipelineaperturephotometrycatalog,andnoperceivablespatialvariations.Therefore,webelievethezeropointcorrectiontotheISPI 50

PAGE 51

Miller 2007 ),rstproducingJH,JKandHKsetsandthenmergingtheJHandHKentriesiftheysharedacommonHcatalognumber.TheoverlapregionsbetweentheUKIDSSpointingweretoonarrowtoeffectivelymeasureourphotometricerrorsbythemethodweusedforISPI.Thereforewecharacterizedbothourttingerrorandcompletenessbyarticialstartests.TheprocedurewasthesameasfortheISPIarticialstartest;e.g.weusedtheDAOPHOTfunctionaddstartoplaceauniformlyspacedgridofarticialstarsofthederivedPSFshapeintooneofourimages.Thepositionofthegridandthemagnitudeofthestarsthereinwereassignedrandomlyforasetof10images.WeperformedtheDAOPHOTreductionoftheseimageframeswitharticialstarswithexactlythesamedetectionthresholds,skyannulusandttingradiiasforouroriginalreductionoftheUKIDSSdata.InthesamemannerasfortheISPIdatawecountedanarticialstarasrecoveredifitwasfoundwithin0.500ofitsinjectedpositionandweexcludedanyarticialstarsthatfellwithin0.500ofarealstarintheISPIimages.Toobtainthecharacteristicerroratagivenmagnitudewegroupedthearticialstarsbyinputmagnitudeandmeasuredthe3-clippedstandarddeviationofthemoutminmagnitudewithin0.5magintervals.Weadoptedthisvalueasour1errorforagivenmagnitude.Sincetheinputmagnitudeswereexactlyknownwedidnotscalethe1errorbyp 2-3 51

PAGE 52

Irwinetal. ( 2004 ).TheVISTAimageswereastrometricallycalibratedtothe2MASScatalog.THeVISTAearlysciencedataoftheGCbecameavailableatalatestageinthisdissertationsoourcharacterizationofthedataislimited.TheJ,HandKsimagescompletelysubtendedtheISPIeld.WeextractedthesubsetsoftheVISTAimagesthatcontainedtheISPIeld,inordertosaveonprocessingtime.WeperformedDAOPHOTPSF-ttingonthereducedimages,usingasingleiterationofthesourcedetectionprocedure.Themeanpixelsizeof000.34wasclosetothe000.3valuefromISPI;thereforeweadoptedthesamettingradiiandskyannulusparametersasforourISPIreductiontomaintainconsistencyinthetting.Atthisstagewehad3pointsourcecatalogsintheJ,HandKsbands,thathadnotbeenzeropointcorrected.ThereasonwedidnotproceedfurtherwasduetoproblemsobtainingsufcientlyaccurateastrometryfromtheVISTAWCSheader.Theresultofthisproblemwasthatwewouldhavetouseamatchingradiusof1.500tocross-matchtheVISTAsourcesbetweenlterscausinganunmanageablenumberofincorrectlymatchedsources.WeexpectthereisaneasysolutionbecauseoftheclaimbytheCASUwebsitethat0.0005calibrationhadbeenattainedwithrespectto2MASS.Forthepurposesofthiswork,however,wechosetoapplyzeropointstotheVISTAdatawhenwecross-matchedforsmallsub-eldsoftheISPIdatainChapter4.WemergedtheJHKsltercatalogsonlyforparticularsourcesofinterest,wherewecouldvisuallyinspecttheresultstopreventpotentialmismatches. 52

PAGE 53

BFigure2-1. (A)ThreecolorJHKscompositeimageoftheInfraredSidePortImager(ISPI)Galacticcenterdata.Eachsideis170inlength.Northisupandeastistotheleft.(B)ThepointingpatternoftheISPIobservations,overlaidontheKsimage.Themagentaboxrepresentsthe10.5010.50eldofviewofasinglepointingwithISPI.Thefullframeisa170170regionandconsistsof5suchpointings. Figure2-2. ThedifferenceinKsmagnitudeversusmeanKsinanoverlappingregionoftheISPIGCeld. 53

PAGE 54

CharacteristicsofdifferentimagingdataoftheGalacticCenter(GC). TelescopeFiltersPSFFWHMKmaglimitAreaReferenceInstrument(00)(50%comp.) 2MASSJHKs2.50011.5FulleldSkrutskieetal.2006VLT/ISAACJHKs0.60016.05050Bandyopadhyayetal.2005MAGELLAN/PANICJHKs0.50015.4100100Laycocketal.2005IRSF/SiriusJHKs1.10014.0FullFieldMauerhanetal.2009ISPI/CTIO4mJHKs0.90014.5170170DeWittetal.2010UKIDSSGPSJHK0.850014.7FullFieldLucasetal.2008VISTAJHKs0.80014.5FullFieldEmersonetal.2006Hubble/NICMOSPa0.200H=17.5FullFieldWangetal.2010VLT/NAOSHKsL00.010017.040004000Schodeletal.2010

PAGE 55

Magnitudeversuserror,calculatedfromthestandarddeviationofKsinFigure 2-2 Figure2-4. 10020HbandimageofanareawithintheoverlappingregionbetweentwooftheISPIpointings. Figure2-5. ImageofthesameregionshowninFigure 2-4 withthevalueofeachpointrepresentingthemedianoffsetinmagnitudesbetweenthetwoISPIpointings,forthe150starsnearesttoeachpoint.Thestretchoftheimageis-0.1to-0.001mag. Table2-2. ISPIpointsourcecatalogcharacteristics. FilterNumberofstars5-Mag.limit,5-Mag.limit,50%CompletenessPoissonerrorttingerrorlimit 55

PAGE 56

100100HbandISPIimageofamatchedareabetweentheISPIandtheUnitedKingdomInfraredDeepSkySurvey(UKIDSS)datasets,toillustratetheregionoftheISPIdatacorrectedbyFigure 2-7 Table2-3. UnitedKingdomInfraredDeepSkySurvey(UKIDSS)pointsourcecatalogcharacteristics. FilterNumberofstars5-Mag.limit,5-Mag.limit,50%CompletenessPoissonerrorttingerrorlimit 56

PAGE 57

100100imageofthelocalmedianoffsetbetweenthemagnitudeofstarsinISPIandinUKIDSSwithinFigure 2-6 Table2-4. VisibleandInfraredSurveyTelescopeforAstronomy(VISTA)pointsourcecatalogcharacteristics. FilterNumberofstars5-Mag.limit,Poissonerror 57

PAGE 58

ThedifferenceinKsmagnitudeversusmeanKsinanoverlappingregionoftheISPIGCeld. Figure2-9. Magnitudeversuserror,calculatedfromthestandarddeviationofKsinFigure 2-8 58

PAGE 59

BFigure2-10. ISPIphotometricerrorandcompleteness:theerrorperlterderivedusingmultiplemeasurementsofthesamestars.ThecompletenessisthefractionofarticialstarsrecoveredatagivenJ,HorKsmagnitude. 59

PAGE 60

Bandyopadhyayetal. ( 2005 ), Laycocketal. ( 2005 ), Arendtetal. ( 2008 ), Goslingetal. ( 2009 )and Mauerhanetal. ( 2009 ). Bandyopadhyayetal. ( 2005 )usedISAAContheVLTtoimage26eldscontaining77oftheX-raypointsourcesdiscoveredby Wangetal. ( 2002 ).TheydetectasmallexcessincounterpartsintheJandHbands,butndnosignicantexcessinmatchesintheKband. Laycocketal. ( 2005 )observethecentral100100GCareawiththePANICnear-infraredcameraonMagellantoaconfusionlimitofKs=15.4mag.Cross-correlatingtheirIRcatalogtotheX-raycatalogof Munoetal. ( 2003b ),theyndastrongIR/X-raymatchingsignicanceforthesoftX-raysources,butlittlesignicanceforthehardX-raysources.Theycalculatethatnomorethan10%ofthehardX-raypointsourcescanhaveapparentKsmagnitudesbrighterthan15mag.AssumingthesehardX-raysources 60

PAGE 61

Arendtetal. ( 2008 )performsimilarmatchingsimulationsusingacatalogof20000SpitzerIRACpointsourcesdetectedat=3.6,4.5,5.8and8.0mlyingwithinthe170170eldofthe Munoetal. ( 2003b )Chandraobservations.TheyfoundthatthesoftX-raysourcesarecorrelatedtotheIRACcatalog,buttheydetectnosuchcorrelationtothehardX-raysources.However,theynotethatsourcecrowdinglimitstheircatalogtoGalacticcentersourceswith[3.5]<12.4mag,andactualcounterpartsmaybemuchfainterthanthislimit.Sincethesestudieswerepublished,theGalacticCenterX-raypointsourcecataloghasbeenupdatedtoincludeatotalof2Msofaccumulatedobservationsofthe20.8areaaroundSgrA( Munoetal. 2009 ). Munoetal. ( 2009 )detect9017X-raypointsources,thepositionsofmanyofwhichhavebeenrenedfromtheearliercatalogs.Themajorityofthesourcesinthecataloghaveastrometricerrorslessthan0.700. Mauerhanetal. ( 2009 )utilizethesedatatogetherwithnear-infrareddatafromtheSIRIUScameraonthe1.4mtelescopeatSutherlandObservatoryinSouthAfrica,andpresentacatalogof5184potentialnear-infraredcandidatecounterpartstotheX-raysources.Theyndthat,statistically,394oftheNIRmatchestothehardX-raysourcesarerealcounterparts.Thisamountsto5.8%ofthe6760hardX-raysources,consistentwiththendingsof Laycocketal. ( 2005 )whichlimitthispercentagetonomorethan10%.Inthischapter,wecross-correlateourISPINIRcatalogtotheChandraX-raycatalogof Munoetal. ( 2009 ),creatingasetof2137NIRcandidatecounterpartstoX-raysources.WestatisticallyidentifythelocationsofthetruecounterpartsonaNIRcolor-magnitudediagram.Wealsoisolatethecharacteristicsthatmakelikelyauthenticcounterpartseasiertondamongstthelargenumberofspuriousmatches.These 61

PAGE 62

Munoetal. ( 2009 )catalogthatliewithinourISPIeld(seex2.2.1).WeexcludeanyX-raysourceswithX>2.000,whereXisthe95%positionaluncertaintyfortheX-raysources,inarcseconds.The71sourceswithlargererrorshaveamediannumberofpossiblematchesof5each,whichwouldprecludeusefulfollow-up.Fortherestofthisworkweonlyconsidertheremaining4268X-raysources.Wecross-correlatethese4268X-raysourceswiththeISPIcatalogusingaX+0.200matchingradius,where0.200representstheIRcatalogastrometricerror.Wedonotusethequadraturesumoftheseerrors,p 3-1 ).OurcatalogofISPI/X-raymatchesisavailableonlineasasearchabledatabasecalledChandraGalactica.Thewebsiteurlishttp://galcent.astro.u.edu.InFigure 3-1 weshowmagnitudeandcolorhistogramsforalltheISPIsourcesandthecandidatematchestoX-raysources.ThehistogramfortheX-raycandidatecounterpartIRsourcesisoverplottedinredandnormalizedto 62

PAGE 63

3-2 and 3-3 wedisplayaH-Ks/KsCMDandaJ-H/H-Kscolor-colordiagramofthematchedNIRsourcesandtheentireISPIcatalog.TheCMDclearlyshows2populationsofsources:sourceswithlowreddeninginanearlyverticallineatH-Ks=0.2magandsourceswithhighreddeningcenteredatH-Ks=2.0mag.Thesourceswithlowreddeningarenearbyeldsources,whilethesourceswithhighreddeningshouldbeneartheGC.Thecolor-colordiagramshowsthatthesourcesaredispersedalongthereddeningvector.Byeye,theX-raysourcematchesmarkedincolorontheseguresshowlittledifferencefromtheoverallISPIcatalog.ThiswouldbeexpectedeitherifthesourcesactuallyarenotdistinctfromtheoverallNIRsourcepopulationorifmostofthematchesarespurious.Toevaluatethequantityandcharacteristicsoftheprobabletruematches,weneedtohaveanideaofthenumberofmatchesthatwouldhappensimplybychance. 63

PAGE 64

3-4 weshowthehistogramofpositionerrorsforthe4268X-raysourceswithpositionaluncertaintiesunder200andforthe1864X-raysourceswithNIRmatches.Thefractionofsourceswithamatch(shownintherightpanelofFigure 3-4 )isabout27%fortheX-raysourceswitherrorsof0.300,andreachesnearly90%around1.100.Thisimpliesthatthesourceswithlargerpositionalerrorshavemorespuriousmatches,aswouldbeexpected.2.HardnessratiosSoftX-raysarenotexpectedtopenetratethegasanddustcolumntotheGalacticCenter.TheX-rayhardnessratioprovidesawaytodividetheX-raysourcesintothosethatareheavilyabsorbedandthereforeatorbeyondtheGalacticCenter,orsoftandthereforelocatedwellinfrontoftheGC.Weadoptthesamehardnessratiocriteriaasin Munoetal. ( 2009 ).TheydeneHR0=(hs) 64

PAGE 65

2-1 .Thelanescorrespondtolinesofsightwithhigherextinctionandareseenasregionswithmorereddenedcolorsandfewerstarcounts.WeattempttoquantitativelylocatethesehigherextinctionareasbycomputingtheaveragestellarsurfacedensityinthevicinityoftheX-raysources.Wecalculatethestellarsurfacedensityusingstarswith101.0ISPI,denseandregionswithvalues1.0ISPI,sparse.Theintentionisthatsparseareasindicateregionsofhighextinctionanddenseareasindicatewindowsoflowerextinction.5.IRcatalogpropertiesThecharacteristicsoftheinfraredcatalogincludetheJ,HandKsmagnitudesandtheJH,JKs,andHKscolors.TwomajorpopulationscanbeseenintheHKsdistributioninFigure 3-5 .Thereisalessreddenedpopulation,centeredatHKs=0.2magandamorereddenedpopulationcenteredatHKs=2.0mag.WeseparatethetwoatHKs=1.0mag,andcallsourceswithHKs<1.0mag,unreddenedandsourceswithHKs>1.0mag,reddened.WederivedcorrespondingcriteriaforJHandJKsusingtheGalacticbulgeextinctionratiosfrom Nishiyamaetal. ( 2008 ).ThesewereJH<1.76mag,JKs<2.76magforunreddenedsourcesandJH>1.76mag,JKs>2.76magforreddenedsources. 65

PAGE 66

3-2 .FromthedenitionofNobs: 66

PAGE 67

67

PAGE 68

68

PAGE 69

(ANran,1Nran,2)2)22Nran,1+(Nran,1(ANobs,1Nobs,2) (ANran,1Nran,2)2)22Nran,2(3) (ANran,1Nran,2)2)22Nran,1+(Nran,2(ANobs,1Nobs,2) (ANran,1Nran,2)2)22Nran,2(3)ThevariabledenitionsforthisderivationarecollectedinTable 3-3 .InthenextsectionwebrieyignorethebiasesassociatedwiththematchingsimulationsinordertogetanillustrativeviewofthecontentofthematchedX-ray/NIRcatalog.Thenweapply 69

PAGE 70

70

PAGE 71

3-4 weshowtheRrealvaluesanduncertaintiesforcombinationsoftheX-raysourceproperties.22%ofsoftsourceshavedetectedmatchesintheISPIcatalogcomparedtoonly1.6%ofthehardsources.Bothsoftandhardsourcesshowabiastowarddetectionofbrightsourcesoverfaintsources,withabrightsourcebeingmorethantwiceaslikelytohaveadetectedcounterpart.Thereisaslight(6%3%)biastowarddetectingsoftsourcesinsparseregionsoverdenseregions.Thiseffectmaybecausedbythecrowdinglimitsbeingmoresevereindenseregions.Thereisnosignicantdifferencebetweendenseandsparseregionsseeninthehardsources.However,thelownumberstatisticsforhardsourceswouldmaskaneffectofsimilarmagnitudetothesoftsources.ThemainpointtodrawfromthisisthathardsourceshaveaparticularlylowrateofIRdetectioninthedata,atleast4timeslowerthanforsoftsources.WeconcludethatmostofthesourcesintheX-raycatalogaretoofaintintheNIRtobedetectedbytheISPIobservations,whichisconsistentwiththendingthat57%ofX-raysourcesdonothaveacandidatematchatall.WediscussthecharacteristicsofthesourceswithoutNIRmatchesinthenextsection.

