325 and 610 MHz Radio Counterparts of SNR G353.6$-$0.7 a.k.a. HESS J1731$-$347

0.51 MB·

by  A. J. Nayana

MNRAS000,1–9(2016) Preprint12January2017 CompiledusingMNRASLATEXstylefilev3.0
−
325 and 610 MHz Radio Counterparts of SNR G353.6 0.7
−
a.k.a. HESS J1731 347
Nayana A.J.1⋆, Chandra, Poonam1, Roy, Subhashis1, Green, David A.2, Acero, Fabio3,
Lemoine-Goumard, Marianne4, Marcowith, Alexandre5, Ray, Alak K.6 and
Renaud, Matthieu5
7
1 1National centre forRadio Astrophysics, Tata Institute of Fundamental Research, PO Box 3, Pune 411 007, India
0 2Astrophysics Group, Cavendish Laboratory, 19 J. J. Thomson Avenue, Cambridge CB3 0HE
2 3Laboratoire AIM, IRFU/SAp - CEA/DRF - CNRS- Universit´eParis Diderot, Bat. 709, CEA-Saclay, Gif-sur-YvetteCedex, France
4Centre d’Etudes Nucl´eaires de Bordeaux Gradignan, Universit´eBordeaux, CNRS/IN2P3, 33175 Gradignan, France
n
5Laboratoire Universet Particules de Montpellier(LUPM), Universit´ede Montpellier, CNRS/IN2P3, Montpellier, France
a
6Tata Instituteof Fundamental Research, Homi Bhabha Road, Mumbai 400001, India
J

0
1

] AcceptedXXX.ReceivedYYY;inoriginalformZZZ
E
H
ABSTRACT
.
h HESSJ1731−347a.k.a.SNRG353.6−0.7isoneofthefiveknownshell-typesupernova
p remnants (SNRs) emitting in the very high energy (VHE, Energy > 0.1 TeV) γ-
-
ray domain. We observed this TeV SNR with the Giant Metrewave Radio Telescope
o
r (GMRT) in 1390, 610 and 325 MHz bands. In this paper, we report the discovery of
t 325 and 610 MHz radio counterparts of the SNR HESS J1731−347 with the GMRT.
s
a VariousfilamentsoftheSNRareclearlyseeninthe325and610MHzbands.However,
[ the faintest feature in the radio bands corresponds to the peak in VHE emission. We

