XMMU J0541.8-6659, a new supernova remnant in the Large Magellanic Cloud

0.38 MB·English

by  M. -H. Grondin

Astronomy&Astrophysicsmanuscriptno.A˙new˙SNR˙in˙the˙LMC˙Grondin-et-al˙TwoColumns (cid:13)c ESO2012
January6,2012
XMMUJ0541.8-6659, a new supernova remnant
⋆
in the Large Magellanic Cloud
M.-H.Grondin1,M.Sasaki1,F.Haberl2,W.Pietsch2,E.J.Crawford3,M.D.Filipovic´3,
L.M.Bozzetto3,S.Points4,R.C.Smith4
1 Institutfu¨rAstronomieundAstrophysikTu¨bingen,Universita¨tTu¨bingen,Sand1,72076Tu¨bingen,Germany
2 Max-Planck-Institutfu¨rextraterrestrischePhysik,Giessenbachstrasse,85748Garching,Germany
3 UniversityofWesternSydney,LockedBag1797,PenrithSouthDC,NSW1797,Australia
4 CerroTololoInter-AmericanObservatory,Casilla603,LaSerena,Chile
2
1 Preprintonlineversion:January6,2012
0
2 ABSTRACT

n
Context.ThehighsensitivityoftheXMM-Newtoninstrumentationofferstheopportunitytostudyfaintandextendedsourcesinthe
a
Milky Way and nearby galaxies such as the LargeMagellanic Cloud (LMC) in detail. The ROSAT PSPCsurvey of the LMC has
J
revealedmorethan700X-raysources,amongwhichthereare46supernovaremnants(SNRs)andcandidates.
5
Aims.We have observed the field around one of the most promising SNR candidates in the ROSAT PSPC catalogue, labelled

[HP99]456withXMM-Newton,todetermineitsnature.
]
Methods.WeinvestigatedtheXMM-Newtondataalongwithnewradio-continuum,nearinfraredandopticaldata.Inparticular,spec-
E
tralandmorphologicalstudiesoftheX-rayandradiodatawereperformed.
H
Results.TheX-rayimagesobtainedindifferentenergybandsrevealtwodifferentstructures.Below1.0keVtheX-rayemissionshows
. theshell-likemorphologyofanSNRwithadiameterof∼73pc,oneofthelargestknownintheLMC.Foritsthermalspectrumwe
h
estimateanelectrontemperatureof(0.49±0.12)keVassumingnon-equilibriumionisation.TheX-rayimagesabove1.0keVreveal
p
alessextendedsourcewithintheSNRemission,located 1′ westofthecentreoftheSNRandcoincidentwithbrightpointsources
-
o detectedinradio-continuum. Thishardcomponent hasanextentof0.9′ (i.e.∼13pcatadistanceof∼50kpc)andanon-thermal
r spectrum.ThehardsourcecoincidesinpositionwiththeROSATsource[HP99]456andshowsanindicationforsubstructure.
t Conclusions.WefirmlyidentifyanewSNRintheLMCwithashell-likemorphologyandathermalspectrum.AssumingtheSNRto
s
a beintheSedovphaseyieldsanageof∼23kyr.Weexplorepossibleassociationsofthehardnon-thermalemittingcomponentwitha
[ pulsarwindnebula(PWN)orbackgroundactivegalacticnuclei(AGN).

1 Keywords.galaxies:MagellanicClouds,ISM:supernovaremnants
v
2
8 1. Introduction to 1994. For a description of the ROSAT mission and PSPC
0
detectors,seeTru¨mper(1982),Briel&Pfeffermann(1986)and
1 The study of supernovaremnants (SNRs) is crucialfor a com-
Pfeffermannetal. (1987). This survey revealed 758 sources
. plete understandingof the chemicalcompositionandevolution
1 (Haberl&Pietsch, 1999, hereafter labelled HP99), among
0 oftheISMinagalaxybecauseoftheirenergyandmatterinputs which 46 sources were classified as firmly identified SNRs or
2 into the interstellar medium (ISM). The sample of SNRs stud-
candidates.
1 ied in our Galaxy is biased because of the high absorption in
SeveralSNRsintheLMChavebeeninvestigatedusingob-
: theGalacticdisk.Therefore,observationsofnearbygalaxiesal-
v servations with ROSAT (Filipovic´etal., 1998; Williamsetal.,
lowustoperformunbiasedpopulationstudiesandconstrainthe
Xi physicalpropertiesofthesourcesindetail.Itisthusalsopossi- 1999), Chandra X-ray Observatory (Hughesetal., 2006;
Sewardetal., 2010) and XMM-Newton (Williamsetal., 2004;
r ble to better understandthe evolutionand structure of the ISM
a intheMilkyWay. Klimeketal., 2010; Crawfordetal., 2010) satellites, allow-
ing a more detailed view of their morphologies and spectra.
Located at a distance of ∼ 50 kpc to the Earth
Badenesetal. (2010) have studied the size distribution of the
(Freedmanetal., 2001; Macrietal., 2006), the Large
SNRs in the Magellanic Clouds (MCs), which has a maxi-
Magellanic Cloud (LMC) offers the ideal laboratory for
mum at ∼40 pc and may extend up to sizes of ∼100 pc. With
studying a large sample of different types of objects (such
an extent of over 100 pc, SNR 0450-70.9 and SNR 0506-
as SNRs) in greater detail than in any other galaxy. Since
6542 (DEM L72) are among the largest SNRs detected in
its first detection in X-rays (Marketal., 1969), the LMC has
theLMC(Williamsetal.,2004;Cajkoetal.,2009;Desaietal.,
been extensively observed, but the major step forward came
from more than 200 observations in a 10◦ × 10◦ field centred 2010; Klimeketal., 2010), which may be highly evolved (age
up to 100 kyr). The size distribution of the MC SNRs as well
on the LMC, which have been performed with the ROSAT
as those in our Galaxy or the neaby spiral galaxy M33 can-
Position Sensitive Proportional Counter (PSPC) from 1990
notbeexplainedonlybytheSedovexpansionmodelforSNRs,
⋆ Based on observations with XMM-Newton, an ESA Science but seems to be largely affected by the ambient ISM densities
Mission with instruments and contributions directly funded by ESA (Bandiera&Petruk, 2010; Badenesetal., 2010, and references
MemberstatesandtheUSA(NASA). therein).