PAGE 72

2-1 ,whichareJ=18.4mag,H=16.4magandKs=14.5mag.Theunmatchedsoftsourcesarelikelytobecoronallyactivelatetypestars( Munoetal. 2003b ).Coronalactivityatthe>1027ergs=slevelisseenatallspectraltypes 72

PAGE 73

Feigelsonetal. 2004 ).Foranillustrativecase,weconsiderthataJ=9magM0VstarwithnoextinctionwouldbedetectedinourISPIdataouttoadistanceof800pc. Munoetal. ( 2003b )determinethatthesoftsourcesintheChandraGCdatacanresideasfaroutas4kpc,whichleavesasubstantialvolumeforcoronallyactivelatetypedwarfstarstoresideandyetbeundetectableinourISPIimaging.TheunmatchedhardX-raysourceshaveanumberofdiversecandidates:AGNs,isolatedpulsarsandmagnetars,magneticCVsandLMXBs. Munoetal. ( 2003b )calculatethatthe170170centralGCeldshouldhavebetween20and100AGNcontributingtothehardsourcecounts.AGNthatareseenthroughtheGCwillbeveryfaint. Bandyopadhyayetal. ( 2005 )notethattherearenoAGNintheHubbleDeepFieldNorthsurveybrighterthanK=17mag.AfterAK>2.5magextinctionisaddedon,thismagnitudeiswellbeyondoursensitivity,whichmeansthatthereshouldbenodetectedcounterpartsforthe102AGNintheX-raycatalog.Isolatedneutronstarscanbebrightatveryyoungages,butexceptfortheCrabPulsar,thebrightestKbanddetectedneutronstaristhemagnetar1E1547.0-54.08withK=18.2mag,atadistanceof9kpc( Mignani 2009 ).Weconcludethatforallbutthemostexceptionalsystems,isolatedpulsarsintheGCshouldbeundetectableinourISPIdata.LMXBsandmagneticCVssharethesametypesofdonorstars,usuallylatetypeKdwarfs( Jonker&Nelemans 2004 ; Knigge 2006 ).Wheninquiescence,thesesystemscanbedominatedbythelightofthesecondarystar.InthecaseofaK0dwarfstar,asystemintheGalacticCenterwouldbeK21mag,assumingAK=2.5mag( Baganoffetal. 2003 )andaGCdistanceof8kpc( Gillessenetal. 2009 ; Cox 2000 ; Carroll&Ostlie 1996 ).Thus,themajorityofcanonicalLMXBsandCVswouldbeundetectedatGCdistancesinourISPIdata. 73

PAGE 74

3-5 .Thereare387singleIRmatchestoX-raysoftsources.AnalysisoftherandomizedX-raycatalogmatchingsuggeststhat23012areprobablerealmatches,forarateof603%.Weseethatalmostall(96%)thesoftX-raysourcesmatchedtounreddenedNIRsourceswithaJdetectionarereal,whileatrivialnumberofmatchestoreddenedX-raysourcesarereal.Thelackoftruesoft/reddenedsourcesisnotsurprising,becausetheX-rayattenuationimpliedbythereddenedIRcolorsshouldcauseanintrinsicallysoftsourcetobedetectedasahardsource,orbecompletelyextinguished. 3-6 ).Suchalowrateoftruematchesisaproblemforobservingcampaignsthatseektofollowupthesesources.Wethereforetrytondthesourceparametersthatmaximizethereturnrate.Inparticularwewishtolocateprobablerealmatcheswhicharebothhardandreddened,whichwouldsuggestthattheyexistatorbeyondtheGalacticCenterdistance.WepresentthestatisticsforanumberofpropertiesandcombinationsofpropertiesinTable 3-6 .ThemostpromisingpropertyseemstobewhetherornotacandidatecounterparthasaJ-banddetection.InTable 3-6 ,weseethatthereare9635probabletrue 74

PAGE 75

3-6 ).TheX-rayfaintsourceswithJbanddetectionshavealowerrateofexpectedrealmatches,12%.WesearchedforotherparametersthatcanraisethepercentageoftruesourceswithintheX-rayfaintcategory,andwendthatifthesourcesarematchedto13.0Ks<14.0magIRsources(inadditiontotheirbeingreddenedanddetectedintheJ-band),andtheX-rayerrorcirclesarerestrictedtoX1.000,thepercentageprobabilityofbeingrealrisesto40%(seelastrowofTable 3-6 ).The13.0Ks<14.0magcriterionissuggestedbythecolor-magnitudeanalysisinx3.3.Incombination,thesetwocategories,showninthebottomtwolinesofTable 3-6 ,have98X-ray/NIRcandidatematches,with447probablerealmatches.InFigure 3-5 weplottheJ-H,J-KsandH-Kshistogramsforallthecandidatematchesfollowedbythecolorhistogramsforthecomponentofthematchesthatareexpectedtobereal.Thenumbersofsoftsourcematchesareshowninblueandthehardsourcematchesareshowninred.Itisapparentthattheratioofrealtocandidatesourcesismuchhigherforsoftsourcesthanhardsources.Also,softsourceswith 75

PAGE 76

3-5 and 3-6 toestimatethetotalnumberofdetectedtrueNIRcounterpartswithinthemultiplymatchedsources. 685=0.420.02.IntheISPI170170area, Mauerhanetal. ( 2009 )nd324candidatematchestosoftX-raysources;200oftheseareunreddened.Theydonotreportontheprobablenumberoftruematcheswithinthiseld,buttheyndthatwithintheentire20.8eld,890outof1007bluematchesarereal.Iftheratioof0.88holdsforthe170170subeld,thentheydiscovered177unreddenedrealcounterpartstosoftX-raysourcesinthe170170ISPIGCeld.TheyreportessentiallyzerorealreddenedsoftsourcecounterpartsmakingtheirtotaldetectedfractionofsoftX-raycounterparts, 76

PAGE 77

685=0.26.Ourvalueof0.420.02islarger,whichweexpectbecauseourcataloghastwiceasmanydetectedsourcesastheSIRIUScatalogwithinthisareaduetoourNIRdata'sgreaterdepthandresolution. 3-6 .Sincethesesourcesaremultiplymatched,itispossibleforanX-raysourcetohaveamatchformorethanoneofcriteria1-4.ToreducethecomplexitywenotethattherealmatchprobabilityforX-raysourceswithJbanddetectedmatchesismuchhigherthanforX-raysourceswithoutJbanddetectedmatches.Therefore,weassumethatifoneofthesemultiplymatchedhardsourceshasaJdetectedcounterpart,thepossibilitythatanyadditionalmatcheswithoutaJdetectionistherealcounterpartisnegligible.Thereare65multiplymatchedhardX-raysourceswithatleastoneJdetectedmatch.Four(4)havebothunreddenedandreddenedJbanddetectedmatches,28havereddenedJbanddetectedmatchesand33haveunreddenedJbanddetectedmatches.Forthe4sourceswithbothreddenedandunreddenedmatches,weestimate0.40.1realreddenedmatchesand0.90.2realunreddenedmatches.The28X-raysources 77

PAGE 78

3583=0.040.01hardX-raysourceswithdetectedcounterparts. Mauerhanetal. ( 2009 )foundarateof0.0580.015withdatathatisslightlyshallowerindepthbutthatspansalargerareaaroundtheGC;nevertheless,ourfractionisstillconsistentwith Mauerhanetal. ( 2009 )withintheerrors.Forafairercomparison,weusetheirstatisticsonthe80radiusregionaroundSgrA(=266.41726,=29.00798).Theycalculatethenumberofrealhardreddenedcounterpartswithinthisregiontobe46.223.1sourcesoutof617candidatematches.Ourtotalforthisregionis7333sourcesoutof1018X-raysourceswithcandidatematches.Thisfractionisconsistentwith Mauerhanetal. ( 2009 )andourgreatertotalislikelyduetothegreaterdepthandresolutionoftheISPIimagingwithrespecttotheSIRIUSdata. 78

PAGE 79

3-6 and 3-7 weshowtheCMDoftheentireISPIcatalogoverplottedbyboxeswhicharecolor-codedtorepresentthenumberofprobabletruecounterpartswithinaCMDbin.HereweonlyincludeX-raysourceswithasingleIRmatch.Theboxcolorsareblack,yellow,greenandblue,andcorrespondto0,15,59,and>9realmatches,respectively.Binswithlessthana2signicance(Nreal,1 3-6 weseethattherealsoftsourcesaremostlyconnedtoHKs<1mag,and8.01mag,and9.01magaremoremysterioussincethehighreddeningimpliedbytheIRcolorshouldalsomaketheX-raycomponentappearhard.However, Mauerhanetal. ( 2009 )donotethatforsourcesnearthehard/softthreshold,theuncertaintycouldcauseahardsourcetobelabeledasasoftsource,andviceversa.AnotherpossibilityisthatthesesystemsmayhaveinternalabsorptionthataffectstheIRcomponentmorethantheX-raycomponent.Finally,for 79

PAGE 80

Mauerhanetal. 2010b ).Inx3.1.5wefoundthat96%ofthe242singlymatchedsoftX-raysourcecounterpartswithunreddenedcolorsandJdetectionsarelikelytobeauthentic.WeplottheirpositionsovertheISPICMDintherightpanelofFigure 3-6 .However,becauseweusedunreddenedcolorsandaJbanddetectionasacriteriainordertoenhancetheprobability,anyrealreddenedmatchestosoftX-raysourceswillnotrepresentedinthisgure. 3-7 ).AlloftheseCMDbinsresideinthe1
PAGE 81

Munoetal. ( 2009 )arguethatthemajorityofhardX-raysourcesmustbefartherthan4kpcfromtheSun,becauseoftheextinctioncolumnrequiredtomakesoftsourcesappearashard.WeadoptthisvalueastheminimumdistanceforhardX-raysources.Forthemaximumdistance,weuseavalueof8kpc,areasonablevalueforthedistancetotheGalacticCenter.ThereisthepossibilityofobservingbrightIRcounterpartsonthefarsideofthebulge(withdistance>8kpc).However,thestrongestcandidateCMDbinslieinthe1<(HKs)<2magrange,whichismostlytothebluesideofthemodalvalueofHKs=1.8magfortheISPIcatalog,implyingthattheymaylieonthenearsideoftheGC.Theintrinsiccolorsformainsequence,giantandsupergiantstarsrangefrom0.06<(HK)0<0.36mag,and0.06<(HK)0<0.1magforallspectraltypesexceptforM( Cox 2000 ).Weusethisdispersioninintrinsiccolorandthe1<(HKs)<2maglimitsofthecolorbintocalculatetherangeinabsoluteKsextinction,AKs.UsingextinctionratiosfortheGalacticCenterofAV:AJ:AH:AKs=1.00:0.188:0.108:0.062( Nishiyamaetal. 2008 )wendthatsourceswith1<(HKs)<2magareexperiencing0.86
PAGE 82

Mauerhanetal. ( 2010b )observedKbandspectraoftheNIRcounterpartstoCXO174532.7-285617,CXO174536.1-285638,CXO174555.3-285126,CXO174617.0-285131whichallliewithinthe10
PAGE 83

83

PAGE 84

Eikenberry 2008 ). 84

PAGE 85

NumberofX-raysourceswith0,1,2,or>3astrometricmatchesintheISPIJHKscatalog. NumberofMatchesOccurrencesPercentage 0242556.8%1161037.7%21924.5%>3411.0% MagnitudeandcolorhistogramsandcumulativedistributionsforallsourcesintheISPIcatalog(inblack)andforjusttheIRsourcematchestoX-raysources(inred).ThehistogramsfortheX-raymatchedIRsourcesarenormalizedtothehistogramsoftheentireIRcatalog. 85

PAGE 86

Colormagnitudediagram(CMD)oftheISPIcatalog(inblack)withcandidatecounterpartstoChandraX-raysourcesmarkedinmagenta. Table3-2. VariabledenitionsforcalculatingnumbersofrealcounterpartstoX-raysources,withunspeciedIRproperties

PAGE 87

Color-colordiagramoftheISPIcatalog(inblack)withcandidatecounterpartstoChandraX-raysourcesmarkedingreen. BFigure3-4. (left)Distributionsofpositionalerrorsizesofall4268ChandraX-raysourcesintheISPIGCeld(solidline),andforthesubsetoftheseX-raysourceswithoneormorenearinfrared(NIR)matches(dashedline).(right)FractionofX-raysourceswithoneormoreNIRastrometricmatchesversustheX-raypositionalerrorsize. 87

PAGE 88

HistogramsoftheNIRcolorsofmatchestosoft(bluehistogram)andhard(redhistogram)X-raysources.Frames(a),(b)and(c)showtheJH,JKsandHKsdistributionsforallcandidatecounterpartsandframes(d),(e)and(f)showthesamecolordistributionsfortheestimatednumbersofrealcounterparts(seeSectionsx3.1.5andx3.1.6).ThedottedlinesindicateourcutbetweenunreddenedandreddenedISPIsources. Table3-3. VariabledenitionsforcalculatingnumbersofrealcounterpartstoX-raysourceswithagivenIRcounterpartproperty.

PAGE 89

RelativematchingfractionforX-raysources. CriteriafrealfSignicance AllX-raysources0.0520.0077.7Softsources0.220.0213.6Hardsources0.0160.0072.2Soft,faintsources0.150.028.0Soft,brightsources0.420.0313.3Softsources,denseregions0.180.028.3Softsources,sparseregions0.230.0210.4Hard,faintsources0.0010.0080.2Hard,brightsources0.050.013.8Hardsources,denseregions0.0120.0091.4Hardsources,sparseregions0.020.011.7 MatchingstatisticsforsoftX-raysourcesmatchedtoasingleIRsource CriteriaNobsNranNrealR=Nreal 89

PAGE 90

MatchingstatisticsforhardX-raysourcesmatchedtoasingleIRsource CriteriaNobsNranNrealR=Nreal Table3-7. RangesofMKspossibleforrealsourcesincolormagnitudediagram(CMD)binsfromFigure 3-7 .Thecolorrangeis1.0<(HKs)<2.0magforallthreebins. ApparentmagnituderangeMKsford=4kpcMKsford=8kpc

PAGE 91

BFigure3-6. (left)HKs,KscolormagnitudediagramofallISPIsources,inred,withthepositionsofprobabletruecounterpartstoX-raysoftsourcesshownbythecoloredboxes.Thecolorcodingisyellow:1-5sources,green:5-9sources,andblue:9ormoresources.Blackboxescontaincountswithlowerthan2signicanceandareconsistentwith0realcounterparts.Theboxesareundersizedforclarity.(right)HKs,KscolormagnitudediagramofallISPIsources,indarkred.Overplottedinblueareasetof242softX-raymatcheswithJbanddetectionsandlowreddening.These242IRsourcesshouldhavea96%likelihoodofbeingthetruecounterpartstotheirmatchedX-raysources.Seex3.1.5fordetails. 91

PAGE 92

BFigure3-7. (left)HKs,Kscolormagnitudediagram(CMD)ofallISPIsources,inred,withthepositionsofprobabletruecounterpartstohardX-raysourcesshownbythecoloredboxes.Thecolorcodingisyellow:1-5sources,green:5-9sources,andblue:9ormoresources.Blackboxescontaincountswithlowerthan2signicanceandareconsistentwith0realcounterparts.Theboxesareundersizedforclarity.(right)HKs,KscolormagnitudediagramofallISPIsources,indarkred.Overplottedareasetof98hardX-raymatcheswithreddenedcolors,Jbanddetections,smallpositionalerrors(X1.000)andareeitherX-raybrightorX-rayfaintwith13Ks<14mag.These98IRsourcesshouldhavea45%likelihoodofbeingtruecounterpartstotheirmatchedX-raysources.Seex3.1.6fordetails. 92

PAGE 93

93

PAGE 94

Leahy&Ananth 1992 ). Leahy&Ananth ( 1992 )monitoredCygX-1andfoundJandKvariabilitywithanamplitudeof0.3magover6dayintervalswhichtheyattributetoaccretioncyclesrelatedtotheorbitalperiod.AnotherclassofsupergiantHMXBisthesuperfastX-raytransients(SFXTs)discoveredbytheINTEGRALsatellite. Pellizzaetal. ( 2006 )foundinterdayvariationswith0.2magamplitudeintheKsbandforIGRJ17544+2619,anSFXTwithanO9Ibsecondary.BeHMXBscanvaryduetochangingaccretionratesontotheneutronstarorbydisruptionofthecircumstellardiskaroundtheBesecondaryfollowingacloseorbitalpassageoftheneutronstarprimary( Reig 2011 ).TheBeHMXB4U0115+634wasobservedby Negueruelaetal. ( 2001 )tovaryintheIbandbetween12.46magand13.46magwhenitwasmonitoredfrom1993to1998,duetobothaccretionontotheNSandBediskdisruptions. 94

PAGE 95

Mauerhanetal. 2010b ).NearbyCWBshavebeenobservedtovaryintheNIR.Forexample, Taranova&Shenavrin ( 2011 )monitoredWR140,aWCclassWolf-RayetstarinorbitwithaO4-5Vstarfor9yearsintheJHKbandsandfoundvariabilityamplitudesof0.5-1.5magover50-daytimescales,whichtheyattributetodustejectioncycles. Kenyon 1986 ).Eightypercentofthe Belczynskietal. ( 2000 )catalogofsymbioticstarscontainanaccretingwhitedwarfandanormalredgiantbranchMstar.Theremaining20%ofthecatalogconsistsofWDswithasymptoticMgiantstarcompanions,includingMirasandsemi-regularvariables.PhotometricvariabilitycanariseeitherfromwindaccretionprocessesorfrompulsationsintheMIIIstar.TypicallyWD/RGBsymbioticsshowepisodesofaringwithVamplitudesof0.5magduringthequiescentphaseinadditiontoviolentoutburstsof3magnitudesevery10-15years( Kenyon 1986 ).ForWD/AGBsymbiotics,theNIRvariabilitycanalsobeinuencedbylargeamplitudestellarpulsations.GX1+4isasymbioticstarwithanormalMgiantandawind-fedneutronstarcompanion.ItiscommonlyclassiedasanLMXBinadditiontoitsdesignationasasymbioticstar. Chakrabarty&Roche ( 1997 )showthattheKbandmagnitudeofGX1+4changesby0.2magover4yeartimescales,duetovaryingaccretionrates. Ritter&Kolb 2003 ).Forblackholeandneutronstartransients 95

PAGE 96

Tanaka&Shibazaki 1996 ).Inprinciplethesesourcescouldbeseeninourdata,ifcaughtduringanoutburst.ThreecasesofLMXBsthatwouldbepersistentlyvisibleinourdataareGX1+4,GRS1915+105andV404Cygni.GX1+4,asdiscussedinx1.1.3,has0.2Kbandvariationsoverspansofyears.GRS1915+105isablackholecandidatewithaK/MIIIdonorstarthathasbeenpersistentlyactivesinceabrightX-rayoutburstin1992( Greineretal. 2001 ).VeryfewX-raysourcesinthe Munoetal. ( 2009 )catalogareseentoreacheventheminimumX-rayluminosityofGRS1915+105andassuchthelargeKbanductuationsof>1.5mag( Fenderetal. 1997 ; Neiletal. 2007 )maynotbeveryrepresentativeforthelowX-rayluminositypopulationintheGC.Sinceitsoutburstin1992,theleastX-rayluminousstatehasbeentheso-calledplateaustate. Eikenberryetal. ( 2008 )foundmillisecondtimescaleinfraredickeringduringtheplateaustateforGRS1915+105withKamplitudesof0.2-0.3mag.V404CygniisablackholecandidateLMXBwithaKIIIsecondarywhichunderwentitslastoutburstin1989.Ithasbeeninaquiescentstateeversince,withatypicalLXof1033ergss1.Photometricmonitoringindicatesthateveninthisquiescentstate,V404Cygnivariesatthe0.24maglevelinKfromellipsoidalmodulationsoftheKIIIsecondaryduetothecloselyorbitingblackhole( Zuritaetal. 2004 ). 96

PAGE 97

Matsunagaetal. 2009 ).Cepheidvariablesarethehorizontalbranchstageofsomeintermediatemassstars.TheirspectraaretypicallyclassiedasF-GI/IIstars.Theparticularsub-typeofCepheidsknowntoexistintheGalacticBulgearemetalpoorCepheids,designatedTypeIICepheids.TypeIICepheidshaveperiodsbetween1-10daysandKsamplitudesofabout1mag( Groenewegenetal. 2008 ). Matsunagaetal. ( 2009 )monitoredtheinner200300oftheGalaxywithsimultaneousJHKsphotometrynding1364longperiodvariableswith100-350dayperiodsandKamplitudesof0.2-2.5mag.Theyalsond50shorterperiodvariablesbetween0.1and60dayswhichtheyattributetoeclipsingbinariesandtypeIICepheids.Atthetimeofthiswriting,thecoordinatesoftheshorterperiodvariablestarshavenotbeenpublished.However,inx4.2.4weareabletousethepublishedcoordinatesoftheLPVdetectionsfrom Matsunagaetal. ( 2009 )totesttheeffectivenessofour2epochphotometryforidentifyingvariablestars. 97

PAGE 98

Miller 2007 ),withalargematchradiusof100tocross-matchthepositionsofsourcesintheISPIandUKIDSScatalogs.Inthesecondstage,weevaluatedeachcross-matchedpairagainstthetypicalmedianoffsetsofneighboringcross-matchedpairsofstars.Weusedthenearest50pairsofstarsforevaluatingthedistributionoflocaloffsets.Thisgroupof50starstypicallysubtended4500and1500inradiusforJandH/Kbandsrespectively.ThedistributionsofRAandDECoffsetsforthe50neighboringpairsofstarsformedwelldenedGaussiandistributionswithawidthof0.01-0.0500.Onthisbasis,werejectedcross-matchedpairsofsourceswithRAorDECoffsetsthatdifferedby0.100fromthelocalmedianvalues.TheresultingmatchedISPI/UKIDSScatalogcontainsabout75%oftheISPIcatalogsourcesineachband.InFigures 4-1 through 4-3 weshowmagnitudehistogramsfortheISPI,UKIDSSandcross-matchedISPI/UKIDSScatalogsforsourceswithintheboundsofthe170170ISPIobservations.WenotethattheISPIKsbandandUKIDSSKbandarenotidentical(seex4.2.2). 98