explainthisanti-correlationintermsofapossibleleptonicoriginoftheobservedVHE
1
γ-ray emission. We determine the spectral indices of the bright individual filaments,
v
5 which were detected in both the 610 and the 325 MHz bands. Our values range from
6 −1.11 to −0.15, consistent with the non-thermal radio emission. We also report a
7 possible radio counterpart of a nearby TeV source HESS J1729−345 from the 843
2 MHzMolongloGalacticPlaneSurveyandthe1.4GHzSouthernGalacticPlaneSurvey
0 maps.Thepositiveradiospectralindexofthispossiblecounterpartsuggestsathermal
1. origin of the radio emission of this nearby TeV source.
0 Key words: ISM:supernovaremnants–ISM:individualobjects:SNRG353.6−0.7–
7
radiocontinuum:ISM–radiationmechanisms:non-thermal–accelerationofparticles
1
: – cosmic rays
v
i
X
r
a
1 INTRODUCTION have discovered many very high energy (VHE; Energy >
100 GeV) γ-ray sources from the Galactic plane in the
Galactic cosmic rays are energetic particles believed to
last decade (Degrange & Fontaine 2015). Association of
be mostly composed of hadrons with energies ranging
spatially-resolved shell-like γ-ray sources with SNRs is
from ∼ 0.1 to 108−9 GeV (Blandford & Eichler 1987).
the signature of particle acceleration at supernova shocks.
How are these particles accelerated to these high energies
Studying these objects provides a unique opportunity
remainsafundamentalquestion.Supernovaremnant(SNR)
to probe the cosmic-ray particles that emit high energy
shocks are considered as one of the probable candidates
γ-rays.However,sofarthereareonlyfivespatially-resolved
of these particle acceleration sites (Blasi 2013). Cherenkov
shell-like VHE sources firmly identified as SNRs (for a
telescopes such as the High Energy Stereoscopic System
review,see Acero et al.2015).TheseareRXJ1713.7−3946
(H. E. S. S.), the Major Atmospheric Gamma Imaging
(Aharonian et al. 2004; H.E. S. S.Collaboration et al.
Cherenkov Telescope (MAGIC) and the Very Energetic
2016a), RX J0852.0−4622 (Aharonian et al. 2007;
Radiation Imaging Telescope Array System (VERITAS)
H.E. S. S.Collaboration et al. 2016b), RCW 86
(Aharonian et al. 2009; H.E. S. S.Collaboration et al.
⋆ E-mail:[email protected]
c 2016TheAuthors
(cid:13)
2 Nayana et al.
2016c), SN 1006 (Aceroet al. 2010) and HESS J1731−347 from the SNR (Aharonian et al. 2008). The 1390 MHz ob-
(Abramowskiet al. 2011). servationswerecarriedoutontwoconsecutivedayssoasto
HESS J1731−347 was discovered (Aharonian et al. increase the sensitivity. Details of the GMRT observations
2008)asaVHEγ-raysourceintheH.E.S.S.GalacticPlane are given in Table 1.
Surveyatα =17h31m55s,δ =−34◦42′36′′ withno Thedatawasrecordedwithanintegrationtimeof16.1
J2000 J2000
identifiedcounterpartinotherwavebands.LaterTian et al. s.Forallthefrequencybands,thebandwidthchosenwas33
(2008) discovered a faint shell-type radio source SNR MHzsplitinto256frequencychannels.Alltheobservations
G353.6−0.7 with an angular size of ≈ 30 arcmin, in spatial were taken with Automatic Level Control (ALC) on. The
coincidence with HESS J1731−347 in the Southern Galac- ALC controls the amplitude gains of the antennas with a
tic Plane Survey (SGPS) at 1420 MHz (Haverkornet al. negativefeedbackloopattheoutputofeachantenna.Iten-
2006) and Molonglo Galactic Plane Survey (MOST) at suresasafe operatingpointwithoutrunningtheelectronics
843 MHz (Green et al. 1999). Tian et al. (2008) derived intosaturation. 3C286 was used as thefluxdensity calibra-
an integrated flux density of 2.2±0.9 Jy at 1420 MHz by tor at all frequencies. At 325 MHz, sources J1714−252 and
azimuthally averaging the radio intensity in rings about J1830−360 were used as the phase calibrators. J1830−360
the centre of the SNR. Later a deeper γ-ray observa- and J1714−252 were used as phase calibrators at the 610
tion of HESS J1731−347 revealed the shell-like morphol- and 1390 MHz observations, respectively. The flux density
ogyinγ-raysaswell(Abramowski et al.2011),whichmade calibratorwasusedtocalibrateantennagains,andthephase
HESS J1731−347 the fifth member of the VHE shell SNR calibrators were used to correct for phase variations due to
group. HESS J1731−347 was also observed with the Fermi atmospheric fluctuations. The phase calibrators were also
Large Area Telescope (LAT), but no GeV counterpart used for bandpasscalibration.
was detected (Yanget al. 2014). HESS J1731−347 was de- The system temperature (T ) correction was imple-
sys
tected in X-rays in archival ROSAT data (Aharonian et al. mentedbydeterminingtheself-powersofeachantennawith
2008; Tian et al. 2008). The X-ray emission was found to ALCoffinaseparatetestobservationrun,takeninDecem-
be non-thermal (Tian et al. 2010; Abramowskiet al. 2011; ber2015inthe325,610and1390MHzbands.TheSNRand
Bamba et al. 2012), indicating that theelectron population thefluxcalibratorwereobservedfor10minuteseach.Power
producing these non-thermalX-raysare of TeV energies. A equalisationwasdoneontheSNR.Antennaself-powers(to-
compact object XMMU J173203.3−344518 was detected at talpowermeasuredbyantenna)wererecordedforallanten-
thecentreoftheremnantbyHalpern & Gotthelf(2010a)us- nas, for each timestamp (16.1 s) and for all channels.
ingXMM-Newtonarchivaldata.Thisobjectwasconsidered
as a central compact object (CCO) associated with SNR
G353.6−0.7 (Aceroet al. 2009; Halpern & Gotthelf 2010a;
Tian et al. 2010). Tian et al. (2008) estimated the distance
3 DATA ANALYSIS