1
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
Furthermore, multi-frequency observations of several SNR
candidateslocatedintheLMChaveenabledtheirfirmidentifica-
tionbasedonmorphologicalandspectralcriteria(Bojicˇic´etal.,
2007; Crawfordetal., 2010) and have revealed a strong cor-
relation between the X-ray sources and the emission observed
by the Magellanic Cloud Emission Line Survey (MCELS;
Smithetal.,2000).Indeed,anenhancementofthe[Sii]andHα
coincident with the X-ray emission can be observed in most
cases. In particular, the ratio [Sii]/Hα is often higher than 0.4
(Levensonetal.,1995;Williamsetal.,2004).However,several
SNRsintheMagellanicCloudssuchasSNRLMCJ0528-6714
or/andSMCSNRJ010505-722319donothaveanyopticalemis-
sion.
While the emission of the gas shocked by the shock waves
of SNRs is mainly of thermal nature, there can also be a pul-
sar or a pulsar wind nebula (PWN) in a SNR, which produce
non-thermalemission.Pulsarsarerapidlyrotatingneutronstars
characterised by short periods(up to a few 10 s) and high sur-
face magnetic fields (Manchesteretal., 2005). The dissipation
oftherotationalenergyofpulsarsviamagnetisedparticlewinds
canbeattheoriginofPWNe(Gaensler&Slane,2006).Ahigh
percentage of PWNe known in our Galaxy have been detected
Fig.1. Smoothed RGB three-colour image of combined
in X-rays, and present a power-law spectra with a mean spec-
exposure-correctedXMM-Newton EPIC pn, MOS1 and MOS2
tralindexofΓ=–1.8±0.6(Kargaltsev&Pavlov,2008,2010).
data (red: 0.2 - 1.0 keV; green : 1.0 - 2.0 keV; blue : 2.0 - 4.5
SensitiveX-rayobservationshaveenabledthedetectionofsev-
keV, square root scale). The instrumental background was es-
eral PWNe and candidateswithin the Magellanic Clouds, with
timated using the method described in Section 3.2.1 and sub-
similar properties as PWNe in the Milky Way (Gaensleretal.,
tractedfromtheimages.Therenormalisationfactorwasderived
2003;Williamsetal.,2005;Owenetal.,2011).
from the shaded detector corners. The field of the radio image
presentedinFigure5isoverlaidforcomparison(greensquare).
There are now over 50 well-established SNRs in the
ThepositionoftheROSATsource[HP99]456isrepresentedby
LMC (Badenesetal., 2010; Klimeketal., 2010, and refer-
ences therein) and some additional ∼ 20 SNR candidates ayellowcircle.
(Bozzettoetal.,inprep.).Thiswouldcompriseoneofthemost
complete samples of SNRs in external galaxies. Therefore, it
is of prime interest to study LMC SNRs and compare them
with SNRs in other galaxies such as M33 (Longetal., 2010),
M83 (Dopitaetal., 2010), the Small Magellanic Cloud (SMC; 2. Observationsanddatareduction
Filipovic´etal., 2005; Payneetal., 2007; Filipovic´etal., 2008)
andourGalaxy(Stuparetal.,2008;Green,2009). 2.1.X-rays
The SNR candidates in the ROSAT PSPC catalogue TheXMM-NewtonsatelliteisanX-rayobservatoryoperatedby
have been classified based on the X-ray spectrum and theEuropeanSpaceAgency(ESA).Thesource[HP99]456has
spatial extent. Additional comparison to radio data taken been proposed for observations with XMM-Newton (Obs. Id.
with the Molonglo Observatory Synthesis Telescope (MOST; 0651880101,P.I.:M.Sasaki).Thispaperpresentsresultsofthe
ν=843 MHz; Turtle&Mills, 1984) and with optical data of X-rayanalysisofa20ksobservationobtainedonthissource.
MCELShasshownthatthereareROSATPSPCsourceswithra- The source [HP99] 456 has been observed on 2010 June
diooropticalcounterpartsindicativeofanSNRbutwithahard 06 (from 04:18:56 to 09:52:50 UT) with the European Photon
X-rayspectrumandthereforenotypicalSNRcharacteristicsin Imaging Camera (EPIC) in full-frame mode and thin filters.
X-rays. The source [HP99] 456 is one of the most promising Using the EPIC MOS1, MOS2 and pn CCDs, it offers the op-
new SNR candidates of this kind, with hard emission detected portunity to perform sensitive X-ray observations of a field of
by ROSAT PSPC and a possible radio or optical counterpart. theskyofdiameterof30′.Moredetailedtechnicaldescriptions
It has been recently re-investigated through new observations of the EPIC cameras are presented by Turneretal. (2001) and
withXMM-NewtonandtheAustraliaTelescopeCompactArray Stru¨deretal.(2001).
(ATCA; Hughesetal.,2007). TheEPICdatawereanalysedwithSASv10.0.01.Theexpo-
suretime,afterremovingperiodsofhighbackground,is∼11ks.
Inthispaper,wereportontheresultsoftheanalysisofnew
Pixelsflaggedasbadwerenottakenintoaccountandscreening
XMM-Newton and ATCA follow-upobservationsof the source
onthepatterns(from0to12forMOS;from0to4forpn),corre-
[HP99] 456. Section 2 presents the observations and analysis
spondingtothecanonicalsetofvalidX-rayevents(calibratedon
techniques. Results of the multi-frequency analyses are pre-
theground)wasappliedfortheimageandspectralanalyses.We
sented in Section 3. Section 4 presents the discussion on the
performedsource detection by using the SAS tasks eboxdetect
SNRpropertiesandonthedifferentscenariostoexplainthenon-
andemldetect.
thermalemission.Inparticular,weexploreapossibleassociation
ofthissecondcomponentwithaPWNoranAGN.Conclusions
arepresentedinSection5. 1 ScienceAnalysisSoftware:http://xmm.esac.esa.int/sas/.