PAGE 99

4-4 through 4-6 weshowthemagnitudeversusthemagnitudedifferenceforcrossmatchedsourcesinUKIDSSandISPI.Themagnitudedifference,mUKIDSSmISPIiscenteredaroundzeroforalllters,exceptatthebrightends(J<10.5mag,H<11.5magandK<10.7mag)andthefaintends(J>18mag,H>16andK>14).Onthebrightendthisdivergenceiscausedbynon-linearityandsaturationintheUKIDSSphotometry.Onthefaintend,themUKIDSSmISPIvaluestendtoshowsystematicallybrighterISPImagnitudes.ThisisduetoEddingtonbias,wheresourcedetectionalgorithmswhenoperatingnearthedetectionlimitsaremorelikelytodetectsourceswithbrightuctuationsthanfaintuctuations.WeavoidthesetworegionsbymakingmagnitudecutsontheISPImagnitudevaluesof12
PAGE 100

4-4 4-5 and 4-6 showthemagnitudeversusmagnitudedifferencefortheJHK=KsltersasderivedbyISPIandUKIDSS.Wesplitthecross-matchedsourcesinto0.5magintervalsandconstructedahistogramofabsolutevaluesofthemagnitudedifferenceofsourceswithintheseintervals.WetaGaussianfunctiontothehistogramandadoptedtheGaussianwidthofthetasthetypical1errorforeachmagnitudeinterval.Figure 4-10 showsourmeasurementsofthe1errorsinthecross-matchedsourcemagnitudesandthe4thorderpolynomialttothesevalues.Toquantifythesignicanceofthemagnitudedifferencesseeninasourceacrosstwoepochswecalculatedthe1errorvaluefortheISPImagnitudeofthesourceusingthepolynomialtanddivideditfromtheabsolutevalueofthemagnitudedifference.Thisgivesthesignicanceofthevariationintermsofthetypicalphotometricscatter.Weuseda3thresholdtodenevariablecandidates.InFigures 4-11 4-12 and 4-13 ,weshowthemagnitudeversusmagnitudedifferencewithsourceswith3-5differencesmarkedinblueand>5markedingreen.Oneobviousfeatureoftheseplotsisthatthereisasignicantlyhighernumberof 100

PAGE 101

Matsunagaetal. ( 2009 )toourISPI/UKIDSScrossmatchedsourcestogaugehoweffectivelyour2-epochphotometrycanidentifyvariabilityinthesesources.904LPVswerematchedtoISPI/UKIDSSsourcesThevastmajorityoftheremaining400LPVsintheMatsunagacatalogwerelocatedoutsideofourISPIGCeld.Figures 4-14 4-15 and 4-16 showthemagnitudeversusmagnitudedifferencesofallcross-matchedsources,withtheknownLPVsmarkedinred.ThemajorityoftheLPVsstandoutwithlargemagnitudedifferencesacrossthe2epochs.ThereisalsoanappreciablefractionofLPVswhichdonotstandoutandappeartohavelittlemagnitudedifferenceacrosstheISPIandUKIDSSepochs.ThisismostlikelyduetotheparticulardutycyclesoftheseLPVsandthetimesinthatcyclethatweobservedthem.WeanticipatethatadditionalepochswouldidentifytheseLPVsasvariables.Ofthe904LPVsintheISPIregion,869haveJ,HorKmagnitudeswhichfallwithinourlimitsforreliablecross-matchedphotometry(12.0
PAGE 102

4-17 4-18 and 4-19 showthemagnitudeversusmagnitudedifferencerelationfortheentireISPI/UKIDSSmatchcatalog,withtheNIRcounterpartstoX-raysourceswhicharealsopresentinthecross-matchedISPI/UKIDSSmagnitudesshowninred.Thegreenpointsshowthemagnitudedifferencesforthe16sourcesinthevariabilitycatalogthathavebeenspectroscopicallyidentied.Theseinclude12CWB/sgHMXBcandidatesdiscoveredbyvariousauthors(e.g. Miklesetal. ( 2006 ), Mauerhanetal. ( 2010b )).TheotherfoursourcesareourtwoOstardiscoveries(XID1944andXID947),aBeHMXB(XID3275)andaprobablesymbioticX-raybinary(XID6592),thediscoveriesofwhicharepresentedinChapter5.Only2spectroscopicallyconrmedX-raysourceshavevariabilitythatwedetectwith3signicance:XID264andXID6592.ThecounterparttoXID264isaCWBcandidatecontainingaWN7-8star( Mauerhanetal. 2010b ).XID6592islikelytobeasymbioticbinarywithanM7IIIsecondary(seeChapter5).ThecounterparttoXID6592isalsocategorizedasaLPVby Matsunagaetal. ( 2009 ).Table 4-1 showsallspectroscopicallyconrmedGCX-raysourceswithinourISPIGCeld.WelistthemUKIDSSmISPIvaluesandthecorrespondingsignicanceofthemagnitudedifference.ManyoftheseIR-brightsourcesweresaturatedinUKIDSSobservations,causingeitheramisseddetectionorunreliablephotometry.Ininstanceswherethemagnitudesarebrighterthanourlimitsforreliablephotometry,wemarkthesignicanceassat.toindicatesaturateddetections. 102

PAGE 103

4-20 .TheUKIDSSimageontherightappearstohaveanirregular,patchybackgroundincomparisontotheISPIimageontheleft.Weconsidervariablesourceswherewecanseetheeffectofpoorbackgroundskysubtractionwithin100ofthesourcepositiontobeunreliable.ResidualstarsintheUKIDSSskyframeswillhaveasubtractiveeffectontheaffectedstarsinthetarget 103

PAGE 104

4-21 weshowthepatternofsourcedetectionsinthevicinityofthevariablecandidatecounterparttoXID3687.TheNIRcounterparttoXID3687isdetectedas2sourcesintheUKIDSScatalogbutonly1sourceintheISPIcatalog.Generally,thisshouldhavetheeffectthattheUKIDSSderivedmagnitudeisfainterthantheISPImagnitude.SincetheUKIDSSimagingisslightlyhigherinresolutionitshouldbemorecommonforUKIDSStocorrectlysplitblendedsourceswhereISPIdoesnot.ThiswillcontributetoanexcessofspuriousvariablesourceswithpositiveKUKIDSSKISPIvalues.Weproceededbyvettingthe92candidatevariableswithintheX-raycounterpartsagainstskybackgroundandsourceblendingissues.Initiallytherewere39candidatesintheKband,39intheHbandand29intheJband.Someofthesamecandidateswerefoundtovaryinmorethanoneband.Wefoundblendingissuesin6Jcandidates(21%),15Hcandidates(38%),and16Kcandidates(41%).Skysubtractionartifactswerefoundfor5Hcandidates(13%),and8Kcandidates(21%).TherewerenoJbandcandidateswithobviousskysubtractionproblems,mostlikelyduetothefactthatthestellardensityintheJbandismuchlowerthanH/K,leadingtomorereliableskysubtractionusingon-sourceskies.Intotal,48ofthe923variabilitycandidateswithinourX-raycounterpartscatalogpassedvisualinspection.Threesourceshad3variabilityinallthreebandsofJ,H,K.Therewasonesourceeachwith3variabilityinJ,HorH,Kbands.Theremaining43sourcesexhibitedvariabilityinonlyoneband. 104

PAGE 105

Matsunagaetal. 2009 ),sixwerefoundtovaryoverthetwoepochswith>3signicance.Onlyoneofourvariabilitycandidateswaspartofthelistof98highprobabilitycandidates(XID2683)thatweidentiedinChapter3.Tables 4-2 and 4-3 listtheassociatedX-raysource,NIRcoordinates,andcross-matchedphotometryforthe48candidatevariableswithinthecatalogofNIRcounterpartstoX-raysources. 105

PAGE 106

4-22 weshowtheexampleoftheKISPIversusKVISTAKISPIscatterplotfortheVISTA/ISPIcross-matchedset.TheKVISTAKISPIdistributionisclusteredarounda0.0magvalue,whichwouldonlyoccurifthemajorityofmatchedsourceswerecorrectlymatched.Wealsoperformedavisualinspectionofasampleofthematchedsourcesovertheimagesandconrmedthatthematchesaremostlycorrect.ThemagnituderangesthatweselectedfortheISPI/UKIDSSvariableswithsecurephotometryappearedtobefreeofsaturationorEddingtonbiaseffectsfortheVISTAcatalog.VISTAandISPInominallyusethesameJHKslterset,derivedfromthe2MASSsurvey.Weassumethatthedifferencesareminimal.FortheUKIDSSKband,weusedthevaluesthatweconvertedtoKs,ISPIwhenwecross-matchedtheISPIandUKIDSSphotometry.Wecalculatedthe1errorofmVISTAmISPIandmVISTAmUKIDSSinthesamemannerasformUKIDSSmISPIinx4.2.3,e.g.byttingaGaussianfunctiontothehistogramofmVISTAmISPI/mVISTAmUKIDSSin0.5magincrements,andadoptingthevalueoftheGaussiantasour1error.Weshowour1errorsversusmagnitudefortheexampleoftheKscross-matchofVISTAandISPIinFigure 4-23 .Weta4thorderpolynomialtotheseerrorvaluesandobtainedananalyticapproximationofthe1cross-matcherrors.The1cross-matcherrorsforVISTA/ISPIandVISTA/UKIDSSallshowedverysimilarvaluesandtrendstotheISPI/UKIDSSvalues,typicallywithaminimumerroroorof0.05mag.ThisgivesuscondencethatthequalityofourVISTAphotometricreductionissufcienttoaddvaluetoourvariabilityanalysis.OursourcesofinterestwithintheVISTAcatalogincludedour48ISPI/UKIDSSX-raymatchedvariables,the15X-raymatchedMatsunagaLPVs(ofwhich6are 106

PAGE 107

4-1 .WelocatedoursourcesofinterestintheVISTAcatalogbyusinga100.5matchingthresholdradiusforeachoftheJ,HandKscomponentsoftheVISTAcatalog.EachsourcematchwasvisuallyinspectedintheVISTAtoensurethatthecross-identicationwascorrectandunaffectedbythesource-splittingseentoaffecttheUKIDSS/ISPIcross-matches.TheVISTAimagingappearedtobefreeoftheskysubtractionissuesintheUKIDSSimages.Oursearchreturned42/48VISTAmeasurementsofourISPI/UKIDSSvariablecandidates,13/15oftheMatsunagaLPVX-raymatchedcandidatesand12/16ofthespectroscopicallyidentiedGCX-raysources.MostofthesourcesthatwerenotfoundinVISTAwereheavilysaturatedandrejectedbyDAOPHOTduringthePSF-ttingphotometryoftheVISTAdata.WecomparedtheVISTAmagnitudestotheUKIDSSandISPImagnitudestondcaseswherethemagnitudedifferencewassignicant,e.g.(mVISTAmISPI=UKIDSS)/3.0).Therewere15LPVvariablesmatchedtoX-raysources.6appearedtohave3variabilityintheISPI/UKIDSSepochs.TheVISTA/ISPIandVISTA/UKIDSSphotometrydifferencesuncovered8as3variables,including4fromthesetfoundwithonlyISPIandUKIDSSand4additionalLPVnotfoundtovaryintheISPI/UKIDSS.Intotal,theVISTAepochincreasesthefractionofknownLPVvariablesdetectedinour3catalogsfrom6 15to10 15.Forthespectroscopicallyidentiedvariables,theadditionoftheVISTAepochmarkedoneadditionalsourceasavariable,whilerediscoveringoneoutoftwooftheoriginallyidentiedmembersofthespectroscopicallyidentiedset.ThenewvariablethatwasidentiedwasthecounterparttoXID2162(GCIRS13E).Thissource,whichissaturatedforUKIDSSHbutnotforISPI/VISTAHexhibiteda0.27mag(6.2)increaseinmagnitudewithrespecttoISPIepoch.XID2162,hasbeenidentiedasaprobableWRcollidingwindbinary( Cokeretal. 2002 ). 107

PAGE 108

4-4 wepresenttheVISTAepochvariabilitymeasurementsinthecaseswhereasourcewasfoundtovaryby3foroneormorebands,foreitherVISTA-ISPIorVISTA-UKIDSS.22oftheoriginal48ISPI/UKIDSSvariablesarefoundtovaryinVISTA/ISPIorVISTA/UKIDSS.Theadditionalepochaddscertaintytotheconrmationofeachvariable.WeanticipatethatoncewextheastrometryissuewiththeVISTAdatathatasignicantnumberofnewvariableswillbeidentied.ThisadvancemayallowustodiscoveradditionalvariablesmatchedtoX-raysourcesbeyondthoseidentiedbyISPI/UKIDSS,providingadditionalIRtargetsforfurtherfollow-upspectroscopy. 4-24 ,weshowtheISPI(HKs)versusKscolormagnitudediagramforallISPIsources,withourcandidatevariablesmatchedtoX-raysourcesshowninblueforsoftsourcesandredforhardsources.InFigure 4-24 weplotthesameISPICMDwiththeLPVsfrom Matsunagaetal. ( 2009 )markedingreen.Thereare15LPVsfrom Matsunagaetal. ( 2009 )matchedtoX-raysources;14havebothHandKsdetectionsinourISPIphotometry.WeshowtheseLPVsinFigure 4-24 withblue(soft)andreddiamonds(hard).NearlyalloftheMatsunagaLPVsourceslieatcolorssignicantlyredderthanthedominantbulgepopulationat(HKs)=2mag.LPVshavelateMIIIspectraltypes.Theirinfraredcolorscanbeheavilyaffectedbycircumstellardust,duetotheaccumulatedmasslosscharacteristicoftheAGBstage.MgiantstarslaterthanM7IIIintheIRTFlibraryhaveintrinsiccolorsofbetween0.4<(HKs)<1.8mag,withalargescatterevenwithinthesamespectralsubtypes( Rayneretal. 2009 ).OurISPI/UKIDSScandidatevariablesmatchedtoX-raysourcesaredispersedacrosstheCMD.Forthepurposeofthisdiscussionwedividethemintoforegroundsources(HKs)<1.3mag(10soft,3hard),11withnormalbulgepopulationcolors1.3<(HKs)<2.2mag(4soft,7hard)and16sourceswithcolorsredderthanthebulge(HKs)>2.2mag(13hard,3soft). 108

PAGE 109

Matsunagaetal. ( 2009 ),becausetheircompletenessdropssharplyafterK10magandour16veryredcandidateshaveaKsrangeof10-14mag.ThefactthatthedominantvariabilityinthesesourcesisstellarinorigindoesnotmeanthattheyarenotalsothetruecounterparttotheirX-raysources.Forexample,wediscoveredtheprobablesymbioticbinarycounterparttoXID6592fromthisregionoftheCMD.(discussedinChapter5).Wedonotexpectthe3matchedtosoftX-raysourcesfromthisregiontobetruecounterpartsbecausetheirX-rayspectraimplyforegrounddistances.Thevariableswithnormal1.3<(HKs)<2.2magbulgecolorsconsistof4softX-raymatchesand7hardX-raymatches.Again,thesoftX-raymatchesareunlikelytohaveinfraredcolorsthisred,andthereforeareprobablyspuriousmatches.ThecauseofvariabilityfromthisregioncouldbefromCepheidIIvariables,orpulsationsfromredgiantsbelowtheredgiantbranchtip,withperiodsof<50daysthatwouldnothavebeenidentiedasLPVs( Itaetal. 2002 ).NormalcolorbulgevariablestrulymatchedtohardX-raysourcescouldbeanumberofactivebinarycandidates,suchasHMXBs,RGBLMXBsorhardX-raysymbioticbinarieswithnormalredgiantstardonors.Thevariableswithforegroundcolorspredominantlycontainsoftsources.AspreviouslyarguedinChapter3,weexpectthatthecounterpartmatchisrealinmanyofthesecases.CandidatesforsoftX-rayvariablesintheforegroundcouldbeAlgolbinaries,RSCVnbinaries,cataclysmicvariablesandnormalsoftspectrumsymbioticbinaries. 109

PAGE 110

4-5 liststhevariablecandidatesforwhichweobtainedspectraaswellastheircolorsandmagnitudesandourderivedspectraltype.WefoundthecandidatevariablematchedtoXID6592(foundinbothourvariablelistandtheMatsunagavariables)tobeaM7IIIwithabrightBrackettemissionline(seeFigure 5-7 ),whichweinterpretasevidenceforahardX-raysymbioticstar.Noneoftheothercandidatesforwhichfortookspectrahadobviousemissionlines.OtherthanXID6592,onlytwosourceshadwellconstrainedspectraltypes,withM4-6IIIspectra.TwooftheothershadspectrawithcharacteristicsofM8+IIIstars,(seeFigures 5-20 5-21 ).ThesestarsmaybefromalaterevolutionarystageoftheAGBphase,withthecircumstellardustfromaccumulatedmasslossmodifyingtheirspectralappearance( Oudmaijeretal. 1995 ).ThevariablecandidatesmatchedtosoftX-raysources(XID6019andXID6010)hadpuzzlingspectrawhichwerenoteasilytyped.XID6010seemedsimilartotheM8+starsfromtheshapeofitsCOabsorptionandjaggedcontinuumappearance(Figure 5-22 ,butitliesinaregionoftheCMDthatisbothfainterandbluerthanthebulkoftheLPVs.XID6019hadanearlyfeaturelessspectrum.WespeculateinChapter5thatitcouldbeapost-AGBcandidatewithheavydustdilutionofthephotosphericfeatures.AsidefromXID6592itisunclearifanyofourspectroscopicallyobservedvariabletargetsaretrulyassociatedtotheirX-raysources.IftheyaretrueX-raycounterparts,thentheirlatetypeMspectraimpliessometypeofsymbioticstar. 110

PAGE 111

111

PAGE 112

Figure4-2. 112

PAGE 113

Figure4-4. 113

PAGE 114

Figure4-6. 114

PAGE 115

DifferenceinJltermagnitudes. Figure4-8. DifferenceinHltermagnitudes. 115

PAGE 116

DifferenceinK/Ksltermagnitudes. Standarddeviationofthescatterinthemagnitudedifferencesforcross-matchedNIRsourcesinISPIandUKIDSS. 116

PAGE 117

Figure4-12. 117

PAGE 118

Figure4-14. Matsunagaetal. ( 2009 )areoverplottedasreddiamonds. 118

PAGE 119

Matsunagaetal. ( 2009 )areoverplottedasreddiamonds. Figure4-16. Matsunagaetal. ( 2009 )areoverplottedasreddiamonds. 119

PAGE 120

Figure4-18. 120

PAGE 121

Figure4-20. ISPIKs(left)andUKIDSSK(right)imagesoftheregionaroundthecounterparttoXID2983.TheUKIDSSimagehasskysubtractionartifactsthatmakethissource's0.3magvariabilityunreliable. 121

PAGE 122

VariabilityinknowncounterpartstoGCX-raysources. AssociatedJukidssJISPIJUIHukidssHISPIHUIsig.KukidssKs,ISPIKUIX-raySourceID(mag)signif.(mag)signif.(mag)signif. XID2470.061.48sat.-sat.-XID264-0.263.91-0.0821.90-0.1342.63XID348-0.0240.290.0471.02sat.-XID514---0.0280.43-0.0921.67XID935--sat.-sat.-XID947-0.0550.980.0020.05-0.0490.96XID1359-0.0150.390.0130.23-0.0170.30XID1733-0.0741.64-0.0240.46sat.-XID1862-0.0992.110.0150.34-0.0190.37XID1944sat.-sat.-sat.-XID2162-0.0791.88sat.-sat.-XID3150-0.0821.91-0.0460.98-0.0731.4XID3275---0.0240.41-0.0240.36XID6592---0.2686.23sat.-XID7189---0.0200.420.1071.95XID7665--sat.-sat.-

PAGE 123

ISPIKs(left)andUKIDSSK(right)imagesoftheregionaroundthecounterparttoXID3687.Thegreenboxpointsindicatewheresourceshavebeendetectedintherespectiveframes.TheUKIDSSsourcedetectionndstwostarsinthestellarblend,whereasISPIdetectsthemasonesource.Thevariabilityvalueof0.24magisthereforeconsideredunreliable. 123

PAGE 124

VariablesourcecounterpartstosoftX-raysourceswith3variationsacrosstheISPIandUKIDSSobservationepochs. 5953266.235381-28.860450soft12.86-0.426.655980266.241037-28.857102soft15.62-0.163.3714.23-0.142.6713.13-0.030.506010266.246021-28.856563soft15.78-0.7515.7314.29-0.6412.2112.96-0.263.986019266.247587-28.861300soft13.56-0.5715.8112.01-0.142.4610.75-0.122.136024266.247971-28.855720soft12.68-0.154.4110.85-0.020.436034266.250364-28.859141soft14.00-0.123.1913.03-0.173.8812.26-0.183.216040266.250659-28.858147soft14.17-0.123.1313.00-0.081.8112.320.020.4012266.258528-28.955975soft14.140.154.0013.830.102.1114.09-0.404.146268266.293015-29.081294soft13.960.215.5713.420.204.576378266.317310-29.155602soft13.930.338.8910.72sat.-437266.337178-28.971874soft13.830.092.4613.87-0.4710.1113.100.142.06711266.356861-29.036005soft16.390.315.1912.300.132.499.950.25sat.823266.364548-28.914347soft15.220.143.2114.680.020.2814.210.010.061084266.375421-29.103717soft16.850.283.7716.270.242.111378266.387513-28.916128soft13.160.133.7112.72-0.010.2412.380.020.371474266.390246-29.023347soft14.920.142.2112.170.183.292443266.425861-29.025544soft16.850.263.5115.630.394.812612266.432978-29.122733soft15.170.143.262750266.440202-28.939393soft15.25-0.051.0714.470.203.673026266.453706-28.921896soft14.430.183.3712.560.020.343053266.455620-29.021960soft12.760.304.823263266.469948-29.030054soft14.440.193.5911.680.071.293459266.485297-29.078745soft15.330.153.4214.900.111.7914.560.131.03

PAGE 125

MagnitudedifferenceKVISTA-KISPIversusKISPIforthecross-matchedVISTA/ISPIcatalogs,afterzeropointingtheVISTAdata. Figure4-23. 1errorsfromthecross-matchedKsmagnitudesofVISTAtoISPI.Thebestt4thorderpolynomialtoourerrormeasurementsisshownwiththedashedline. 125

PAGE 126

VariablesourcecounterpartstohardX-raysourceswith3variationsacrosstheISPIandUKIDSSobservationepochs. 6049266.252292-28.861704hard15.84-0.153.1014.67-0.111.9813.94-0.080.826123266.265626-28.851343hard15.360.429.5713.950.357.3313.040.365.386277266.294367-28.861502hard11.090.153.5610.07sat.6300266.299023-29.163732hard13.170.133.02264266.319688-28.973623hard16.64-0.263.9113.10-0.081.9111.06-0.132.636406266.323271-29.153778hard16.22-0.223.9514.81-0.081.3314.02-0.080.86930266.368953-28.999865hard15.87-0.275.5511.980.000.079.84sat.6592266.369945-29.161870hard13.03-0.276.231016266.373041-29.027871hard16.250.403.6213.190.263.711067266.375577-29.046889hard13.380.133.0510.64sat.6687266.389351-29.133241hard13.80-0.5511.6010.99-0.295.611480266.390577-29.075724hard14.830.233.8211.05-0.050.941747266.400488-29.054453hard13.28-0.263.621817266.403523-29.021604hard13.66-0.303.696752266.404389-29.152165hard15.39-0.101.4113.32-0.233.222164266.415377-29.035308hard13.65-0.415.162211266.417186-29.014798hard15.42-0.527.0712.13-0.346.112236266.417973-29.010954hard18.92-1.164.3612.71-0.214.582264266.419358-29.009343hard14.88-0.162.5912.45-0.203.452463266.427636-28.942741hard13.650.143.0911.360.040.702683266.436555-28.990445hard12.950.163.7211.150.142.693222266.467237-29.029354hard16.39-0.594.8613.38-0.344.603691266.511567-29.056758hard17.770.231.8412.960.235.4010.420.17sat.3740266.518725-29.119653hard17.70-0.100.8313.850.193.9411.970.111.967680266.573823-28.909493hard14.460.264.9111.320.091.84

PAGE 127

Tableofcandidatevariableswithdetected3variabilityinVISTA. 630013.170.254.120.122.01659213.030.040.640.315.121214.140.122.19-0.030.5613.830.131.970.030.4614.090.181.890.586.1771116.390.283.02-0.030.2712.300.061.08-0.071.149.950.060.54-0.201.92244316.850.252.31-0.010.1215.630.424.810.030.31601913.560.010.200.5810.312.010.081.670.234.4710.75-0.060.830.060.84637813.930.529.600.193.510.72668713.80-0.203.210.345.3610.99-0.172.740.111.81601015.78-0.719.340.040.5614.29-0.547.890.101.4812.96603414.00-0.112.070.000.1013.03-0.233.72-0.061.0212.26-0.223.41-0.040.57612315.360.141.85-0.283.8313.950.020.28-0.334.8613.040.080.88-0.283.18101616.250.768.040.353.7613.190.425.270.162.03216413.65-0.293.020.131.36221115.420.091.180.618.2912.130.010.160.356.08226414.880.111.510.273.7312.450.213.540.416.90322216.39-0.343.400.252.4313.38-0.202.390.131.56595312.86-0.121.750.304.4043713.830.173.050.071.3713.870.162.560.6410.1613.100.192.420.060.69148014.830.243.290.010.1811.05-0.122.07-0.071.24223618.920.050.381.218.7212.71-0.060.940.152.57326314.44-0.030.46-0.233.1611.68-0.101.71-0.172.88369117.770.423.150.191.3912.960.304.960.071.1410.420.131.66-0.040.47

PAGE 128

ISPIcolormagnitudediagramshowingthepositionsofvariablecandidatesmatchedtosoftX-raysources(blue)andhardX-raysources(red). Table4-5. SpectroscopicallyidentiedvariablesmatchedtoX-raysources. XIDOurVariables?MatsunagaLPV?Ks(mag)(HKs)SpectralType 861yesyes8.952.06M8+III6055noyes12.832.45M8+III6010yesno12.941.25M8+III?6687yesyes10.912.86M6III6019yesyes10.721.18post-AGB?930yesno9.902.08M4III6592yesyes9.883.15M7III+Bremission 128

PAGE 129

ISPIcolormagnitudediagramshowingthepositionsoflongperiodvariablesfrom Matsunagaetal. ( 2009 )ingreen.MatsunagavariablesthatarematchedtosoftandhardX-raysourcesaremarkedinblueandred,respectively. 129

PAGE 130

Miklesetal. 2006 ; Hyodoetal. 2008 ; Mauerhanetal. 2007 2010b ).Therehavebeenmorethan30spectroscopicidenticationsofGCX-raysourcesfromtheMuno2009catalog( Mauerhanetal. 2010b ).AllconrmedcounterpartstodatehavebeencollidingwindbinaryorHMXBcandidateswithapparentmagnitudesKs<12mag.OursurveyusedSOAR/OSIRIS,andLBT/LUCIFER1totakeNIRspectraof46NIR/X-raycounterpartswith8.5
PAGE 131

Munoetal. ( 2009 )thatresideneartheGalacticCenterdistance.Weusedour2137sourceNIR/X-raymatchcatalogtoselecttargets.Thiscatalogcontains1565hardX-raysourcematchesand572softX-raysourcematches.ThedivisionbetweensoftandhardX-raysourceswasmadetodistinguishX-raysourceswithlowextinctionthatarewithin4kpcoftheEarthandsourceswithhighextinction,whicharelikelytolieat4kpcandbeyond.WeuseathresholdhardnessratioofHR0<0.175fordeningsoftsources,whereHR0hs h+s,sisthenetsoftphotoncountswith0.21.0magthatwerematchedtohardX-raysources.AsdiscussedinChapter3,thesematchesfrequentlyrepresentchancealignmentsoftheX-rayandNIRsourcecoordinates,whichmakestargetingthetruematchesachallenge.InChapter3weidentiedsetsofnear-infrared/X-raypropertiesthatmaximizedthenumberofprobabletruecounterpartsoverthespuriousmatches.Wefound98reddenedNIRmatchestohardX-raysourceswitha459%likelihoodofbeingtruecounterparts.AllofthesesourceshadX-raypositionalerrorsofX1.0"anddetectionsinJ,HandKs.InadditiontheyhadeitherrelativelybrightX-rayuxes(withfX0.0001countss1)orfaintX-rayuxeswithacounterpartKsmagnituderestricted 131

PAGE 132

Mauerhanetal. ( 2009 ): X)2(5)toparameterizetheclosenessofaNIR/X-raymatch.Inthisequation,istheangularseparationoftheX-rayandinfraredpositionandXisthepositionaluncertaintyoftheX-raypointsource.Thebvaluerunsfrom0.2
PAGE 133

5-1 5-2 ,and 5-3 .IneachtableweincludetheofcialIAUnameofeachX-raysource(e.g.CXOUJ174533.5-285540)andthecorrespondingrecordnumberwithintheChandraX-raysourcecatalogfrom Munoetal. ( 2009 ).Forthesakeofbrevity,weusetheX-rayrecordnumber,designatedbytheprex,XIDforreferringtoX-raysourcesandtheirNIRcounterparts. 133

PAGE 134

DeWittetal. ( 2010 ).Wewereabletoposition7ofthesesourcesonbothmasks.Themaskcenterpositionswere=17:45:50,=-28:57:29(J2000)forMask1and=17:46:00,=-28:54:24(J2000)forMask2.LUCIFER1hadonlyrecentlybeencommissionedforMOSmodeatthetimeofourobservations.Theproceduresformaskalignmentandguidingwerestillbeingoptimizedandwereexpectedtotakeuplargeamountsofoverheadtime.Therefore,insteadofusingknownA0Vtelluricstandardsthatwouldrequirerecenteringtheslitmask,weusedunreddenedsourcesintheMOSmaskeldofviewwhichhadNIRcolorsconsistentwithstarsearlierthanG0.Thisapproachhasthedisadvantagethattheexactspectraltypeofthereferencestarisunknowninadvanceandmustbederivedaspartofourspectroscopicanalysis,whichisaffectedbytheatmospherictransmissionspectrumthatwedesiretoremove.WedescribetheprocedureforcorrectingforatmospherictransmissionforLUCIFER1inx5.2.2.2.Mask1andMask2weredesignedfor18and21targets,respectively.Eachtargetslitwas0".56.0".Duringobservations,theMOSmaskswerealignedtobrightreferencestarswithinthemaskeldofviewandanABBAnoddingsequencecontainingeight100secondexposureswastakenforeachmask.Thenodlengthwas2.500.Weattemptedtoselectonlytargetsthatwerefreeofstarsintheskypositionsofthenodpatterninordertofacilitatecleanskysubtraction.Thiswasnotpossiblefor1 2ofthetargets,andinthesecasesweselectedtargetswithclearskyinjustoneofthenod 134

PAGE 135

Warneretal. 2008 ).Webeganthereductionprocessbysubtractingdarkframesfromallimagestoremovethedetectordarkcurrentandthendividingbyspectralateldstakenwitheachslitmasktocorrectforthegratingblazefunctionandthedetector'swavelengthsensitivity.ThenwesubtractedthespectralframesoftheBditherpositionfromtheAditherposition,pairingtheAandBframesbythetimetheyweretaken.Inprinciple,theresultantsky-subtractedspectrashouldhaveanyatmosphericOHemissionlinesremoved,andcontainonlysignalfromthetargetspectra.Thesky-subtractedframesincludetwospectraforeachtarget,onepositiveandonenegativevalued.Wetooktheabsolutevalueofeachsky-subtractedframeandcoaddeditwithanappropriatelyshiftedversionofitself,sothattheresultantimagecontainedonespectrumpertarget.Thisstepiscalleddoublesubtraction.Wecoaddedalldouble-subtractedframes.Atthispointwecorrectedthecurvatureofthespectrallineandcontinuumshapesofthe2-Dspectrabyperformingarecticationsothatthespatialandspectralchannelsofthe2-Dwereallalignedinxandypixelspace.Weextracted1-Dspectrabysummingthe2-Dspectrainthespatialdirection.Attheskysubtractionphase,amedianimagewasalsocreatedwhichisolatestheOHskylinesratherthanremovingthem.Werectiedthisskylineimageusingthesametransformationasforthesky-subtractedspectra.Wecalibratedthewavelengthsolutionforeachtargetspectrumusinga2ndorderpolynomialttothecentersofatmosphericOHemissionlineswithknownwavelengthinthemedianimage.Forsometargetstherewasacontaminatingstarinthetargetslitthatwasacquiredattheskypositionofoneof 135

PAGE 136

Davies ( 2007 ).Weextractedthecenterpartsofthemedian-stacked2-Dspectra(whichisolatedtheOHskyspectra)togenerateatemplateoftheOHsky-linesforeachtargetspectrum.Wetapowerlawtothethermalcontinuumintheskytemplatespectrumandremovedit.Thentheskylineswerebrokenintogroupswhichsharethesameintensityvariations( Davies 2007 ).ThesedifferentcomponentsoftheskytemplatewerethenaddedorsubtractedtovisuallyremoveresidualOHemissionlines.Toproducetellurictransmissionspectraweneededtoremoveintrinsicabsorptionlinesfromthetelluricspectrum,whichrequiresknowledgeofthespectraltype.WechosetargetswithJKs<0.1magandwhichwereunblendedinourISPIimaging.AtthisthresholdourtelluricreferencespectraltypesshouldhavebeenlimitedtoG0typesandearlier( Ducatietal. 2001 ).O/BstarshaveHeIabsorptionfeaturesat=2.112=3mand1.700m.Neitherofourselectedstandardsshowedthesefeatures,whichconstrainsthespectraltypestoA0-G0.Forthesespectraltypes,theonlystrongabsorptionfeaturesshouldbetheBrackettseries.BrwaseasilydetectedinKband,aswasBr10-4throughBr16-4inHband.WeusedasynthetictelluricabsorptionspectrumcreatedforMaunaKeaconditionsatsec(z)=2.0withATRAN( Lord 1992 ).Wescaledthisspectruminpositionandwithanexponentanddividedinfromourtelluricspectrainordertobestremovethedeepabsorptionbandsat2mand2.05mintheKbandand1.6mand1.63mintheHband.Thisallowedustosee,forthemostpart,theprolesofBrackettabsorptionfeatures,thoughthetelluricline-correctedregionswerestillquitenoisyduetotheusageofanarticiallyderivedtransmissionspectrum. 136

PAGE 137

Castelli&Kurucz 1994 )toremovetheBrackettlinewhichislikelytoexistinthisregionofourtelluricstandardstars,butweconsidertheresultantsignalfrom1.8<<1.95mtobehighlysuspect.Aftertheseintrinsicstellarabsorptionlineswereremoved,werestoredthetelluriclines(whichhasbeenonlypartiallyremovedbythesynthetictelluricspectrum),andnormalizedtheresultantspectrum.Atthispointwehadanormalizedtelluricspectrumwiththeunderlyingstellarfeaturesfromwhichitwasderivedremoved.WedividedthistelluricspectrumfromallourLUCIFER1targets. DeWittetal. ( 2010 ).WefocusedontargetsthathadKs<12.5mag.Intotal,weobserved30candidatecounterpartstoX-raysourceswithOSIRISonAugust19-20,2010.TheexposuresweretakenwithABBAnodding,wherethenodlengthwascarefullychosentoavoidcontaminatingstarsintheskypositionsforbothtargetsontheslit.WetookaspectrumoftheA1VstarHD160461atleastonceperhourforthepurposeoftelluriccalibration.Thisstarwasnevermorethan0.05differentinairmassfromanyofourGalacticCentertargets. 137

PAGE 138

5-1 weshowthespectrafromLBV1806-10takenonsuccessivenights.Thesecondnightspectrumshowsfourlargeregularlyspaceddipsinthecontinuumthatwerenotpresentintherstnightspectrum.Weobservedsimilarfeaturesinroughly50%ofourtargets.Byvisualexaminationwenotedthatthefeaturechangedinamplitudeandcentralpositionbutthespacingandrelativeamplitudesofthedipsappearedtostayconstant.ThisisaknownbutunsolvedissuewithOSIRISsinceitsinstallationonSOARfromitspreviouslocationatthe4mCTIOtelescope(R.Blum,priv.comm.).WecreatedatemplatefortheshapeofthesefeaturesbyttingGaussianfunctionstothefourpitsintheratioofthenight2andnight1OSIRISspectraofLBV1806-10.ThepittingtemplateisshownatthebottomofFigure 5-1 .Foreachtargetthatwasaffectedwevisuallymatchedtheamplitudeandpositionofthepittingtemplateanddivideditfromthetargetspectrum.ThetelluriccorrectionwasperformedbyrstremovingtheintrinsicBrackettabsorptionofthetelluricstandardbyttingaGaussianfunctiontotheBrackettpositionanddividingitout.Wethennormalizedthetellurictransmissionspectrumto1.0anddividedtheresultanttelluricstandardspectrumoutofallthetargetspectra. 138