2
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
2.2.Radio-continuum 2. Ahardemittingregion:above1.0keV,theX-rayanalysisre-
vealsalessextendedhardercomponent.Itscentreislocated
The field of [HP99] 456 was observed with the Australia ∼1′ westfromthecentreofthesoftshell-likeemissionand
TelescopeCompactArray(ATCA)on2010November29,with presentsanelongatedmorphologyoflength∼0.9′.Thepo-
anarrayconfiguration6C,atwavelengthsof3 and6cm (9000
sitionisconsistentwiththeROSATpositionof[HP99]456.
and 5500 MHz), and a bandwidth of 2 GHz (ATCA project
C2367). The observationswere carried out in snap-shot mode,
totalling about 1 hr of integration over a 12 hr period. PKS
Thepresenceofthehardemittingregioncoincidentwiththe
B1934-638wasusedforfluxandbandpasscalibrationandPKS
ROSATsourcesuggeststhat[HP99]456isunrelatedtotheshell-
0530-727was used for phase calibration. Standard calibration,
like emission region. Therefore, we designated the soft, shell-
editing and imaging techniques (Sault&Killeen, 2011) were
like source XMMUJ0541.8-6659accordingto its approximate
used.Largebandwidthmultifrequencyclean(Sault&Wieringa,
central coordinates.Additionaldetails on the positionsand ex-
1994)wasusedtodeconvolvetheimage.Wepointoutthatinter-
tents of both emitting regions are summarised in Table 2. The
ferometerssuchastheATCAsufferfrommissingfluxowingto
countratesoftheEPIC-pnobservationsarelistedinthelastcol-
thelackofshortspacings,whichsignificantlyaffectstheoverall
umn.
detectionofextendedemissionlikethatfromSNRs.
Wealsousedvariousotherradioobservations(SeeTable1)
including 843 MHz by Millsetal. (1984), 1377 MHz from
Hughesetal. (2007) and 4800 MHz & 8640 MHz taken from
themosaicprojectofDickel(2005).
2.3.Optical
The Magellanic Cloud Emission Line Survey (MCELS)2 was
carried out from the 0.6 m University of Michigan/CTIO
Curtis Schmidt telescope, equipped with a SITE 2048× 2048
CCD, which gave a field of 1.35◦ at a scale of 2.4′′pixel−1
(Smithetal., 2006; Winkleretal., inprep). Both the LMC and
SMCweremappedinnarrowbandscorrespondingtoHα,[Oiii]
(λ=5007Å), and[Sii] (λ=6716,6731Å),plusmatchedredand
green continuum bands. All data were continuum subtracted,
flux-calibratedandassembledintomosaicimages(asmallsec-
tionofwhichisshowninFigure6).
3. Dataanalysis
3.1.X-raymorphology
The morphological analysis was performed in three energy
ranges:0.2–1.0keV(softband),1.0–2.0keV(mediumband),
Fig.2. Smoothed image of exposure-corrected XMM-Newton
2.0 – 4.5 keV (hardband).For each energyinterval,exposure-
EPIC data in the 0.2 – 1.0 keV energy range. The dashed cir-
correctedimages were obtained using SAS toolsusing the fol-
cle shows the region considered for the estimation of the local
lowingmethod.
background. The thick circle represents the region used to de-
First,X-raydatawerecleanedusingtheselectionprocedure
rivethespectrumofthesoftemissionfromtheshell-likestruc-
described in the previous section. We computed event images
ture. Cancelled circles show the position of the bright sources
and exposure maps using the filtered dataset. Then we applied
excludedfromthespectralanalysis.
amasktoremovebadpixelsfromthethreeinstruments.Wedi-
videdeachresultingeventimagebythecorrespondingexposure
map and smoothed the resulting images with a Gaussian filter
usingasmooth.Finally,weaddedtheeventimagesfromthedif-
ferentinstrumentsforeachenergyband.
Figure 1 shows the RGB exposure-correctedimage around
3.2.X-rayspectralanalysis
theunidentifiedsource[HP99]456obtainedwiththeEPICdata
in the three bands defined above (red: 0.2 – 1.0 keV; green : The following sections present the results of the subsequent
1.0– 2.0keV;blue: 2.0– 4.5keV) aftersubtractionofthe in- spectral analyses of the two spectral components mentioned
strumental background(see caption). The emission seen in the above.Wefirstdescribethemethodusedtoestimatethecontri-
hardestbandissimilartothemediumband.Thismorphological butionfromtheintrinsicdetectorandX-raybackgroundforeach
analysisrevealstheexistenceoftwodistinctemittingregions: emittingregion.Thespectralanalysiswas performedusing the
XSPEC v 12.6.0 package (Arnaud, 1996; Dorman&Arnaud,
1. A soft emitting region: below 1.0 keV, the X-ray emis-
2001).
sionisdominatedbya brightthermal(formoredetails,see
Pointsourcesthatweredetectedusingthemethoddescribed
Section3.2)component.Itpresentsashell-likemorphology
inSection2.1wereexcludedfromthespectralanalysis.Weused
withanextentof∼5.0′×4.6′.
databetween0.3keVand5.0keVbecausenosignificantemis-
2 MCELS:http://www.ctio.noao.edu/mcels/ sionisdetectedathigherenergies.
3
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
Table1.Integratedradiofluxdensitiesof[HP99]456.
ν λ R.M.S BeamSize S Reference
Total
(MHz) (cm) (mJy) (′′) (mJy)
843 36 1.5 43.0×43.0 0.045 Millsetal.(1984)
1377 20 1.5 45.0×45.0 0.029 Hughesetal.(2007)
4800 6 0.6 30.0×30.0 0.008 Dickel(2005)
8640 3 0.6 12.0×12.0 0.004 Dickel(2005)
5500 6 0.1 2.8×2.2 — ThisWork
9000 3 0.1 2.8×2.2 — ThisWork
Table 2. Morphologicaldetails on the two emitting regions.The last columnlists the countratesestimated from the analysisof
EPIC-pndatabetween0.3and5.0keV.