PAGE 139

5-24 139

PAGE 140

5-4 .TheX-raypropertiesofeachsourcearelistedinTable 5-5 ( Munoetal. 2009 ). 5-2 .Thespectrum,observedwithLUCIFER1,clearlyshowsemissionfromtheBrackettseriesfromH7-4toH18-4,aswellasHeI2.0589m(seeFigures 5-3 and 5-4 ).ThereisalsoamarginaldetectionoftheMgIIdoubletat2.138mand2.144m.Noabsorptionlinesareseeninthespectra,andinparticularthereisnoevidenceofaCOfeatureat2.295m.AtasignaltonoiselevelS=N=40perresolutionelement,theabsencerulesoutspectraltypeslaterthanF0.ThelineidenticationsandpropertiesfortheXID3275NIRspectraarelistedinTable 5-6 .OnthebasisofthepresenceoftheBrackettseriesemission,andthelackofabsorptionfeatures,thisstarappearssimilartospectraofBestars.Bestarsareyoungmassivestarsthatarerotatingneartheirbreak-upspeed,andthecast-offmattercreatesanequatorialcircumstellardisk.Thediskmaterialoftenuorescesduetotheultra-violetemissionfromtheBstarphotosphere. Steeleetal. ( 1999 )and Steele&Clark ( 2001 )presentacatalogofHandKbandspectraof57optically-typedBestarsbetweenO9-B9.BrackettemissionandHeIemissionaretheidentifyingcharacteristics 140

PAGE 141

Reig 2011 ).BeHMXBsareknownforhardpower-lawX-rayemission,whichshouldbedetectablebyChandraattheGCdistance,makingthisaplausiblecandidateidentication.TypicallyBeXRBscontainaneutronstarwithaneccentricorbitof100days.AccretionepisodesoccurduringtheclosestapproachoftheneutronstarandoftenexhibitX-raypulsationsduetotheneutronstarrotation.Thehighstatescanbeveryluminous,reachingLX=1038ergss1,whilequiescentluminositiesaretypically10331035ergss1.Ingeneral,BeHMXBswithfainterandsteadieremissionareassociatedwithlongerperiodsandlowerorbitaleccentricities.ForaBeHMXB,XID3275hasanunusuallylowluminosityof1032ergss1.GiventhelargerangeofBrackettstrengthsfoundforordinaryBestars(EW=8AinabsorptiontoEW=24Ainemissionforthesourcesinthe Steeleetal. ( 1999 )sample)thereisnoclearevidencefororagainstaccretionactivitycontributingtotheemissionfeaturesinourspectrum.Theequivalentwidth(EW)oftheBrackettemissionintheXID3275counterpartis-10.3A,whichistypicalforBestars( Clark&Steele 2000 ).MgIIisonlyseeninstarsearlierthanB4( Steele&Clark 2001 ).ThereforeweprovisionallyconstrainourcounterparttoaO9-B3star.TheemissionlineFWHMareconsistentwithanorigininthecircumstellarmaterial.UsingtheFWHMBr-vsin(i)relationshipfrom Steele&Clark ( 2001 )theFWHM=20Atranslatestovsin(i)=164kms1.TheBestarsintheClarksamplehavearangeof90kms1
PAGE 142

5-5 ),whichimpliesadistanceclosetothatoftheGalacticCenter.O9-B3starshaveintrinsiccolorsof(JH)0=0.12magor0.10<(HK)0<0.05mag,accordingto Ducatietal. ( 2001 ).BeHMXBsareoftenredderthanthisduetothecircumstellarmaterial;thetypicalrangegivenin Reig ( 2011 )is0.1<(JK)<1.1mag,whichtranslatesto0.02<(HK)<0.2mag,approximatingtheabsorptionratioswithvaluesfromtheGCinterstellarabsorption( Nishiyamaetal. 2009 ).Therefore,weadoptamidpointvalueof(HK)0=0.1magforXID3275andaGCdistancevalueof8.3kpc( Gillessenetal. 2009 ).UsingextinctioncoefcientsderivedfortheGalacticCenterby Nishiyamaetal. ( 2008 )wecalculateanextinctionofAKs=2.373.ThedereddenedmagnitudesarethusJ=10.54mag,H=10.65magandKs=10.71mag.WecalculateanabsolutemagnitudeofMKs=3.81mag.Mainsequence09-B3starshaveabsolutemagnitudesof0.8
PAGE 143

Munoetal. 2009 ).Wenote,however,thattherewereanumberoflonggapsinthecoverage:two400daygapsandtwo260daygapsduringwhichXID3275couldhaveundergoneanoutburstandreturnedtoaquiescentstate.TheimplicationcouldbethatXID3275isalargeseparation,loweccentricityversionoftheaverageBeHMXB.FiveBeHMXBsareknowntohaveorbitaleccentricitiessmallerthane=0.2andwithP30days( Reig 2011 ).DuetothelackofcloseapproachestotheBestar,thesesystemsareobservedtohavesteadyluminositiesof1033ergss1.Alternatively,XID3275maybelongtotheCassubclassofBeHMXBs.ThesesystemshavehardX-rayemissiondominatedbya12keVplasma,insteadoftheusualhardpower-lawseeninotherBeHMXBs( LopesdeOliveiraetal. 2010 ).Theirluminosityisfoundintherangeof1032-1033ergss1,whichmakesthemareasonablecandidatefortheX-rayluminosityseeninXID3275.CassystemsalsodonotdisplaythelargeamplitudeoutburstscommontotraditionalBeHMXBs.Instead,smalleramplitudevariabilityisseenonalltimescales,fromhourstomonths( LopesdeOliveiraetal. 2010 ).ThehourlyickeringhasbeenusedbysomeauthorstoarguethatCassystemshostaWDinsteadofaneutronstar,byanalogytothehightemperaturethermalplasmasandickeringseeninsomeCVs( Reig 2011 ).Thelargestobservedamplitudevariabilityhasbeenafactorof3overtimescalesof50-90days. Munoetal. ( 2009 )donotndevidenceforuxvariationsbetweenthe35observationsofthissource;howeverweagainnotetheirregularityoftheChandracoverage.Thequestion 143

PAGE 144

50ascommonasneutronstarBeHMXBsandhavelongerquiescentperiods( Reig 2011 ).ThequiescentuxofBHBeHMXBsmaybeexpectedtobelowerthanforNSBeHMXBsbecauseoftheeventhorizon'sabilitytoadvectmaterialacrosstheboundaryinsteadofignitingnuclearburningonthesurfaceofaneutronstar.BlackholesareseentobelessX-rayluminousduringquiescencethanneutronstarsforLMXBs( Remillard&McClintock 2006 ).Forafavorableinclinationtheradialvelocitymeasurementscouldhelpdiscriminatebetweenaneutronstarorwhitedwarfandonecontainingablackhole. Skrutskieetal. 2006 ),SIRIUScatalog( Matsunagaetal. 2009 ),andourISPIdata(seeFigure 5-6 ). Matsunagaetal. ( 2009 )identifyitasalongperiodvariablebasedonHandKsphotometricmonitoring,buttheywerenotabletoidentifyaperiod.ThissourcewasalsodetectedinallfourbandsoftheIRACcatalog( Ramrezetal. 2008 ),with[3.6]=8.252mag,[4.5]=7.817mag,[5.8]=7.221mag,and[8.0]=7.320mag.TheX-rayuxforXID6592is2103photonss1,whichplacesitinthebrightest0.2%ofhardX-raysourcesinthe Munoetal. ( 2009 )catalog.Thefactthat 144

PAGE 145

Matsunagaetal. 2009 )impliesthatthechanceforaspuriousmatchisverysmall.TheKbandOSIRISspectrum,showninFigure 5-7 ,showsdeepCObandsbeyond=2.295manddeepNaIandCaIabsorptionlines(2.2062/2.2090mand2.26311/2.26673m,respectively)aswellasadepressioninthecontinuumbelow=2.09mduetosteamabsorption.WeidentifythespectrumasapotentialaccretingsystemonthebasisofthelargeBremissionline.However,aswediscussbelow,therearenon-binaryinterpretationsaswell.OurlinemeasurementsfromthespectrumarelistedinTable 5.6 .WespectrallytypedthecounterparttoXID6592bycalculating2differencestotheIRTFspectrallibrary( Rayneretal. 2009 ).Firstwettheslopeofthecontinuumbetween2.1mand2.285mandremovedtheslopefromboththecounterpartandtheIRTFtemplates.ThenweshiftedthetemplatespectratothevelocityoftheXID6592velocity,usingacrosscorrelationfunction.Tocalculatethe2value,weassumedthetemplatespectrumhadnegligiblenoiseandusedthestandarddeviationbetweenthecounterpartspectrumandasmoothedversionofitselfasthenoisevalue.Wecalculatedthe2valueintheregionbetween=2.24mand=2.35m,whichcontainsthesegmentoftheCObandacquiredbyOSIRIS,aswellastheCafeatureat2.285m(Table 5.6 ).ThebestmatchforXID6592intheIRTFspectrallibrary( Rayneretal. 2009 )wasfoundtobeaM7IIIstar(HD108849),whichisasemi-regularvariablestar.HD108849alsohasadepressionbluewardof2.1m,indicativeofwaterabsorptionfoundtypicallyinverylatetypethermallypulsatingasymptoticgiantstars.SupergiantshadgoodtstothecontinuumbetweentheCOedgesandgenerallyhadequallygood2valuesasthelateMgiants,buttheirCObandminimaovershotthelevelsinXID6592by15%andmore.XID6592wouldalsobeunderluminousfora 145

PAGE 146

Blumetal. ( 2003 ),whichareusedforndingthetemperaturesoflatetypeGalacticCentergiantsandsupergiantsandcullingoutlongperiodvariables.WithalargeH2Oindexof20%theXID6592counterpartwouldhavebeenidentiedasaborderlineAGBstarorlongperiodvariable(LPV)inthe Blumetal. ( 2003 )study.ThecounterparttoXID6592isidentiedasaphotometricvariablebyouranalysisinChapter4andby Matsunagaetal. ( 2009 ). Matsunagaetal. ( 2009 )classifythesourceasanirregularlongperiodvariable.Theywerenotabletondapulsationperiodintheirobservations.InFigure 5-8 weshowtheirHandKslightcurvesoftheXID6592counterpart,withtheadditionofourISPIandUKIDSSphotometry.TheISPIKsbandandHbandseemtobeoutliersfromtheSIRIUSHandKslightcurve;bothourmeasurementsaresignicantlybrighterandmayindicateaaringepisodeinthedata.Wewouldnotinsistonthisinterpretation,sincethesourceisverybrightandourmeasurementcouldbeaffectedbypixelnon-linearity.DetailedcomparisonofourISPIphotometryandtheSIRIUSphotometryshouldbeperformedtoverifythepossiblearingepisode.TheUKIDSSHvalueisconsistentwiththetrendoftheSIRIUSHbandlight-curve.WedonotplottheUKIDSSKmeasurementbecauseitwassaturated.ThelongperiodvariabledesignationisconsistentwiththebehaviorofthebesttspectrumofthestarHD108849.However,thelightcurvemayalsoreectaccretionactivity.TheintrinsiccolorforthetemplateM7IIIis(HK)0=0.414mag( Rayneretal. 2009 ).WeusetheXID6592counterpartmeancolorfromtheobservationsof Matsunagaetal. ( 2009 )((HKs)=2.89mag),andanextinctionscalingofAV:AH:AKs=16.15:1.74:1( Nishiyamaetal. 2008 2009 ),tocalculateAKs=3.3magandAV=53.2magfortheextinctiontoXID6592. 146

PAGE 147

5-9 ),soitmayeitherliebeyondtheGCdistance,orbesubjecttointrinsicextinctioninadditiontowhatispresentinthebestttemplate(meaningthat(HK)0>0.414mag).WeadopttheGCdistanceof8.3kpcbutnotethatitmayrepresenttheminimumdistancetothisobject.FortheGCdistancewecalculateanabsolutemagnitudeMKs=7.91mag,givenaspectraltypeofM7III.ThisvaluefallswithintherangeofGCLPVsspectrallytypedby Blumetal. ( 2003 ),whichrangefrom7.762.3m,duetotheirregularshapesofHeNeArlinesinOSIRISatthe 147

PAGE 148

Wallace&Hinkle ( 1997 )spectralatlasincludes9spectraofoCeti(Mira)takenatdifferentphasesacrossthepulsationperiod.WemeasuredtheBrackettEWsofwithinthespectraofdifferentpulsationphasesandfoundittovarybetween0and1.170.2A,signicantlysmallerthantheEWwemeasureforthissource. Chakrabartyetal. ( 1998 )measuredtheblueshiftedemissioninthesymbioticbinarypulsarGX1+4andfoundvelocitiesupto25070kms1forHeI1.083.GX1+4isknowntocontainapulsar. Chakrabartyetal. ( 1998 )consideredwhethertheblueshiftedemissionlinecouldoriginatefromtheionizationofthewindoftheMgiant,orfromawinddrivendirectlyfromthepulsaroraccretiondisk.SinceMgiantwindspeedsareexpectedtohavearangeof10-30kms1( Dupree 1986 ),theyconcludeinfavorofanoutowdrivenbythepulsaroraccretiondisk.Ourvalueof8020kms1isclearlyinconsistentwiththeknownvelocityvaluesofself-shockedemissionfromMiras.ThereforewebelievethatthereisasecondbodyinorbitaroundtheXID6592redgiantcomponentthatiseitherionizingtheredgiant'swindordirectlydrivinganoutow.ThehighvelocityismarginallyinconsistentwithanionizedRGBwind,andfavorsadirectlydrivenoutowfromanaccretiondiskorthecompanionobject.Twentypercentoftheobjectsinthe Belczynskietal. ( 2000 )symbioticbinarycataloghaveirregularvariablesorMiravariablesassecondarystars.Theirmasslossatthisevolutionarystagemakesthemeffectivemassdonorstoanycompactobjectsin<100dayorbits( Murset&Schmid 1999 ).TheremainderofsymbioticsecondariesconsistofnormalrstascentMtypegiants.Mostsymbioticstarsdiscoveredsofarcontainwhitedwarfcompactobjects.ThemajorityofthesesourcesshowX-rayscharacteristicof0.4-1keVplasmas,asaresultofquasi-steadynuclearburningofaccretedmaterialontheWDsurface( Muersetetal. 148

PAGE 149

).TheexceptionsareNSsymbioticX-raybinariesandthenewlyemergingclassofhardX-rayemittingwhitedwarfsymbioticbinaries.TheWDorNSidenticationisoftenmadebyX-rayspectraltswhichsimultaneouslysolveforaccretionrate(bymeasuringtheinternalabsorptionoftheX-rays)andtotalX-rayluminosity.Theamountofenergyliberatedbyaccretionontoacompactobjectisproportionaltoitssurfaceradius.Thus,oncetheaccretionrateandresultingluminosityisknown,theradiusofthecompactobjectcanbeconstrained( Luna&Sokoloski 2007 ).Theradiusofawhitedwarfhasanegativedependenceonthemass( Provencaletal. 1998 ),allowingX-rayspectralttingtosolvefortheWDmass( Luna&Sokoloski 2007 ).Thereareseveralsymbioticbinariesnowknowntohostneutronstarprimaries,includingGX1+4,4U1700+24,4U1954+31,IGRJ16194-2810,IGR16393-4643andIGR16358-4726( Luna&Sokoloski 2007 ; Nespolietal. 2010 ).TheseobjectsdisplayX-rayluminositiesbetween1032-1034ergss1( Masettietal. 2002 ; Nespolietal. 2010 ).GX1+4hasbeenobservedinoutburstat1037( Staubertetal. 1995 ). Masettietal. ( 2006 )tookopticalspectraofthecounterpartsto4U1954+319and4U1954+31duringperiodsofrelativelylowX-rayluminosityof1032ergss1.Theyfoundtheopticalspectraforbothsourceswerecompletelydevoidofemissionlines,withthecontinuumdominatedbythesecondaryMIIIstars.ThelackofBalmerlinesintheopticalmakeitunclearwhetherthesesourceswouldhavedetectableBrackettemissionintheNIR. Nespolietal. ( 2010 )tookKbandspectraofIGR16393-4643andIGR16358-4726andfoundtenuousdetectionsofBrackettandHeI2.0581ineach.ThespectrumofGX4+1takenintheKbandby Chakrabartyetal. ( 1998 )hasstrongemissionatBrackettandnootheremissionfeatures.Thereareatleast13knownhardspectrumsymbioticsthoughttohostwhitedwarfs,includingRTCru,TCrB,CHCyg,CD-573057,SS7317and8othersrecentlydetectedbytheX-raytelescopeontheSwiftGammaRayObservatory( Luna&Sokoloski 2007 ; Smithetal. 2008 ; Lunaetal. 2010 ). Luna&Sokoloski ( 2007 )modeltheX-rayspectra 149

PAGE 150

Sion&Ready 1992 ).Similarlyhighwhitedwarfmassesarecalculatedfortheothersystemsinthisclass( Lunaetal. 2008 ; Smithetal. 2008 ; Lunaetal. 2010 ). Eze ( 2011 )suggestthatthissourceclasscouldbeamajorchannelforTypeIaSupernovae.SNIa'sareusuallythoughttoresultfromthecollapseofChandrasekharmass(1.4Mo)COwhitedwarfs.Variousaccretingbinarieswithwhitedwarfshavebeenproposedastheprogenitors,butmostofthesesystemsareeithertoorarefortheobservedSNIarateorunlikelytoaccreteenoughmasstoreachtheChandrasekharmasslimit( vanKerkwijketal. 2010 ).HardspectrumWDsymbioticsmaysignifysystemsthathavenearlysucceededinreachingtheChandrasekharmassandthusmayprovetobeamajorchannelforSNIa( Eze 2011 ).TheX-rayluminosityrangefoundinhardX-rayWDsymbioticsisbetween1032-1034ergss1( Luna&Sokoloski 2007 ; Smithetal. 2008 ).TodateweknowofnopublishedNIRspectraofanyofthesesources,buttheopticalspectraareknowntohaveBalmerlineemissioninatleastthreecases( Cieslinskietal. 1994 ; Mukaietal. 2003 ).ThisgivesusreasontoexpectBrackettintheNIRfromhardspectrumWDsymbiotics.Thus,wehave2strongcandidatesforthenatureofthissystem;NSsymbioticsandhard-spectrumWDsymbiotics.EachofthesecandidateshavetheluminosityrangeandhardspectralcharacteristicsofXID6592.ThenatureofthecompactobjectcouldinprinciplebeconstrainedwithobservationsofNSburstsorpulsations,orbyX-rayspectralmodelingoftheX-raysource,thoughitisunclearwhetherthe150photon 150

PAGE 151

5-10 ).Itsinfraredcolors(HKs)=0.78magnominallymeanthatitshouldbecloserthan4.0parsecs,andwasthereforenotinourhighprioritylist.However,therelativelybright,hardX-rayspectrumandthematchtoabrightNIRsourcemadeitanunusualandinterestingtargetandweincludeditinourOSIRISobservations.ThecounterpartKbandspectrumclearlyshowsemissionfromHeI2.0587mandHeI2.112mandaswellasamarginallydetectedabsorptionlineatHeII2.189m(seeFigure 5-11 ).ThelineequivalentwidthsandotherparametersarepresentedinTable 5-9 .ThepresenceoftheHeIIabsorptionandtheHeIemissionlinesarecharacteristicsofOtypestars.Byeye,thespectrumismostsimilartoO3-7Isupergiantspectrapresentedin Hansonetal. ( 2005 ),duetotheappearanceoftheemissionlines.Radiativemodelingofearlytypesupergiantsshowthatthefeatureat2.112/2.113misablendofNIII,HeI,OIIIandCIII( Martinsetal. 2008 ).ThemodelspectraalsohaveCIVat2.07and2.075m,whicharemarginallydetectedinourspectrum.ThetheoreticalintrinsiccolorsandmagnitudesforOstarswerecalculatedby Martins&Plez ( 2006 ).ForO3I-O7Itheyndcolorsof(JH)o=0.11magand(HK)=0.10magandamagnituderangeof5.53
PAGE 152

( 2006 )is4.95.1kpc.ThisdistanceseemshighconsideringthattherelativelyunreddenedNIRcolorsandthatthehardnessratioHR0=0.168isneartothehard/softthresholdvalueofHR0=0.175.Thesethresholdvalueswerecalculatedby Munoetal. ( 2009 )tondsourceswithinadistanceof4kpc.Wealsoexpecttheobservedcolorwouldbe(HKs)>1fordistanceslargerthan4kpc( Mauerhanetal. 2009 ).IfweusedthelowerboundvalueofMK=5.27magforOsupergiantsdiscoveredintheGC,weobtainadistanceof3.7kpc.Duringthepreparationofthismanuscript,thecounterparttoXID1944wasalsoidentiedviaPaschenexcessdetectedinnarrowbandimagingfromHubble/NICMOSobservations( Mauerhanetal. 2010a ).TheytookaKbandspectrumwithSOAR/OSIRISon27May2010.Thoughtheydonotpublishlinemeasurements,theirspectraandoursappearconsistent,includingthenotableabsenceofdetectableBremissionorabsorption.TheyconcludeonthebasisoftheirspectraltypingandextinctionanalysisthatthisisanO4-6IsupergiantbycomparisontotheO4-6Ispectrapresentedby Crowtheretal. ( 2006 ). Mauerhanetal. ( 2010a )calculateadistanceof3.6kpc,basedtheiradoptionoftheMK=5.14magvaluefromasecondOI4-6starintheGCannouncedinthesamepaper.TheyalsomodeltheChandraspectrumandndthatthesoftandhardX-rayemissioncanbeaccountedforwithathermalplasmawithkT=0.90.2keV.SincethistemperatureiswithinthetypicalrangeforisolatedOsupergiants,theyconcludethatthereisnoevidenceforinvokingbinarywinds.Ourdataareconsistentwiththeirconclusions. 5-12 .WeobservedtheKbandspectrumwithOSIRIS.WeidentifyitasaprobabletruecounterpartbecauseofthelackofapparentCOfeaturesat>2.295mandbecauseofthemarginaldetectionsofanHeI/NIIIP-Cygnifeature 152

PAGE 153

Hansonetal. 2005 ).ThisspectrumisdifculttotypebeyondthelackofCOabsorptionbecausethetellurictransmissionwaspoorlycorrectedbythetwoobservationsoftheA1Vstartakenbeforeandafterthistarget(seeFigure 5-13 ).Thismaybeduetothefactthattheslitpositionangleofthetelluricreferencespectrawastakenneartheparallacticangle,and100awayfromthepositionangleusedfortheobservationofXID947.Thespectrumofthissourcemaythereforebeaffectedbywavelengthdependentslitlosses,exacerbatedbythehighairmassatthetimeoftheobservations(sec(z)=2.1).WeproceedbyadoptingtheintrinsiccolorsofOstarsfrom Martins&Plez ( 2006 ):(HK)=0.10magandJH=0.11mag.Weadopt(HKs)=1.78magforthereddeningandcalculateAJ=7.26mag,AH=4.16mag,AKs=2.40magandAV=38.8magusingextinctionscalingsfrom Nishiyamaetal. ( 2008 ).TheextinctiontocolumndepthconversionratioNH 5-14 ).ThereforeweadopttheGalacticCenterdistanceof8.3kpcforthissource.Forthisstar,wecalculateabsolutemagnitudesofMJ=5.62mag,MH=5.46mag,MKs=5.39mag.ThismagnitudeiscompatiblewitheitherOI3-9.5supergiantsandO6IIIgiantsorO3Vdwarfs( Martins&Plez 2006 ).UsingtheChandraPIMMSproposalplanningtool,andthesameGCdistanceof8.3kpc,wecalculateLX=1.91031ergss1.ThisisonthelowendfortheknownOsupergiantsthataredetectedintheGC( Mauerhanetal. 2010b ).AsingleOVofOIIIstarcouldinprinciplegeneratethisluminosityinX-rayphotons,frominternalshockswithinitswind.ThetypicalscalingoftheX-rayuxtothebolometricluminosityisLbol Oskinova 2005 ).Forabolometriccorrectionof-4.4( Mauerhanetal. 2010a ),log(Lbol)=5.5.ThisyieldsaratioLbol 153

PAGE 154

Oskinova 2005 ).However,ourPIMMSspectralmodelcalculatedthetotaluxofonlyapowerlawcomponent,extrapolatingtheuxtosoftX-rayenergiesthatwereabsorbedinourspectrum.IftherewasanadditionalsoftthermalX-raycomponenttheresultantuxwouldalsobeabsorbedbutitwouldnotbetakenintoaccountbyourPIMMSluminositycalculation.MostofthesourcesusedtoderivethecanonicalLbol Oskinova 2005 ).Therefore,XID947mayhaveatypicalX-rayluminosityforanOstar,whenallspectralcomponentsaretakenintoaccount.WeconcludethatXID947islikelytobeanO-typestarofuncertainevolutionaryclass,thatmaybeisolated,orinacollidingwindconguration,orquietlyaccretingontoacompactobject.Clearly,ahigherS/Nratiospectrumisneededtomakeanysignicantconclusionsaboutthissource. 5-16 .SincethisistheonlyclearfeatureinthespectrumwecomparetheBrequivalentwidthstothosewemeasuredforstarsintheIRTFspectrallibrary( Rayneretal. 2009 ). 154

PAGE 155

5-15 weplottheIRTFlibraryBrackettEWsagainstspectraltypeforthespectralsequenceofF0VtoG9.5V,andweplaceXID3178andXID3390ontheplotbasedontheirBrequivalentwidths.ThescatterintherelationseeninFigure 5-15 andthemeasurementuncertaintyintheBrackettstrengthsleadustoconsideraspectraltyperangeofF3V-F6VforthecounterparttoXID3178.ThemagnitudesandcolorsforF2V,F5VandF8Vstarsfrom Carroll&Ostlie ( 1996 )forVmagnitudesand Ducatietal. ( 2001 )for(VH)and(VK)colors,arelistedinTable 5-10 .WeadoptabsoluteKsmagnitudesand(HKs)colorsofMKs=2.83mag,(HKs)=0.047magandMKs=3.10mag,(HKs)=0.037mag,forF3VandF6Vstars,respectively.Usingthe Nishiyamaetal. ( 2009 )and Predehl&Schmitt ( 1995 )extinctionratios,wendextinctionvaluesandcolumndensitiesofAKs0.7andNH=61021cm2fortheF3-F6possibilities.ThedistanceandX-rayluminositiesaretherefored=680860pc,andLX=(0.61.6)1027ergss1,consideringtheNIRphotometricuncertaintyandrangeinpossiblespectraltypes.TheX-rayluminosityiswellwithintherangeexhibitedbyotherelddwarfswithcoronalactivity( Feigelsonetal. 2004 );thusweconcludethatthissourceisacoronallyactiveeldFVstar. 5-16 ),showsaprominentBrabsorptionline,withanequivalentwidthEW=5.70.7AandFWHM=383A.TheBrackettstrengthplacesitinrangeoftheFandGdwarfsasshowninFigure 5-16 .Withthe 155

PAGE 156

5-10 ,weadoptabsoluteKsmagnitudesand(HKs)colorsofMKs=2.92mag,(HKs)=0.043magandMKs=3.27mag,(HKs)=0.03mag,forF4VandF8Vstars,respectively.InthesamemannerasforXID3178,wecalculateadistancerangeof4902.295m.BecauseofthelowsignalattheredendofthetargetspectrumwendthatmainsequenceKstarswithsubtypesofK2-K7areequallyacceptabletsforthisspectrum(seeTable 5-11 ).ThisgivesarangeofAKof0.29-0.46mag,andMKsof4.24-4.79mag.Thecolumndensityisbetween2.6-4.11021cm2.Thedistanceisbetween440and530pc.Theux,givenbyPIMMSfor)-305(=02is5.66.61014ergss1cm2.Fortheassumeddistancerange,thisworksoutto1.32.21030ergss1.CataclysmicvariablessuchasdwarfnovaeorpolarsareplausiblecandidatesforXID7214becauseoftheirhardX-rayspectra,andtypicalluminosityrangeof10301032ergss1( Mukai&Shiokawa 1993 ; Munoetal. 2004 ).ItisnotunusualforaccretioninducedlineemissioninCVstobeabsent( Howelletal. 2010 ).TheX-rayuxfromXID 156

PAGE 157

Munoetal. 2009 ).X-rayvariabilityofthismagnitudeiscommonlyobservedinCVsintheeld.Onthebasisofthelatetypedwarfspectrum,hardX-rayspectrumandX-rayvariability,weconcludethatthissourceisalikelycataclysmicvariable. 157

PAGE 158

Rayneretal. 2009 ),whichcontainsmultiplerepresentativesofmostsub-typesandluminosityclassesofstarslaterthanF0.TheCObandheadatwavelengths>2.295misthemosteffectivefeaturefortypinglatetypestarsintheKbandduetoitsmonotonicallydecreasingdepthwitheffectivetemperature( Blumetal. 2003 ).AsimilareffectwithsmallermagnitudeisseeninthedepthoftheNaandCaabsorptionlinesintheKband( Rayneretal. 2009 ).ThespectraltypesforthemajorityofourhighlyreddenedsourceswerelaterthanG.Toconstraintheluminosityclasses,weconsideredthataG0dwarfwithMK=3.56mag,andwithanapparentmagnitudeofourfaintestcandidate(Ks14)wouldonlybe1200pcaway( Carroll&Ostlie 1996 ; Ducatietal. 2001 ).Ourtargetselectiononlyincludedsourceswith(HKs)>1placingtheirdistancesat4kpcandbeyond( Mauerhanetal. 2009 ).ThereforewewereabletoexcludetheIRTFlibrarydwarfspectrafromourspectralttingefforts.SupergiantsofG-MtypeshaveabsolutemagnitudesofMK=-8magandbrighter.Atadistanceof10kpc(onthefarsideoftheGC),andwithaAKs=3magextinctionthesestarswillhaveKs<10mag.Therefore,weonlyexploretstosupergiantlibraryspectraforourtargetswithKs<10mag.WeobtainedcoverageoftheKbandoutto=2.38mforallOSIRIStargets.Asdiscussedinx5.2.1,theLUCIFER1spectrawavelengthrangevariedwiththepositionontheslitmask.WeprioritizedtheacquisitionoftheBrackettlineat=2.1661m.Unfortunately,asaresult4outof15ofourLUCIFER1targets(thatwerenotalreadydiscussedinx5.3),havenocoverageoftheCOregionatallandwereliedonthelessprominentfeatureslikeNaandCatovisuallytypethesestars.Another5LUCIFER1spectrahadpartialcoverageoftheCObandregionwithwavelengthcutoffsbetween 158

PAGE 159

5-12 welistourbestttemplatesforeachsourcefoundthroughourquantitativespectraltyping.Welistthe(HK)0valuesofthebesttlibraryspectra,takenfromTable11of Rayneretal. ( 2009 ),andcalculatetheKsbandextinctionforeachtarget,usingourISPI(HKs)valuesandextinctionratiosfrom Nishiyamaetal. 159

PAGE 160

2009 ).WeapplytheseextinctionvaluestotheISPIphotometrytondthedereddenedapparentKs,0magnitudeforeachsource.Table 5-13 liststhespectraltyperangeswefoundforsourcesthatrequiredvisualtyping.Duetothelargerangeinpossiblespectraltypesineachcase,wedonotattempttondtheextinctionanddereddenedphotometryforthesesources.Asanillustrationoftherangeinspectraltypesforourresults,weshowspectrafromtheG5IIIcounterparttoXID7204inFigure 5-18 andtheM6IIIcounterpartofXID2212inFigure 5-19 ,bothtakenwithOSIRIS.ThecounterpartstoXID861andXID6055hadveryirregularspectralappearances.WeshowtheirspectrainFigures 5-20 and 5-21 .XID861isabrightK=8.95magsourceandobtainedexcellentsignal,comparabletothatseenforXID2212inFigure 5-19 .Therefore,webelievethatthejaggedshapeofthecontinuumisrealandnotaproductofpoorsignal.TheCOlineshapesaresimilartoM8+IIIspectrafromtheIRTFlibrary.AlsotheXID861spectrumshowsabroaddepressionattheblueendoftheKspectraduetowatervaporabsorption.Thisfeatureisseeninthemajorityoflibrarystarspectrawith>M7III.XID6055doesnothaveashighS/NasXID861,butbycomparisonoftheCOfeaturesandcontinuumshape,itappearssimilarto>M8+IIItypes.Bothofthesesourceswereidentiedaslongperiodvariablesby Matsunagaetal. ( 2009 ).Longperiodpulsationsareseeninallgiantswith>M5III,whichsupportsourchoiceofspectraltype. Munoetal. ( 2009 )topreferentiallyselectsourceswithlowX-rayabsorption,consistentwithadistanceoflessthan4kpc.However,itisnotinfallible,asillustratedbythe Hyodoetal. ( 2008 )observationsoftheGCWolfRayet/X-raysourceCXOGCJ174645.2-281547,whichhasahardnessratioofHR0=1.TheNIRcounterparttoCXOGCJ174645.2-281547hasareddened(HKs)=1.71mag,whichisnormally 160

PAGE 161

Mauerhanetal. 2010a ).TheX-rayspectrumdoeshaveahighenergycomponentwhichappearsatenergiesabove3.3keV,with4.32105photonss1.WesearchedourmatchcatalogforsourceswithsimilarcharacteristicstothisknownGCWRbinary:nominallysoftX-rayspectra,butwithahighhardX-raycountratematchedto(HKs)>1.0magNIRsources.Weselectedtwosuchsourceswithveryhighbvalues,XID6010(b=0.92)andXID6019(b=0.98).ThecountratesforXID6010andXID6019forE>2keVwere0.00011and0.00022photonss1,respectively,meaningtheywouldbedesignatedasbrightsourcesinourstudyevenifjustthehardsourcephotonsareconsidered.NeitherNIRspectrumcontainedanyemissionlines(Figures 5-22 and 5-23 ),suggestingthatthesematchesarenotauthenticcounterpartstotheX-raysources,despitethehighbvalues.BothspectrawereirregularwithrespecttoIRTFlibraryspectra.XID6010lookssomewhatsimilartoXID6055andXID861,withCOshapesandjaggedcontinuumcharacteristicofverylatetypeMIIIstars.However,wedonotseeastrongdepressionfromwatervaporat<2.1m,whichisinconsistentwiththeM8+IIIlibraryspectra.ThecounterparttoXID6019isnearlyfeatureless.ThepositionsofthedeepestCObandsat<2.295mmayshowsubtleevidenceofabsorption.EarlyFIIItypestarshavesimilarlyweakCO,butalsohaveadeepabsorptionBrlinethatwouldcertainlybeseeninourXID6019spectrum. Matsunagaetal. ( 2009 )reportthatXID6019isaLPV.IfthespectrumwastrulythatofanF/GIII,itwouldbesimilartothetypeIICepheidswhichhaveperiodsshorterthanafewdaysanditwouldnotbeclassiedasanLPVby Matsunagaetal. ( 2009 ).Thiscouldbeadustenshroudedpost-AGBsource,suchasseenbyKbandobservationsby Oudmaijeretal. ( 1995 ). 161

PAGE 162

5-24 ).Group2:hardX-raysourceswithJHKdetections,(HKs)>1.0mag,X<100.0andanX-rayphotonuxoffx<0.0001countss1,and13
PAGE 163

6910=0.380or22.4likelytruematcheswithintheremaining59candidates.The1errorinthisvalueis0.0959=5.3.Ifweconsiderjustthe15GroupIcandidatesobservedwithOSIRISandLUCIFER,weseethattherearelikelytobe7.61.8truecounterpartsthatdidnotexhibitclearemissionlinesignatures.IfXID13isincluded,theremainingcandidateswithoutemissionlinesshouldcontain7.01.9truematches.Intherstcasethesignicanceis4.2andinthesecondthesignicancefallsto3.7.WelooktothebparametertoseewhetheroursourceselectionwithinGroup1mightbebiasedagainsttruecounterparts.InSectionx5.1wedemonstratedthatmedianbvalueofagroupofsourceswhichliescloseto1.0demonstratesahighlevelofastrometriccoincidence,andsuggestthatNIRandX-raysourcesarerelated.ThemedianvalueofbforallGroupIsourcesis0.52.Themedianbforthe15sourceswesearchedwithOSIRIS/LUCIFER1is0.54.Thesevaluesaresimilar,suggestingthatoursourceselectionwasnotskewedagainsttruecounterparts.TheGroupIspectraltypes 163

PAGE 164

164

PAGE 165

Rayneretal. ( 2009 )listtheintrinsic(HK)0colorsforallthespectraintheirlibrary.Weadoptthe(HK)0valueofthebestttemplateforeachofourGroup1sourcestocalculatetheAKsandAVbandextinction,usingaAV:AH:AKs=16.1:1.73:1extinctionratio( Nishiyamaetal. 2009 ).Weconvertedthesevaluestothehydrogencolumndensity,NH,usingtheratiofrom Predehl&Schmitt ( 1995 ).Then,usingtheChandraproposaltoolPIMMS,wecalculatetheunabsorbedX-rayuxfrom2-8keV,givenourvalueofNH,theapparentX-ray 165