Emitting Ra Dec Size Positionangle pncountrate
region (hh:mm:ss.d) (◦:′:′′) (′) (◦) (ctss−1)
Softemission 05:41:51.5 -66:59:02.8 5.0×4.6 –45 (18±2)×10−2
Hardemission 05:41:39.4 -66:58:45.8 0.9×0.45 –15 (2.8±0.3)×10−2
3.2.1. Estimationoftheinstrumentalbackground from Starketal. (1992) and was used for the absorption in the
Galaxy.
We estimated the contribution from the intrinsic instrumen-
tal background using the filter wheel closed (FWC) data
(Freybergetal.,2001a,b).Theintrinsicbackgroundiscomposed 3.2.3. Softemissionregion
ofinternalelectronicnoiseaswellascontinuousandfluorescent
X-ray emission induced by high-energy particles. This back- The X-ray emission below 1.0 keV is dominated by a struc-
ground is measured by operating the EPIC instruments in the ture presenting a shell-like morphology with a maximal di-
“closed”filterwheelposition,wherenophotonsfromastrophys- ameter of ∼ 5.0′ and centred at RA(J2000) = 05h41m51.5s,
ical sources can be observed. The FWC spectrum needs to be DEC(J2000)=–66◦59′03.8′′.Photonslocatedwithintheregion
renormalisedusingthecontinuumathigherenergies. correspondingto the hard X-ray emitting region, as defined in
Table 2, were excluded from the source spectrum to avoid any
contamination,asweredetectedpointsourcesvisibleinFig.1.
3.2.2. EstimationoftheX-raybackground
The spectrum of the shell-like structure was obtained after
The contribution of the X-ray local background emission was thesubtractionoftheinstrumentalbackground.TheX-rayback-
derived from the same observationas the source itself. We de- ground was estimated using a local background spectrum ex-
fined a region located close to our source (see Figure 2), and tracted from the same data. The source spectrum and the local
extractedthespectrumfromtheobservationandtheFWCdata. backgroundweremodelledsimultaneouslyinXSPECbyusing
We estimated the renormalisation factor of the FWC spectrum abackgroundmodelconsistingofthecomponentsasexplained
withrespecttotheobservedspectrumusingthetotalcountrates intheprevioussection.Allparametersofthethreebackground
measuredabove5keVinthecornersoutsidethefieldofviewof componentswerefixedexceptthenormalisations.Thenormali-
theEPICcamera.Thevalueoftherenormalisationfactoris1.3. sationsofthebackgroundcomponentsandtheparametersofan
The spectrum of the diffuse emission after subtracting the additionalmodelcomponentfortheshellemissionwerefreefit
instrumental backgroundcan be modelled as a sum of the fol- parameters.
lowingthreecomponents:
– a soft component, modelling the emission from the local
bubble with a collisional plasma, non-equilibrium model
Table3. XMM-Newtonspectralresultsofthesoftthermalemit-
(NEI; Borkowskietal., 1994, 2001; Hamiltonetal., 1983;
Liedahletal.,1995)3assumingalowtemperature(0.1keV); ting region in the 0.3 - 5.0 keV energy range. Model is VNEI
withabundancesfixedto0.5ofthesolarabundances.
– a hard component modelled with an absorbed NEI model
withatemperatureof0.3keV,whichstandsfortheemission
Parameter Value
fromtheGalactichalo;
N (×1021cm−2)intheGalaxy 0.6(fixed)
– an absorbed power-law of spectral index Γ = 1.46, to de- H
N (×1021cm−2)intheLMC 0(fixed)
scribetheextragalacticunresolvedemission. H
kT(keV) 0.49±0.12
The parameters are fixed to values similar to those used by Ionisationtimescale*τ(scm−3) (1.5±0.7)×1010
Kuntz&Snowden(2010). Absorbedflux(0.3–5.0keV,erg/cm2/s) 2.5×10−13
Absorbedluminosity(0.3–5.0keV,erg/s) 5.1×1034
The two last components are convolved with a Tu¨bingen-
Unabsorbedluminosity(0.3–5.0keV,erg/s) 8.9×1034
Boulder ISM absorption model (TBABS). The Galactic fore-
Reducedχ2 1.27
groundhydrogencolumndensityof0.6×1021cm−2wasderived
Degreesoffreedom 86
3 For more details on XSPEC models, please see *:Theionisationtimescaleτ=net,whereneistheelectronnumber
http://heasarc.nasa.gov/xanadu/xspec/manual/manual.html densityandtistheageofthegas.
4
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
The spectral analysis of the soft circular emitting re-
gion reveals a thermal spectrum. It can be modelled
with a single-temperature non-equilibrium ionisation colli-
sional plasma model (VNEI; Borkowskietal., 1994, 2001;
Hamiltonetal.,1983;Liedahletal.,1995)withatemperatureof
(0.49±0.12)keVwithanabsorptionofN =0.6×1021 cm−2
H
fortheGalacticforeground(withsolarabundances)andavalue
ofN fortheabsorptionintheLMC.Theabundanceswerefixed
H
to 0.5 of the solar system abundancesfor the emission and ab-
sorption taking place in the LMC. This value is the standard
mean value for the LMC (Russell&Dopita, 1992). The value
of N forthe LMC turnedoutto tend towards0 with an 90 %
H
C.L.upperlimitof0.4×1021 cm−2.Thisparameterwasthere-
forefixedto0.
The resulting spectrum is presented in Fig. 3. The corre-
spondingspectralparametersaregiveninTable3.
Fig.4. XMM-Newton EPIC-pn spectrum of the hard emitting
regionwithapower-lawmodelfit.Thesolidlineisforthehard
emission plus backgroundspectra. Here, the backgroundis the
sum of the shell emission and the local emission. The back-
groundspectrumandshellspectrumwereevaluatedasdescribed
in Section 3.2.4. The backgroundspectrum was not subtracted
fromthesourcespectrumbutwasmodelledsimultaneouslyand
is included in the spectral model of the source spectrum. The
contributionfromthehardemittingregionisrepresentedinred.