PAGE 166

5-15 .WedonotincludeXID13,becausewewereunabletodetermineaspectraltypefromourpreliminarydatareduction.WenotetheextinctionforXID1330,XID6023,andXID2663issignicantlylowerthantheaveragevalueofAV=35magfortherestofthesources,implyingthatthesemaylieatcloserdistances. Mauerhanetal. ( 2009 )calculateanupperlimitofAV=22.3magforadistanceof4kpctowardtheGC.Therefore,thevaluesofLXandMKsmaybe4lowerthanthevaluecalculatedinTable 5-15 forthesethreesources.ThebreakdowninspectraltypesforGroup1is:8starswithG7-K3III,1M1I,and6M1-7IIIstars.OnepossibilityisthatsomeofthesesourcescouldbequiescentLMXBtransients.RGBLMXBssuchasV404Cygni,theblackholebinary,areknowntoexperiencelongdurationquiescentperiodswithaninverseComptonX-rayspectraof)-328(=12andX-rayluminositiesof1032ergss( Asaietal. 1998 ).Inquiescence,thedonorstarspectrumofV404CygniappearstobeacompletelynormalKIIIstar,devoidofemissionfeatures( Shahbazetal. 1996 ; Kharghariaetal. 2010 ).Anotherpossibilityisthatsomeofthesesourcescouldbesymbioticbinaries.WediscusstwotypesofsymbioticstarswithhardX-rayspectrainx5.3.2inregardstotheidentityofXID6592.ThehardspectrumsymbioticswithmassivewhitedwarfshaveX-rayluminositiesrangingfrom1032to-1034ergss1( Luna&Sokoloski 2007 ; Smithetal. 2008 ).TheirdonorstarshaveearlytomidMspectraltypes.OpticalspectrafromthesesourcestendtoshowBalmerlineemission,whichmaymeanthatlowionizationlineslikeBrandHeIwouldbepresentintheinfrared( Cieslinskietal. 1994 ; Mukaietal. 2003 ; Smithetal. 2008 ).ItisunclearwhetherthesesourcesexperiencelowlevelactivityphaseswheretheNIRemissionlineswouldbeunseen.Inthatcase,hardX-rayWDsymbioticswouldbeunlikelycandidatesforthehiddenGroup1X-raysourcecounterparts. 166

PAGE 167

Chakrabartyetal. ( 1998 ).However,fourNSsymbioticsystemshavebeenobservedwithlittleornoemission. Masettietal. ( 2006 )tookopticalspectraofthecounterpartsoftwoNSsymbioticbinaries,4U1954+31and4U1954+319,duringaperiodofrelativelylowX-rayluminosity,LX1032ergss1.Theyfoundtheopticalspectrawerecompletelydevoidofemissionlines,withthecontinuumdominatedbytheM2-5IIIsecondaries.WithalackofBalmerlines,itisunclearwhetherthissourcewouldhavedetectableBrackettemissionintheNIR. Nespolietal. ( 2010 )tookKbandspectraoftheneutronstarsymbioticsIGR16393-4643andIGR16358-4726andfoundverytenuousdetectionsofBrackettandHeI2.0581.Therefore,NSsymbioticbinariesdoseemtohavephaseswithlowemissionlineux,andassuchareplausiblecandidatesforthehiddentruecounterpartswithinGroup1. vandenBergetal. ( 2006 )searchedforsymbioticstarsinBaade'swindowandStanek'swindowusingX-raypositionalmatchestoMgiantsknownfromtheOGLEcatalogs.Theyfound13candidatesmeetingtheircriteriafor>M0IIIstarsmatchedtopredominatelyhardX-raysourceswith1031
PAGE 168

168

PAGE 169

TableofspectroscopicobservationswiththeLargeBinocularTelescopeNearInfraredSpectroscopicUtilitywithCameraandIntegralFieldUnitforExtragalacticResearch1(LUCIFER1)oftheNIRcounterpartsto20sources ,(K)Conditions CXOUGCJ174543.4-285841XID25535/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174544.4-285829XID26425/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174544.6-285806XID26635/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174546.9-285903XID28705/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174547.2-285816XID28905/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174547.7-285719XID29315/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174549.6-285704XID30775/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174551.2-285838XID31785/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174552.0-285507XID32185/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174552.9-285537XID32755/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174554.0-285432XID33345/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174554.4-285455XID33605/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174554.8-285650XID33905/17/2010LBT/LUCIFER25002.07100.5thincirrusCXOUGCJ174557.3-285353XID34965/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174558.1-285303XID72665/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174558.6-285511XID35475/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174601.7-285238XID73725/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174604.8-285439XID74405/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174606.3-285313XID74745/17/2010LBT/LUCIFER25002.06100.5thincirrusCXOUGCJ174607.6-285351XID74975/17/2010LBT/LUCIFER25002.06100.5thincirrus

PAGE 170

TableofspectroscopicobservationswiththeOhioStateInfraRedImager/Spectrometer(OSIRIS)oftheNIRcounterpartsto30sources ,(K)Conditions CXOUGCJ174533.5-285540XID14448/19/2010SOAR/OSIRIS12001.30000.5clearCXOUGCJ174459.5-285108XID60238/19/2010SOAR/OSIRIS12001.08000.5clearCXOUGCJ174500.8-285121XID60558/19/2010SOAR/OSIRIS12001.16000.5clearCXOUGCJ174559.7-290112XID35938/19/2010SOAR/OSIRIS12001.03000.5clearCXOUGCJ174540.1-290055XID22128/19/2010SOAR/OSIRIS12001.30000.5clearCXOUGCJ174533.7-285728XID14708/19/2010SOAR/OSIRIS12001.84000.5clearCXOUGCJ174536.4-290227XID17838/19/2010SOAR/OSIRIS12001.04000.5clearCXOUGCJ174530.3-290341XID11008/19/2010SOAR/OSIRIS12001.03000.5clearCXOUGCJ174529.0-290406XID9788/19/2010SOAR/OSIRIS12001.03000.5clearCXOUGCJ174533.5-290759XID66878/19/2010SOAR/OSIRIS12001.01000.5clearCXOUGCJ174556.0-285056XID72048/19/2010SOAR/OSIRIS12001.01000.5clearCXOUGCJ174607.6-285351XID74978/19/2010SOAR/OSIRIS12001.06000.5clearCXOUGCJ174605.5-285319XID37688/19/2010SOAR/OSIRIS12001.06000.5clearCXOUGCJ174510.1-290515XID62628/19/2010SOAR/OSIRIS12001.12000.5clearCXOUGCJ174507.4-290456XID578/19/2010SOAR/OSIRIS12001.12000.5clearCXOUGCJ174529.6-290227XID10308/19/2010SOAR/OSIRIS12001.36000.5clearCXOUGCJ174527.8-290109XID8618/19/2010SOAR/OSIRIS12001.36000.5clearCXOUGCJ174550.5-290201XID31458/19/2010SOAR/OSIRIS12001.47000.5clearCXOUGCJ174531.6-290048XID12298/19/2010SOAR/OSIRIS12001.64000.5clearCXOUGCJ174532.4-290126XID13308/19/2010SOAR/OSIRIS12001.64000.5clearCXOUGCJ174528.0-290023XID8858/19/2010SOAR/OSIRIS12002.02000.5clearCXOUGCJ174528.5-285959XID9308/19/2010SOAR/OSIRIS12002.02000.5clearCXOUGCJ174528.6-285605XID9358/19/2010SOAR/OSIRIS12001.30000.5clearCXOUGCJ174528.8-285726XID9478/19/2010SOAR/OSIRIS12001.84000.5clearCXOUGCJ174537.9-290134XID19448/19/2010SOAR/OSIRIS12001.04000.5clearCXOUGCJ174528.7-290942XID65928/19/2010SOAR/OSIRIS12001.01000.5clearCXOUGCJ174556.3-285054XID72148/19/2010SOAR/OSIRIS12001.01000.5clearCXOUGCJ174459.0-285123XID60108/19/2010SOAR/OSIRIS12001.38000.5clearCXOUGCJ174459.4-285140XID60198/19/2010SOAR/OSIRIS12001.38000.5clearCXOUGCJ174500.0-285132XID60348/19/2010SOAR/OSIRIS12001.16000.5clear

PAGE 171

TableofspectroscopicobservationswiththeFolded-portInfraRedEchellette(FIRE)oftheNIRcounterpartsto12newsourcesand3repeatedsourcesfromOSIRISandLUCIFER1observations. ,(K)Conditions CXOUGCJ174528.1-290736XID9028/25/2010MAGELLAN/FIRE45001.01000.8clearCXOUGCJ174457.4-290613XID59738/25/2010MAGELLAN/FIRE45001.50000.8clearCXOUGCJ174527.1-290235XID8088/25/2010MAGELLAN/FIRE45001.72000.8clearCXOUGCJ174514.5-290225XID2048/25/2010MAGELLAN/FIRE45001.86000.8clearCXOUGCJ174549.1-290137XID30398/25/2010MAGELLAN/FIRE45001.05000.8clearCXOUGCJ174523.1-285946XID5368/25/2010MAGELLAN/FIRE45001.57000.8clearCXOUGCJ174502.2-285749XID138/25/2010MAGELLAN/FIRE45001.01000.8clearCXOUGCJ174602.8-285624XID36878/25/2010MAGELLAN/FIRE45001.18000.8clearCXOUGCJ174552.9-285537XID32758/25/2010MAGELLAN/FIRE45001.28000.8clearCXOUGCJ174554.6-285404XID33778/25/2010MAGELLAN/FIRE45001.36000.8clearCXOUGCJ174500.0-285252XID60358/25/2010MAGELLAN/FIRE45001.06000.8clearCXOUGCJ174459.4-285140XID60198/25/2010MAGELLAN/FIRE45001.70000.8clearCXOUGCJ174457.8-285125XID59808/25/2010MAGELLAN/FIRE45001.11000.8clearCXOUGCJ174459.0-285123XID60108/25/2010MAGELLAN/FIRE45001.03000.8clearCXOUGCJ174457.1-285107XID59638/25/2010MAGELLAN/FIRE45001.08000.8clear

PAGE 172

SpectraofLBV1806-10takenonsuccessivenights.Inthesecondnightthespectrumisafictedbyaregularlyspacedpittinginthecontinuum.ThelowestlineshowsourGaussiantstotheratioofthetwoLBVspectra.Weusedthispatterntocorrectpittingissuesintherestofourtargetspectra. Table5-4. NearinfraredphotometryofthecounterpartstoX-raysources,takenwiththeISPIinstrument( DeWittetal. 2010 ). X-ray1RADECJJHHKsKsSourceNumber(J2000.0)(J2000.0)(mag)(mag)(mag)(mag)(mag)(mag) 6592266.369970-29.161879--13.030.059.880.05947266.370329-28.95733616.230.0513.290.0511.610.051944266.408300-29.02624110.780.059.540.058.760.053275266.470695-28.92695117.690.0514.750.0513.080.053178266.463361-28.97756614.270.0513.540.0513.060.053390266.478267-28.94700112.150.0511.990.0511.870.057214266.484811-28.84833215.080.0513.760.0513.310.05 172

PAGE 173

Munoetal. ( 2009 ).) X-rayRADECerrorHR0Typelog(FX)SourceNumber(J2000.0)(J2000.0)(")(countss1) 6592266.36997-29.161880.60.716hard-2.673947266.37033-28.957340.51.0hard-4.1101944266.40830-29.026240.3-0.168hard-3.5413275266.47070-28.926950.6-9.0hard-4.0023178266.46336-28.977570.50.083hard-4.3073390266.47839-28.947160.6-0.293soft-4.2587214266.48480-28.848340.4-0.012hard-2.818 173

PAGE 174

B CFigure5-2. Munoetal. ( 2009 )isshowninred.(a)Jband(b)Hband(c)Ksband. 174

PAGE 175

Figure5-4. 175

PAGE 176

LineidenticationsandparametersforthecounterpartspectrumofXID3275. LineIDrestobservedWEQFWHMwavelength(m)wavelength(m)(A)(A) Br2.166102.16561-11.80.8203HeI2.058702.05844-3.81.0233Br1.945101.94454-21.31.5293MgII2.13800-MgII2.14400-H10-41.945101.94454-21.31.5293H11-41.736701.73611-18.71.5353H12-41.681101.68053-13.81.5323H13-41.641201.64074-8.01.5213H14-41.611401.61056-7.01.5213H15-41.588501.58822-6.81.5233H16-41.570501.57023-5.51.5243 176

PAGE 177

TheHKscolorofthecounterparttoXID3275(dashedline),comparedtothecolorsofallNIRsourceswithin20"ofitsposition. B CFigure5-6. Munoetal. ( 2009 )isshowningreen.(a)Jband(b)Hband(c)Ksband. 177

PAGE 178

LineidenticationsandparametersforthecounterpartspectrumofXID6592. LineIDrestobservedEWFWHMwavelength(m)wavelength(m)(A)(A) Br2.166102.164741.8+1.00.2173Na2.2062/2.20902.20530/2.20810-Ca2.2631/2.26672.26185/2.26545Figure5-7. 178

PAGE 179

BFigure5-8. Matsunagaetal. 2009 )plottedwithdiamonds.ThesingleISPIandUKIDSSmeasurementsareoverplottedwithatriangleandsquare,respectively.KwassaturatedforthissourceforUKIDSS. Table5-8. Sp.Type2(D.O.F.=149)reduced2templateID K2III7913.553.1HD132935K6III5229.535.1HD3346M1III3579.224.0HD204724M4III2764.418.6HD214665M6III1119.17.5HD196610M7III721.34.8HD108849M5e-M9eIII1129.57.6HD14386(Mira)M9III1069.77.2IRAS15060+0947M0.5Ib1242.28.3HD236697M1-2Ia-Iab695.64.7HD39801M3-4Iab718.34.8HD14469M5Ib-II2061.613.8HD156014 179

PAGE 180

TheHKscolorofthecounterparttoXID6592,comparedtothecolorsofallNIRsourceswithin20"ofthetargetposition. B CFigure5-10. Munoetal. ( 2009 )isshowningreen.(a)Jband(b)Hband(c)Ksband. 180

PAGE 181

Table5-9. LineidenticationsandparametersforthecounterpartspectrumofXID1944. LineIDrestobservedWEQFWHMwavelength(A)wavelength(A)(A)(A) HeI2.05872.06123-1.20.3303CIV2.0700---CIV2.07502.07854-1.10.2133HeI/NIII2.1200-2.11473-3.60.4393CIII/OIII-2.1350HeII2.189002.188461.30.4323 181

PAGE 182

B CFigure5-12. Munoetal. ( 2009 )isshowningreen.(a)Jband(b)Hband(c)Ksband. 182

PAGE 183

TheKbandspectrumofthecounterparttoXID947.AlsoplottedareanO6Vspectrumfrom Hansonetal. ( 2005 )andanO7Vfrom Wallace&Hinkle ( 1997 ). Figure5-14. TheHKscolorofthecounterparttoXID947,comparedtothecolorsofallNIRsourceswithin20"ofthetargetposition. 183

PAGE 184

ColorsandmagnitudesforFdwarfstars( Carroll&Ostlie 1996 ; Ducatietal. 2001 ) SpectralTypeMK(mag)(JK)(mag)(HK)(mag) F2V2.740.18-0.05F5V3.010.22-0.04F8V3.270.24-0.03 Figure5-15. BrackettequivalentwidthsmeasuredforGandFdwarfs( Rayneretal. 2009 ).TheequivalentwidthsofBrackettinthecounterpartspectraofXID3178andXID3390areplottedinred. 184

PAGE 185

Rayneretal. 2009 ). 185

PAGE 186

Rayneretal. 2009 ). Table5-11. Sp.Type2(D.O.F.=149)reduced2templateID K0V1099.84.3HD145675K1V1052.74.1HD10476K2V904.93.5HD3765K3V990.33.9HD219134K4V922.73.6HD45977K5V930.03.6HD36003K7V949.63.7HD201092M0V1079.64.2HD19305 186

PAGE 187

Figure5-19. 187

PAGE 188

ResultsofspectralttingofNIRcounterpartstoX-raysources. X-rayBestSpec.2/Spec.Type(HK)0AKKs,0SourceNumberTypeRange(mag)(mag)(mag) 1444M3III2.4M1-M4III0.2672.9838.966023G7III3.5G7-G8III0.1072.6908.573593M7III1.4M2.5-7III0.4142.0228.082212M6III2.0M6-7III0.2843.1226.831470G7.5III5.9G3-7.5III0.1302.8509.611783M6III1.3M4.5-7III0.2843.4046.061100M6III1.6M4.5-7III0.2842.5278.94978M1I3.1M1-3.5I0.3032.2386.286687M6III1.9M6-7III0.2843.5327.387204G5III1.9G3-7.5III0.1241.7498.597497(OSIRIS)M4III1.4M1-7III0.2971.9059.847497(LBT)M4III2.1M1-4III0.2971.9059.843768M6III1.3M4.5-7III0.2842.3377.146262M4III1.2M2.5-4III0.2972.21410.0257M5III2.1M4-5.5III0.2292.34910.171030M6III2.4M2.5-7III0.2842.6867.923145M5III2.2M4-5.5III0.2292.4107.461229M2III1.9K5.5-M2.5III0.2712.3429.451330G7III3.3G3-G9III0.1071.05811.47885K1III2.4K1-K2III0.1341.34510.13930M4III2.0M3.5-4III0.2972.4467.453547K1III2.8G9-K2III0.0971.96411.937440M4.5III2.1M2.5-5.5III0.2342.3169.167474M4III0.7M1-4III0.2971.7639.062890M0III2.1M0-3.5III0.2002.38210.642870K1III1.5G9-K1III0.0971.9229.80 188

PAGE 189

VisualspectraltypingofNIRcounterpartstoX-raysources. X-raySourceNumberSpectralTypeRangeReasonforvisualtyping 6055M8+IIIverylatetype861M8+IIIverylatetype3218M1-M7IIINoCO3360K0.5-K3.5IIIpart.CO3334G0-M3IIINoCO3496M4-M7IIIpart.CO7266M4-M7IIINoCO2553K2-k3.5IIIpart.CO2642M4.5-M7IIIpart.CO2663K1.5-3IIIpart.CO2931K3-M5IIIpart.CO3077M1-M7IIIpart.CO 189

PAGE 190

Figure5-21. 190

PAGE 191

Table5-14. ProposedclassicationfromvisualinspectionforreddenedNIRcounterpartstosoftX-raysources. X-rayBestSpec.SourceNumberType 6010M8+III?6019Embeddedpost-AGB? 191

PAGE 192

Figure5-24. PreliminaryNIRspectrumforthecounterparttoXID13,observedwithFIRE. 192

PAGE 193

SpectraltypesandestimatedX-rayparametersforsourcesobservedfromGroup1. 133012.53G7III0.8460.1071.2820.63.70.000181.316E-141.07e+32602311.26G7III0.8750.1071.3121.13.80.000675.128E-144.20e+32147012.46G7.5III0.5640.1302.6943.47.80.000151.827E-141.48e+32287011.73K1III0.1340.0971.9231.05.50.000353.432E-142.80e+32354713.90K1III0.2780.0971.9631.65.60.000131.321E-141.07e+32255313.77K2III0.4580.1402.0235.66.40.000758.218E-146.76e+32336012.96K3III0.1250.1442.4639.77.10.008631.014E-128.33e+33266311.95K3III0.9580.1441.1919.23.40.000128.206E-156.76e+319788.52M1I0.5720.3032.2436.16.50.000363.931E-143.21e+32747411.37M4III0.4140.2971.7628.45.10.000141.281E-141.07e+32744011.47M4.5III0.4990.2342.3237.46.70.000212.317E-141.90e+3222129.52M6III0.9190.3193.0749.58.90.000881.223E-131.01e+33110011.46M6III0.2800.3192.4840.07.20.000323.770E-143.05e+3237689.48M6III0.7320.3192.2936.96.60.000202.220E-141.81e+32349610.79M7III0.2580.4141.9231.05.50.000121.152E-149.89e+31

PAGE 194

194

PAGE 195

195

PAGE 196

196

PAGE 197

197

PAGE 198

198

PAGE 199

Eikenberry 2008 ).Thesurveyhasthecomplementarygoalofprobingthephysicsbehindtheblackholemass/bulgedispersionrelationshipbymeasuringthestarformationhistorywiththestarsthatweretargetedaspossibleX-raycounterpartsbutfoundtobeunrelated.4000otherprobableRGBspectrawillbetakentosupplementthesesources.WewillrevisittheVISTAearlysciencedataandextendourvariabilitystudytondnewvariablecandidateX-raycounterparts.Itislikelythatwecanincreasethedynamicrangeofvariableswecandetectbyspecializedttingofthewingsofsourceswithsaturatedcores,makingthebrightmagnitudesusable.Wewillalsoutilizeoptimalimagesubtractiontechniqueswhichhavebeenextensivelydevelopedforuseinsupernovaesearchestowarddistancecrowdedelds( Alard&Lupton 1998 ).Optimalimagesubtractionalleviatestheneedfordetectingandttingasourceineachepoch,whichisespeciallyimportantfortheblendedsourcesinoureldorsourceswithhighdynamicrangeintheirvariabilityamplitude.WewillalsoparticipateintheVISTAVVVdedicatedsurveyoftheGalacticbulge.ThissurveyperformsdedicatedNIRmonitoringoftheGalacticCentertondvariablestarswhichcanbeusedasdistancemeasures Minnitietal. ( 2010 ).ThesurveyisalsoidealfordiscoveringtransientNIRvariabilityfromX-raysources.WewillalsoattempttohavetheSwiftXRTsatellitemonitortheregionoversomeoralloftheNIRsurveyduration.WewillperformX-rayspectralttingonourBeHMXBcandidateandourhardX-raysymbioticcandidate,withexistingdataandpossiblynewobservationsifnecessary.WeidentiedcanonicalNSBeHMXBsandCassystemsaspossiblesourceclassesforXID3275.InhardX-raysthesetwosourceclassescanbedifferentiatedbyspectrumtypes;powerlawhighenergytailsincanonicalNSBeHMXBsand12keVthermalplasmasforCassystems. 199

PAGE 200

200

PAGE 201

Alard,C.,&Lupton,R.H.1998,ApJ,503,325 Arendt,R.G.,etal.2008,ApJ,685,958 Arnaboldi,M.,etal.2010,TheMessenger,139,6 Asai,K.,Dotani,T.,Hoshi,R.,Tanaka,Y.,Robinson,C.R.,&Terada,K.1998,PASJ,50,611 Baganoff,F.K.,etal.2003,ApJ,591,891 Balick,B.,&Brown,R.L.1974,ApJ,194,265 Bandyopadhyay,R.M.,Shahbaz,T.,Charles,P.A.,&Naylor,T.1999,MNRAS,306,417 Bandyopadhyay,R.M.,etal.2005,MNRAS,364,1195 Bartko,H.,etal.2010,ApJ,708,834 Belczynski,K.,Mikoajewska,J.,Munari,U.,Ivison,R.J.,&Friedjung,M.2000,A&AS,146,407 Binney,J.,Gerhard,O.E.,Stark,A.A.,Bally,J.,&Uchida,K.I.1991,MNRAS,252,210 Blum,R.D.,Ramrez,S.V.,Sellgren,K.,&Olsen,K.2003,ApJ,597,323 Campana,S.,Gastaldello,F.,Stella,L.,Israel,G.L.,Colpi,M.,Pizzolato,F.,Orlandini,M.,&DalFiume,D.2001,ApJ,561,924 Carlstrom,J.E.,Holder,G.P.,&Reese,E.D.2002,ARA&A,40,643 Carroll,B.W.,&Ostlie,D.A.1996,AnIntroductiontoModernAstrophysics,ed.Carroll,B.W.&Ostlie,D.A. Castelli,F.,&Kurucz,R.L.1994,A&A,281,817 Chakrabarty,D.,&Roche,P.1997,ApJ,489,254 Chakrabarty,D.,vanKerkwijk,M.H.,&Larkin,J.E.1998,ApJL,497,L39+ Cieslinski,D.,Elizalde,F.,&Steiner,J.E.1994,A&AS,106,243 Clark,J.S.,&Steele,I.A.2000,A&AS,141,65 Coker,R.F.,Pittard,J.M.,&Kastner,J.H.2002,A&A,383,568 Cox,A.N.2000,Allen'sastrophysicalquantities,ed.Cox,A.N. 201

PAGE 202

Davies,R.I.2007,MNRAS,375,1099 DeWitt,C.,Bandyopadhyay,R.M.,Eikenberry,S.S.,Blum,R.,Olsen,K.,Sellgren,K.,&Sarajedini,A.2010,ApJ,721,1663 Draine,B.T.,&McKee,C.F.1993,ARA&A,31,373 Ducati,J.R.,Bevilacqua,C.M.,Rembold,S.B.,&Ribeiro,D.2001,ApJ,558,309 Dupree,A.K.1986,ARA&A,24,377 Eikenberry,S.S.2008,inAmericanInstituteofPhysicsConferenceSeries,Vol.1010,APopulationExplosion:TheNature&EvolutionofX-rayBinariesinDiverseEnvironments,ed.R.M.Bandyopadhyay,S.Wachter,D.Gelino,&C.R.Gelino,132 Eikenberry,S.S.,Patel,S.G.,Rothstein,D.M.,Remillard,R.,Pooley,G.G.,&Morgan,E.H.2008,ApJ,678,369 Emerson,J.,McPherson,A.,&Sutherland,W.2006,TheMessenger,126,41 Eze,R.N.C.2011,AdvancesinSpaceResearch,47,1999 Feigelson,E.D.,etal.2004,ApJ,611,1107 Fender,R.P.,Pooley,G.G.,Brocksopp,C.,&Newell,S.J.1997,MNRAS,290,L65 Gebhardt,K.,etal.2000,ApJL,539,L13 Ghez,A.M.,Salim,S.,Hornstein,S.D.,Tanner,A.,Lu,J.R.,Morris,M.,Becklin,E.E.,&Duchene,G.2005,ApJ,620,744 Gilfanov,M.2004,MNRAS,349,146 Gillessen,S.,Eisenhauer,F.,Trippe,S.,Alexander,T.,Genzel,R.,Martins,F.,&Ott,T.2009,ApJ,692,1075 Gosling,A.J.,Bandyopadhyay,R.M.,&Blundell,K.M.2009,MNRAS,394,2247 Greiner,J.,Cuby,J.G.,McCaughrean,M.J.,Castro-Tirado,A.J.,&Mennickent,R.E.2001,A&A,373,L37 Grimm,H.,Gilfanov,M.,&Sunyaev,R.2003,MNRAS,339,793 Groenewegen,M.A.T.,Udalski,A.,&Bono,G.2008,A&A,481,441 Hanson,M.M.,Kudritzki,R.,Kenworthy,M.A.,Puls,J.,&Tokunaga,A.T.2005,ApJS,161,154 202

PAGE 203

Hodgkin,S.T.,Irwin,M.J.,Hewett,P.C.,&Warren,S.J.2009,MNRAS,394,675 Howell,S.B.,Harrison,T.E.,Szkody,P.,&Silvestri,N.M.2010,AJ,139,1771 Howell,S.B.,Nelson,L.A.,&Rappaport,S.2001,ApJ,550,897 Hyodo,Y.,Tsujimoto,M.,Koyama,K.,Nishiyama,S.,Nagata,T.,Sakon,I.,Murakami,H.,&Matsumoto,H.2008,PASJ,60,173 Irwin,M.J.,etal.2004,inPresentedattheSocietyofPhoto-OpticalInstrumentationEngineers(SPIE)Conference,Vol.5493,SocietyofPhoto-OpticalInstrumentationEngineers(SPIE)ConferenceSeries,ed.P.J.Quinn&A.Bridger,411 Ishibashi,W.,&Courvoisier,T.J.-L.2010,A&A,512,A58+ Ita,Y.,etal.2002,MNRAS,337,L31 Jonker,P.G.,&Nelemans,G.2004,MNRAS,354,355 Judge,P.G.,Solomon,S.C.,&Ayres,T.R.2003,ApJ,593,534 Kenyon,S.J.1986,Thesymbioticstars,ed.Kenyon,S.J. Khargharia,J.,Froning,C.S.,&Robinson,E.L.2010,ApJ,716,1105 Knigge,C.2006,MNRAS,373,484 Krabbe,A.,Genzel,R.,Drapatz,S.,&Rotaciuc,V.1991,ApJL,382,L19 Lawrence,A.,etal.2007,MNRAS,379,1599 Laycock,S.,Corbet,R.H.D.,Coe,M.J.,Marshall,F.E.,Markwardt,C.,&Lochner,J.2005,ApJS,161,96 Leahy,D.A.,&Ananth,A.G.1992,MNRAS,256,39P Liu,Q.Z.,vanParadijs,J.,&vandenHeuvel,E.P.J.2006,A&A,455,1165 LopesdeOliveira,R.,Smith,M.A.,&Motch,C.2010,A&A,512,A22+ Lord,S.1992,ANewSoftwareToolforComputingEarth'sAtmosphericTransmissionofNear-andFar-InfraredRadiation,NASATechnicalMemorandum103957 Lu,G.-L.,Zhu,C.-H.,&Han,Z.-W.2007,ChineseJournalofAstronomyandAstrophysics,7,101 Lucas,P.W.,etal.2008,MNRAS,391,136 Luna,G.J.M.,Sokoloski,J.,Mukai,K.,&Nelson,T.2010,TheAstronomer'sTelegram,3053,1 203

PAGE 204

Luna,G.J.M.,Sokoloski,J.L.,&Mukai,K.2008,inAstronomicalSocietyofthePacicConferenceSeries,Vol.401,RSOphiuchi(2006)andtheRecurrentNovaPhenomenon,ed.A.Evans,M.F.Bode,T.J.O'Brien,&M.J.Darnley,342+ Maness,H.,etal.2007,ApJ,669,1024 Martins,F.,Hillier,D.J.,Paumard,T.,Eisenhauer,F.,Ott,T.,&Genzel,R.2008,A&A,478,219 Martins,F.,&Plez,B.2006,A&A,457,637 Masetti,N.,Orlandini,M.,Palazzi,E.,Amati,L.,&Frontera,F.2006,A&A,453,295 Masetti,N.,etal.2002,A&A,382,104 Matsunaga,N.,Kawadu,T.,Nishiyama,S.,Nagayama,T.,Hatano,H.,Tamura,M.,Glass,I.S.,&Nagata,T.2009,MNRAS,399,1709 Mauerhan,J.C.,Cotera,A.,Dong,H.,Morris,M.R.,Wang,Q.D.,Stolovy,S.R.,&Lang,C.2010a,ApJ,725,188 Mauerhan,J.C.,Muno,M.P.,&Morris,M.2007,ApJ,662,574 Mauerhan,J.C.,Muno,M.P.,Morris,M.R.,Bauer,F.E.,Nishiyama,S.,&Nagata,T.2009,ApJ,703,30 Mauerhan,J.C.,Muno,M.P.,Morris,M.R.,Stolovy,S.R.,&Cotera,A.2010b,ApJ,710,706 Meurs,E.J.A.,&vandenHeuvel,E.P.J.1989,A&A,226,88 Mignani,R.2009,TheMessenger,138,19 Mikles,V.J.,Eikenberry,S.S.,Muno,M.P.,Bandyopadhyay,R.M.,&Patel,S.2006,ApJ,651,408 Miller,C.J.2007,inAstronomicalSocietyofthePacicConferenceSeries,Vol.382,AstronomicalSocietyofthePacicConferenceSeries,ed.M.J.Graham,M.J.Fitzpatrick,&T.A.McGlynn,19+ Minniti,D.,etal.2010,notavailable,15,433 Morris,M.,&Serabyn,E.1996,ARA&A,34,645 Muerset,U.,Wolff,B.,&Jordan,S.1997,A&A,319,201 Mukai,K.,Kinkhabwala,A.,Peterson,J.R.,Kahn,S.M.,&Paerels,F.2003,ApJL,586,L77 204

PAGE 205

Muno,M.P.,Baganoff,F.K.,Bautz,M.W.,Brandt,W.N.,Garmire,G.P.,&Ricker,G.R.2003a,ApJ,599,465 Muno,M.P.,Lu,J.R.,Baganoff,F.K.,Brandt,W.N.,Garmire,G.P.,Ghez,A.M.,Hornstein,S.D.,&Morris,M.R.2005,ApJ,633,228 Muno,M.P.,etal.2003b,ApJ,589,225 .2004,ApJ,613,1179 .2009,ApJS,181,110 Murset,U.,&Schmid,H.M.1999,A&AS,137,473 Negueruela,I.,Okazaki,A.T.,Fabregat,J.,Coe,M.J.,Munari,U.,&Tomov,T.2001,A&A,369,117 Neil,E.T.,Bailyn,C.D.,&Cobb,B.E.2007,ApJ,657,409 Nespoli,E.,Fabregat,J.,&Mennickent,R.E.2010,A&A,516,A94+ Nishiyama,S.,Nagata,T.,Tamura,M.,Kandori,R.,Hatano,H.,Sato,S.,&Sugitani,K.2008,ApJ,680,1174 Nishiyama,S.,Tamura,M.,Hatano,H.,Kato,D.,Tanabe,T.,Sugitani,K.,&Nagata,T.2009,ApJ,696,1407 Olsen,K.A.G.,Blum,R.D.,&Rigaut,F.2003,AJ,126,452 Oskinova,L.M.2005,MNRAS,361,679 Oskinova,L.M.,Hamann,W.-R.,Feldmeier,A.,Ignace,R.,&Chu,Y.-H.2009,ApJL,693,L44 Oudmaijer,R.D.,Waters,L.B.F.M.,vanderVeen,W.E.C.J.,&Geballe,T.R.1995,A&A,299,69 Paczynski,B.1971,ARA&A,9,183 Paul,B.,Dotani,T.,Nagase,F.,Mukherjee,U.,&Naik,S.2005,ApJ,627,915 Pellizza,L.J.,Chaty,S.,&Negueruela,I.2006,A&A,455,653 Pfahl,E.,Rappaport,S.,&Podsiadlowski,P.2002,ApJL,571,L37 Predehl,P.,&Schmitt,J.H.M.M.1995,A&A,293,889 Pringle,J.E.1981,ARA&A,19,137 Provencal,J.L.,Shipman,H.L.,Hog,E.,&Thejll,P.1998,ApJ,494,759 205

PAGE 206

Rayner,J.T.,Cushing,M.C.,&Vacca,W.D.2009,ApJS,185,289 Reig,P.2011,APSS,332,1 Remillard,R.A.,&McClintock,J.E.2006,ARA&A,44,49 Revnivtsev,M.,Sazonov,S.,Churazov,E.,Forman,W.,Vikhlinin,A.,&Sunyaev,R.2009,Nature,458,1142 Ritter,H.,&Kolb,U.2003,A&A,404,301 Schodel,R.,Najarro,F.,Muzic,K.,&Eckart,A.2010,A&A,511,A18+ Seward,F.D.,&Charles,P.A.1995,ExploringtheX-RayUniverse,ed.Seward,F.D.&Charles,P.A. Shahbaz,T.,Bandyopadhyay,R.,Charles,P.A.,&Naylor,T.1996,MNRAS,282,977 Shapiro,S.L.,&Teukolsky,S.A.1986,BlackHoles,WhiteDwarfsandNeutronStars:ThePhysicsofCompactObjects,ed.Shapiro,S.L.&Teukolsky,S.A. Shu,F.H.1991,PhysicsofAstrophysics,Vol.I,ed.Shu,F.H.(UniversityScienceBooks) Sion,E.M.,&Ready,C.J.1992,PASP,104,87 Skrutskie,M.F.,etal.2006,AJ,131,1163 Smith,R.K.,Mushotzky,R.,Mukai,K.,Kallman,T.,Markwardt,C.B.,&Tueller,J.2008,PASJ,60,43 Staubert,R.,Maisack,M.,Kendziorra,E.,Draxler,T.,Finger,M.H.,Fishman,G.J.,Strickman,M.S.,&Starr,C.H.1995,AdvancesinSpaceResearch,15,119 Steele,I.A.,&Clark,J.S.2001,A&A,371,643 Steele,I.A.,Negueruela,I.,&Clark,J.S.1999,A&AS,137,147 Stetson,P.B.1987,PASP,99,191 Tan,J.C.,&Draine,B.T.2004,ApJ,606,296 Tanaka,Y.,&Shibazaki,N.1996,ARA&A,34,607 Taranova,O.G.,&Shenavrin,V.I.2011,AstronomyLetters,37,30 Tomsick,J.A.,Heinke,C.,Halpern,J.,Kaaret,P.,Chaty,S.,Rodriguez,J.,&Bodaghee,A.2011,ApJ,728,86 206

PAGE 207

vanderBliek,N.S.,etal.2004,inSocietyofPhoto-OpticalInstrumentationEngineers(SPIE)ConferenceSeries,Vol.5492,SocietyofPhoto-OpticalInstrumentationEngineers(SPIE)ConferenceSeries,ed.A.F.M.Moorwood&M.Iye,1582 vanKerkwijk,M.H.,Chang,P.,&Justham,S.2010,ApJL,722,L157 Wallace,L.,&Hinkle,K.1997,ApJS,111,445 Wang,Q.D.,Gotthelf,E.V.,&Lang,C.C.2002,Nature,415,148 Warner,C.,Gonzalez,A.,&Eikenberry,S.2008,FATBOY:DataProcessingPipelineforFLAMINGOS-II,ProceedingsoftheAstronomicalDataAnalysisSoftwareandSystemsConference Weisskopf,M.C.2010,inAmericanInstituteofPhysicsConferenceSeries,Vol.1248,AmericanInstituteofPhysicsConferenceSeries,ed.A.Comastri,L.Angelini,&M.Cappi,525 Wilms,J.,Allen,A.,&McCray,R.2000,ApJ,542,914 Wood,K.S.,etal.1984,ApJS,56,507 Worrall,D.M.,Marshall,F.E.,Boldt,E.A.,&Swank,J.H.1982,ApJ,255,111 Yamasaki,N.Y.,etal.1997,ApJ,481,821 Zoccali,M.,etal.2006,A&A,457,L1 Zurita,C.,Casares,J.,Hynes,R.I.,Shahbaz,T.,Charles,P.A.,&Pavlenko,E.P.2004,MNRAS,352,877 207

PAGE 208

CurtisNoelDeWittwasborninLancaster,Pennsylvania.Hedouble-majoredinastronomyandphysicsatPennStateUniversity,betweenJune1997andDecember2001.HejoinedDr.JianGe'sextrasolarplanetresearchgroupwhileatPennStateandstayedonasresearchstaffuntilAugust2004whenheenrolledinthegraduateastronomyprogramattheUniversityofFlorida.After2006,hejoinedStephenEikenberryandRebaBandyopadhyaytoworkonGalacticCenterresearch,andtohelpdevelopnovelinfraredinstrumentation. 208