Thebottompanelshowstheresidualsfromthebest-fitmodel.
roundingshell-like emission as the localbackground.The nor-
malisationsofthecomponentcorrespondingtotheX-rayback-
groundcomponentswererenormalisedtotheareaandwerefit-
ted, alongwith the spectralparametersof the hard emittingre-
gion.
Fig.3.XMM-NewtonEPIC-pnspectrumandmodelofthe soft Thespectrumofthehardemittingregionisnon-thermaland
emitting region with a VNEI model fit (black points and solid canbe modelledin the 0.3–5.0keV energyrangewith anab-
line). The estimation of the local background spectrum and sorbedpower-lawwithaspectralindexofΓ=1.8±0.3,withan
model (red points and solid line) is described in Section 3.2.1. absorptionofN =0.6×1021cm−2fortheGalacticforeground
H
The background spectrum was not subtracted from the source (fixed,withsolarabundances)andavalueofN fortheabsorp-
H
spectrum but was modelled simultaneously and is included in tionintheLMCofN =(0.58±0.53)×1021cm−2.Thisyields
H
thespectralmodelofthesourcespectrum.Thedifferentspectral anabsorbedfluxof∼9.5 × 10−14erg/cm2/s(reducedχ2=0.22
components are shown separately (red and black dotted lines) for 6 degrees of freedom). The resulting spectrum is shown in
forthebackgroundandsourcespectra.Thebottompanelshows Fig.4.Inparticular,thecontributionfromthehardX-rayemit-
theresidualsfromthebest-fitmodels. tingregionisrepresentedinred.
Analternativetechniquewasusedtoderivethespectrumof
the hard emitting region. Instead of using the FWC data, we
extracted a backgroundspectrum from a nearbyregion located
withintheshellandsubtracteditfromthesourcespectrum.This
3.2.4. Hardemissionregion
methodyieldsconsistentresultsforthehardemission.
The X-ray emission above 1.0 keV is dominated by a narrow
structureoflength∼0.9′located∼1′tothewestofthecentroid
3.3.Radio
oftheshell-likethermalemission,asseenfromFigure1.Above
1.0keV,thehardsourcehasasurfacebrightnessof∼5.8×10−3 New high-resolution (∼2′′) radio-continuum observations re-
counts/s/arcmin2,i.e.∼6timesthebackgroundlevel,whichhas solvedlow-resolutionemissioninto threebrightsourceswithin
asurfacebrightnessof∼9.4×10−4 counts/s/arcmin2. Thereis thehardX-rayemittingregion(Fig.5).Forcomparison,thesize
indication for substructure in the X-ray image, which could be oftheradioimageshowninFig.5isrepresentedinFig.1.Inpar-
causedbymultiplesources.TheROSATsource[HP99]456co- ticular,threebrightsourcescanbeseencoincidingwiththecen-
incideswiththenorthernpartofthenarrowstructure. tralpositionofthishardnon-thermalX-rayemittingregion.The
Toanalysethespectrumofthissecondcomponentagainthe brightest source (the most eastern one) has a spectral index of
instrumentalbackgroundspectrumwassubtractedandtheX-ray α=–0.5±0.1withintegratedfluxdensitiesof(0.85±0.05)mJy
spectrumwasmodelledsimultaneously,thistimeusingthesur- at5.5GHzand(0.67±0.05)mJyat9.0GHz.
5
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
Fig.5. High-resolution ATCA radio-continuum image at 6 cm
(5.5 GHz) overlaid with 20 cm contours (blue) from the low-
resolution mosaic image of the immediate surroundings of Fig.6. Hα emission (arbitrary units) derived from the MCELS
[HP99] 456. Contours are from 3 to 18 mJy/beam in steps of in the surroundingsof [HP99] 456. A contour of the soft ther-
1 mJy/beam and 6 cm beam size (2.8′′ ×2.2′′) is shown in the mal emission of XMMUJ0541.8-6659is overlaid in green for
bottomleftcorner. comparison.
OurestimateoftheSNRoverallradiospectralindex(exclud-
ingcentralpointsource;usinguxdensitiesfrommosaicsurveys
onlyandlistedinTable1i.e.observationsfrom843,1377,4800,
8640MHz)isα=–1.0±0.2,whichismoretypicalforyounger
SNRs.
3.4.Optical
Theemissionfromopticallines(Hαand[Sii])derivedfromthe
MCELSwereconsidered.TheHαemissioninthesurroundings
of[HP99]456arepresentedinFig.6.Thisimageshowsapos-
siblecorrelationbetweenthewesternpartoftheshell-likeX-ray
thermalemissionobservedbyXMM-Newton,thelow-resolution
radio-continuum and the Hα emission line, as seen in Figures
5 and 6. However, we acknowledge that this correlation could
be a chance coincidence because the Hα “cloud” could be an
unrelatedHiiregion.
Low-resolution radio-continuum images at 36, 20, 6 and
3 cm show good alignment with the optical (MCELS) feature
andalsocoincidewiththemaximumemissionfromtheX-rays.
Supernova remnants can show significant Hα, [Sii] and
[Oiii] line emission if they are in the radiative phase. In this
case,thefluxratio[Sii]/Hαisacrucialparametertodistinguish
between SNRs and e.g., Hii regions (Levensonetal., 1995). Fig.7. Soft X-rayemission of XMMUJ0541.8-6659.The con-
However,the[Sii]/Hαratioaround[HP99]456is∼0.3,which toursofthenear-infraredemission(Spitzer,5.8µm)areoverlaid
isnotnecessarilyindicativeofSNRemission. forcomparison.
Hα emission is spatially coincident with the hard emitting
region, but additional studies are required to determine if they
areassociatedornot.
4. Discussion
morphologyand is firmly identified as a new SNR, designated
ThankstothehighsensitivityofXMM-Newton,arecentobserva- XMMUJ0541.8-6659. The nature of the emitting region with
tionofthefieldof[HP99]456hasrevealedtheexistenceoftwo harder spectrum is still unclear, but we examined the possible
emittingcomponents.Thesoftercomponentpresentsa circular identificationasaPWNorabackgroundAGN.
6
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
4.1.PropertiesofthenewlyidentifiedSNR elongated structure located within the SNR, which dominates
XMMUJ0541.8-6659 theX-rayemissionabove1.0keV.Itscentreislocated∼1′from
thecentreoftheSNR. Thishardemissionregionhasanappar-
X-raydatareveala largestructureofsoftemissioncloseto the ent extent of ∼ 0.9′, correspondingto ∼ 13 pc at a distance of
ROSAT PSPC source [HP99] 456 in the LMC. Analyses per-
∼50kpcandanon-thermalspectrum.However,theangularres-
formedindifferentenergyrangesindicatethatthissourcedom-
olutionofXMM-Newtondoesnotallowustoclarifyifthishard
inatestheX-rayemissionbelow1.0keV.
source is indeed extended (with substructure) or if it is a con-
ThespectrumoftheSNRXMMUJ0541.8-6659canbemod-
glomerateofpointsources.AcarefullookattheROSATimages
elledwithanabsorbedcomponentrepresentingaplasmainnon-
reveals that the position of [HP99] 456 is consistent with the
equilibriumionisationcharacterisedbyatemperatureof∼0.49
hard source detected with XMM-Newton and in particular with
keV, which is the average of the typical temperature of previ-
themorenorthernknotinthehardemission(seeFig.1).Hence,
ouslydetectedSNRsintheLMC.We notethatthespectralpa-
apossibleassociationbetweenthesoftandthehardcomponents
rametersquotedinTable3areconsistentwiththeparametersof
and the exact nature of the hard X-ray source remains unclear.
the unambiguouslyidentifiedSNRs in the LMC(Klimeketal.,
The radio observations reveal several point sources at the po-
2010;Levensonetal.,1995;Williamsetal.,1999,2004).Itfur-
sition of the northern X-ray knot. This may point towards an
ther supports the identification of this thermal emission with a
identificationasapulsarwithPWN.However,equivalently,one
SNR.
mayargueforanexistenceofseveralbackgroundsourcessuch
The softemitting regionis firmly identified as a new SNR,
asAGN.Weinvestigatethedifferentscenariosbelow.
in view of its morphology and X-ray spectrum. It has a max-
Firstly,weexaminethepossibilitythatthehardX-rayemit-
imum diameter of ∼ 5.0′, which corresponds to a maximum
ting region is associated to background AGN. With a flux of
extent of ∼ 73 pc at a distance of ∼ 50 kpc. Therefore,
5.1×10−14 erg cm−2 s−1 (0.5– 2 keV), we expectto see ∼ 12
XMMUJ0541.8-6659isoneofthelargestSNRobservedinthe
AGNpersquaredegreeinthesky(Rosatietal.,2002).Wehave
LMC(Williamsetal.,1999;Klimeketal.,2010).
an8%probabilityforanAGNinsidea5.0’x4.5’SNRshell.This
However,thelargeextentofthesourcedoesnotnecessarily
lowprobabilityandthenon-detectionofanysimilarmorpholog-
implyanoldagefortheSNR.Indeed,assumingtheSNRtobein
ical (jet-like) structure either in radio, IR or in optical renders
theSedovphase,thedynamicageofthesourcecanbeestimated
anassociationofthehardX-raycomponentwithanAGNquite
usingtheshocktemperatureasfollows:
unlikely.
k T =1.8×105 Rs 2 keV, (1) Itisworthnoticingthatthedifferentvaluesofabsorptionob-
B s (cid:18) t (cid:19) tainedduringthespectralfitofthehardandsoftemittingregions
mayindicatethattheyarelocatedatdifferentdistancesfromus.
where k is the Boltzmann constant, T is the temperature of
B s
WhiletheabsorptionoftheshellisconsistentwiththeSNRbe-
theshock,whichiscomparableheretotheplasmatemperature
ing located in the LMC, the slightly higher absorption of the
estimatedin Section3.2.3, R isthe radiusofthe shock(inpc)
s
hardemissionsuggeststhatthesourcemightbelocatedbehind
andtistheageoftheSNR(inyr).
Fromthespectralfityieldingaplasmatemperatureofk T ≈ theSNR.
B s
0.49keV,wederivedanageof∼23kyr.Thisvalueiswellbelow Secondly, we consider the possible association of the hard
theageofpreviouslargely-extendedSNRsdetectedintheLMC X-raycomponentwithaPWN.Threeradiosourcesareembed-
(e.g.Klimeketal.,2010;Williamsetal.,2004). dedinthehardandnarrowX-raycomponent.Thebrightestradio
Thisresultmeansthatthelargeextentofthesourceisnotdue sourcepresentsa spectrumthatcanbemodelledwith a power-
toitshighage,butrathertoanexpansioninaprobablyrarefied law of indexα ≈ −1.0whichis consistentwith spectraof pul-
ambientmedium.Thisisnotsurprisingknowingthepositionof sars.Ifthisradiosourceisindeedapulsar,thenitmightpower
XMMUJ0541.8-6659atthenorth-eastpartoftheLMC,where aPWN thatwouldbeseenintheX-raydomain.Timingobser-
thegasdensityisquitelow. vations are requiredto confirm or invalidate this identification.
AsalegacyprojectoftheSpitzerSpaceTelescope,asurvey Furthermore, it is worth noticing that with a spectral index of
oftheMagellanicCloudshasbeencarriedoutcalled“Surveying Γ=1.8±0.3,thespectrumofthehardX-rayemittingsourceis
the Agents of Galaxy Evolution” (SAGE)4 (Meixneretal., similar to spectra of PWNe in the Milky Way (spectralindices
2006)toobtainimagesandspectraofthedustemission.Auni- ranging from –2.5 to –1.2; Kargaltsev&Pavlov, 2008, 2010)
form survey of the LMC was performed in a 7◦ × 7◦ by the butalsotypicalforAGN.
InfraredArrayCamera(IRAC,dedicatedtonearinfraredobser- The multi-frequency observations along with the spectral
vations at wavelength 3.6, 4.5, 5.8, and 8 µm) and Multiband andmorphologicalresultsofourX-rayandradiodataanalyses
Imaging Photometer for Spitzer (MIPS, mid- and far-infrared do not uniquely identify the nature of the hard X-ray emission
observations at wavelengths 24, 70, and 160 µm). The near- embeddedintheSNRXMMUJ0541.8-6659.DeeperX-rayob-
infrared emission (at 5.8 µm) in the region surrounding the servationswithhigherangularresolutionarerequiredtodiffer-
source[HP99]456ispresentedinFig.7.Spitzer-SAGEobserva- entiatebetweenthetwoscenarios.
tions(Meixneretal.,2006)revealdustemissionspatiallycoin-
cidentwiththeeasternhalfoftheshell-likesoftemittingregion.
However, there is no clear indication that this dust emission is 5. Conclusions
associatedtothesourceXMMUJ0541.8-6659.
A recent X-ray observation with XMM-Newton of [HP99] 456
previously discovered by ROSAT led to the identification of a
4.2.Hardemissionregion:aPWNorbackgroundAGN? new SNR (XMMUJ0541.8-6659)in the LMC, which presents
a shell-like morphologyand a soft thermal spectrum. No clear
Energy-dependentmorphologicalstudiesoftheX-rayemission
correlationcanbefoundbetweentheX-rayandopticalemission.
in the surroundings of [HP99] 456 have revealed an extended
Thephysicalproperties(temperature,size,etc.)ofthisnewSNR
4 Spitzer-SAGE:http://sage.stsci.edu/index.php are consistent with SNRs previously identified in the LMC by
7
Grondinetal.,2011:XMMUJ0541.8-6659,anewSNRintheLMC
X-ray observations(Levensonetal., 1995; Klimeketal., 2010; Filipovic´,M.D.,Haberl,F.,Winkler,P.F.,Pietsch,W.,Payne,J.L.,Crawford,
Williamsetal.,1999,2004). E.J.,deHorta,A.Y.,Stootman,F.H.&Reaser,B.E.,2008,A&A,485,63
AdditionalanalysesoftheXMM-Newtondatahaverevealed Freedman,W.L.,Madore,B.F.,Gibson,B.K.,etal.,2001,ApJ,553,47
Freyberg,M.J.,Pfeffermann,E.&Briel,U.G.,2001,Proc.Symposium”New
a harder and narrower emitting region within the shell, which
visions of the X-ray Universe in the XMM-Newton and Chandra era”,
maybeaPWNorbackgroundAGN.Thisemissionislikelythe November2001,ESTEC,Netherlands
counterpartofthehardsourcediscoveredbyROSAT.Follow-up Freyberg,M.J.,Briel,U.G.,Dennerl,J.,Haberl,F.,Hartner,G.,Kendziorra,E.
observations with ATCA revealed several radio point sources &Kirsch,M.,2001,Proc.Symposium”NewvisionsoftheX-rayUniversein
theXMM-NewtonandChandraera”,November2001,ESTEC,Netherlands
coincidentwiththehardemittingregion,oneofwhichcouldbe
Gaensler, B. M., Hendrick, S. P.,Reynolds, S.P.,& Borkowski, K. J.,2003,
apulsar.DeeperobservationsofthehardX-rayemittingregion
ApJL,594,111
with the Chandra X-ray Observatory will help to distinguish Gaensler,B.M.&Slane,P.O.,2006,ARA&A,44,17
betweenthedifferentscenariosandtounveilitsnature. Green,D.A.,2009,BulletinoftheAstronomicalSocietyofIndia,37,45
Haberl,F.&Pietsch,W.,1999,A&AS,139,277
Hamilton,A.J.S.,Sarazin,C.L.&Chevalier,R.A.,1983,ApJS,51,115
Acknowledgements
Hughes,J.P.,Rafelski,M.,Warren,J.S.,Rakowski,C.,Slane,P.,Burrows,D.
The XMM project is supported by the Bundesministerium &Nousek,J.,2006,ApJL,645,117
fu¨r Wirtschaft und Technologie/Deutsches Zentrum fu¨r Luft- Hughes,A.,Staveley-Smith,L.,Kim,S.,Wolleben,M.&Filipovic´,M.,2007,
undRaumfahrt(BMWI/DLR,FKZ50OX0001)andtheMax- MNRAS,382,543
Kargaltsev,O.&Pavlov,G.G.,2008,AIPC,983,171
PlanckSociety.
Kargaltsev,O.&Pavlov,G.G.,2010,AIPC,1248,25
The Australia Telescope Compact Array is part of the
KlimekM.D.,Points,S.D.,Smith,R.C.,Shelton,R.L.&Williams,R.,2010,
AustraliaTelescope,whichisfundedbytheCommonwealthof ApJ,725,2281
Australia for operation as a National Facility managed by the Kuntz,K.D.&Snowden,S.L.,2010,ApJS,188,46
CSIRO. Long,K.S.,Blair,W.P.&Winkler,P.F.,2010,ApJS,187,495
Levenson,N.A.,Kischner,R.P.,Blair,W.P.&Winkler,P.F.,1995,Astron.J,
This work is based in part on observations made with
110,739
the Spitzer Space Telescope, obtained from the NASA/IPAC Liedahl,D.A.,Osterheld,A.L.&Goldstein,W.H.,1995,ApJ,438,L11
Infrared Science Archive, both of which are operated by the Macri,L.M.,Stanek,K.Z.,Bersier,D.,Greenhill,L.J.&Reid,M.J.,2006,
Jet Propulsion Laboratory, California Institute of Technology ApJ,652,1133
Manchester,R.N.,Hobbs,G.B.,Teoh,A.,&Hobbs,M.,2005,Astron.J,129,
under a contract with the National Aeronautics and Space
1993
Administration.
Mark,H.,Price,R.,Rodrigues,R.,Seward,F.D.&Swift,C.D.,1969,ApJ,155,
The Magellanic Clouds Emission Line Survey (MCELS) L143
data are provided by R. C. Smith, P. F. Winkler, and S. D. Meixner,M.,Gordon,K.D.,Indebetouw,R.,etal.,2006,Astron.J,132,2268
Points.TheMCELSprojecthasbeensupportedinpartbyNSF Mills,B.Y.,Turtle,A.J.,Little,A.G.&Durdin,J.M.,1984,AustralianJournal
ofPhysics,37,321
grantsAST-9540747andAST-0307613,andthroughthegener-
Owen,R.A.,Filipovic´,M.D.,Ballet,J.,etal.,2011,A&A,530,132
oussupportoftheDeanB.McLaughlinFundattheUniversityof Payne,J.L.,White,G.L.,Filipovic´, M.D.&Pannuti,T.G.,2007,MNRAS,
Michigan,abequestfromthefamilyofDr.DeanB.McLaughlin 376,1793
in memory of his lasting impact on Astronomy. The National Pfeffermann,E.,Briel,U.G.,Hippmann,H.,etal.,1987,SoftX-rayopticsand
technology; Proceedings of the Meeting, Berlin, Berlin, Society of Photo-
OpticalAstronomyObservatoryis operatedbythe Association
OpticalInstrumentationEngineers,733,519
ofUniversitiesforResearchinAstronomyInc.(AURA),under
Rosati,P.,Tozzi,P.,Giacconi,R.,etal.,2002,ApJ,566,667
acooperativeagreementwiththeNationalScienceFoundation. Russell,S.C.&Dopita,M.A.,1992,ApJ,384,508
M.-H. Grondinis supportedby the BMBF/DLR grant#50- Sault,R.J.,&Wieringa,M.H.1994,A&AS,108,585
0R-1009. Sault, B. & Killeen, N., 2011, Users Guide (Australia Telescope National
Facility)
Seward,F.D.,Williams,R.M.,Chu,Y.-H.,Gruendl,R.A.&Dickel,J.R.,2010,
References Astron.J,140,177
Smith,R.C.,Leiton,R.&Pizarro,S.,2000,ASPConferenceProceedings,221,
Arnaud,K.A.,1996,AstronomicalDataAnalysisSoftwareandSystemsV,eds. 83
G.JacobyandJ.Barnes,p17,ASPConf.Series,101,17 Smith,R.C.,Points,S.,&Winkler,P.F.,2006,NOAONewsletter,85,6
Badenes,C.,Maoz,D.&Draine,B.T.,2010,MNRAS,407,1301 Stark,A.A.,Gammie,C.F.,Wilson,R.W.,Bally,J.,Linke,R.A.,Heiles,C.&
Bandiera,R,&Petruk,O.,2010,A&A,509,34 Hurwitz,M.,1992,ApJ,79,77
Bojicˇic´,I.S.,Filipovic´,M.D.,Parker,Q.A.,Payne,J.L.,Jones,P.A.,Reid,W., Stru¨der,L.,Briel,U.,Dennerl,K.,etal.,2001,A&A,365,L18
Kawamura,A.&Fukui,Y.,2007,MNRAS,378,1237 Stupar,M.,Parker,Q.A.&Filipovic´,M.D.,2008,MNRAS,390,1037
Borkowski,K.J.,Sarazin,C.L.&Blondin,J.M.,1994,ApJ,429,710 Tru¨mper,J.,1982,AdvancesinSpaceResearches,2,241
Borkowski,K.J.,Lyerly,W.J.,&Reynolds,S.P.,2001,ApJ,548,820 Turner,M.L.J.,Abbey,A.,Arnaud,M.,etal.,2001,A&A,365,L27
Bozzetto,L.etal,2011,A&A,inprep. Turtle,A.J&Mills,B.Y.,1984,AstronomicalSocietyofAustralia,5,537
Briel,U.G.,Pfeffermann,E.,1986,NuclearInstrumentsandMethodsinPhysics Williams,R.M.,Chu,Y.-H.,Dickel,J.R.,Petre,R.,Smith,R.C.&Tavarez,M.,
ResearchSectionA,242,376 1999,ApJS,123,467
Cajko,K.O.,Crawford,E.J.&Filipovic´,M.D.,2009,SerbianAstronomical Williams,R.M.,Chu,Y.-H.,Dickel,J.R.,Gruendl,R.A.,Shelton,R.,Points,
Journal,179,55 S.D.&Smith,R.C.,2004,ApJ,613,948
Crawford,E.J.,Filipovic´,M.D.,Haberl,F.,Pietsch,W.,Payne,J.L.&deHorta, Williams,R.M.,Chu,Y.,Dickel,J.R.,etal.2005,ApJ,628,704
A.Y.,2010,A&A,518,35 Winkler,P.F.etal.,2011,inprep
Desai,K.M.,Chu,Y.-H.,Gruendl,R.A.,Dluger,W.,Katz,M.,Wong,T.,Chen,
C.-H.R.,Looney,L.W.,Hughes,A.,Muller,E.,Ott,J.&Pineda,J.L.,2010,
Astron.J,140,584
Dickel,J.R.,2005,AmericanAstronomicalSocietyMeeting207,27,1403
Dopita,M.A.,Blair,W.P.,Long,K.S.,etal.,2010,ApJ,710,964
Dorman,B.&Arnaud,K.A.,2001,AstronomicalDataAnalysisSoftwareand
SystemsX,eds.F.R.Harnden,Jr.,F.A.Primini,&H.E.Payne,238,415
Filipovic´, M. D., Pietsch, W., Haynes, R. F., White, G. L., Jones, P. A.,
Wielebinski,R.,Klein,U.,Dennerl,K.,Kahabka,P.&Lazendic,J.S.,1998,
A&AS,127,119
Filipovic´, M.D.,Payne,J.L.,Reid, W.,Danforth, C.W.,Staveley-Smith, L.,
Jones,P.A.&White,G.L.,2005,MNRAS,364,217
8