totheSNRas3.2±0.8 kpc,byassumingthenearbyHiire-
gion G353.42−0.37 tobe at thesame distance as theSNR. 3.1 Calibration and Imaging
In this paper, we present the Giant Metrewave Radio
The data was analysed using Astronomical Image Process-
Telescope(GMRT, Swarup et al.1991)observationsofSNR
G353.6−0.7inthe325,610and1390MHzbands.Ourdetec- ing System (AIPS) developed by the National Radio As-
tronomyObservatory(NRAO, Greisen2003).Oneantenna
tionsinthe325and610MHzbandsarethelowestfrequency
was not working in the 325 as well as 610 MHz band ob-
detections of the SNR thus far. We compare the shell mor-
servations whereas three antennas were not working in the
phology of the SNR from our low-frequency measurements
1390MHzobservationsduetotechnicalproblems.Thesean-
with those of other published high-frequency results. We
tennaswereexcludedfrom furtheranalysis.Corrupteddata
also measure the spectral index of different filamentsof the
SNR.Thepaperisorganisedasfollows:In§2weexplainthe dueto instrumentalproblems were removedusing standard
GMRTobservationsandin§3webrieflydiscussthemethod AIPS routines. Channels with radio frequency interference
ofdataanalysis.The§4containstheresultsanddiscussions (RFI) were flagged using the task SPFLG 1. This task dis-
and the§5 contains thesummary and conclusions. plays the uv data in a grid with frequency channels in the
x axis and time in the y axis. It also provides interactive
options to inspect and flag the data. Amplitude and phase
calibration were done for single channel and the solutions
2 GMRT OBSERVATIONS were applied to all the channelsin thedata. Bandpass cali-
bration was done using the phase calibrator. At 1390 MHz
We observed SNR G353.6−0.7 with the GMRT at 325, 610
band, the two days of data were combined together after
and 1390 MHz. The 325, 610 and 1390 MHz observations
calibrationusingtheAIPStaskDBCON.Thetargetsource
werecarriedouton2013Oct19.3(UT),2015Aug29.5(UT)
datawasaveragedinsmallfrequencybinssoastoreducethe
and 2013 Dec 19.1,20.1 (UT), respectively. In the 325 and
effect of bandwidth smearing, which reduces the amplitude
610 MHz bands, the GMRT full width at half maximum
ofthevisibility.Thefractionalreductioninthestrengthofa
(FWHM) of primary beam is 85.2 and 43 arcmin, respec-
sourceat aradial distancer from thecentreof field tothat
tively, and only one pointing was necessary to cover the
SNR.Inthe1390MHzband,theFWHMis26.2arcmin,and
observations were taken in four pointings since the field-of-
view (FOV) is less than the source extent. Three pontings 1 Detailed documentation of SPFLG as well as all other
weredonetowardstheSNRregionandonepointingtowards AIPS tasks mentioned in this section is available at
anotherTeV source HESS J1729−345, just 30 arcmin away http://www.aips.nrao.edu/cook.html
MNRAS000,1–9(2016)
Low Frequency Radio counterpart of HESS J1731-347 3
Table 1.DetailsofGMRTobservations.
Frequency Dateofobservation Timeonsource FWHMofprimarybeam Numberof Pointingcentre
(MHz) (UT) (hours) (arcmin) pointings RA(J2000) Dec(J2000)
325 2013Oct19.3 5.4 85.2 1 17h 31m 54.9s 34◦ 42′ 36.00′′
−
610 2015Aug29.5 3.5 43 1 17h 32m 19.6s 34◦ 43′ 30.00′′
−
1390∗ 2013Dec19.1,20.1 2.5 26.2 4 17h 32m 11.9s 34◦ 37′ 60.00′′
17h 32m 29.9s −34◦ 49′ 60.00′′
17h 31m 35.9s −34◦ 44′ 60.00′′
17h 29m 35.9s −34◦ 32′ 60.00′′
−
*Inthe1390MHzband,observations weretakenintwoconsecutive dayswith4pointingstocoverSNRG353.6 0.7andHESSJ1729 345.
− −
at thecentreof field is given by usingtaskFLATN.Thistaskdoesaninterpolationofmulti-
plefieldsproducedbythetask IMAGRintoasingleimage.
θ ν r∆ν
R =1.064 b 0erf 0.833 In the 1390 MHz band, the final map was made by com-
b r∆ν (cid:18) θ ν (cid:19)
b 0 biningmapsofallthreepointingstowardstheSNR.Finally
where θ is the angular size of synthesized beam, ν is the primary beam corrections were made for all themaps.
b 0
centreoftheobservingbandand∆ν isthebandwidthofthe Ourobservationsaresensitiveuptothelargestangular
signal. We averaged a few channels in such a way that the scales of ∼ 41.2, 20.6 and 4.0 arcmin in the 325, 610 and
fractional reduction in the strength of a source at the edge 1390 MHz bands, respectively. Since the size of the SNR
ofthefieldis5percent.Forexample,inthe325MHzband, G353.6−0.7is30arcmin,wedonothaveenoughshortspac-
the adjacent 4 channels were averaged to obtain a channel ings to completely measure the flux density of SNR at the
width of 0.5 MHz and 8 channels were averaged to get a 610 and 1390 MHzfrequencies. However,theeffect of miss-
channel width of 1 MHz in the 610 MHz band. In the 1390 ing flux is small in the 325 MHz map. A simulation was
MHz,10channelswereaveragedtoobtainachannelwidthof done to quantify the flux density missing using AIPS task
1.25 MHz. These sub-channels were stacked together while UVMOD.Thistaskallowstomodifytheuv databyadding
makingimages. amodel.Askymodelofasoliddiscof30arcminindiameter
Multi-facetimagingwasdoneusingthetaskIMAGRas wassimulated and thensampled totheGMRTuv-coverage
thetwodimensionalfastfouriertransformleadstoaflatsky at the 325 MHz. The uv-data was then imaged and more
approximation. Multi-facet imaging takes care of the phase than 97 per cent of flux density was recovered. Thus our
errorsduetothisapproximation.Thenumberof facets was GMRT325 MHz map sufferfrom a maximum of 3 percent
calculatedusingtheAIPStaskSETFC.Atotalof55facets missing flux while imaging a 30 arcmin structure. Simula-
werecreatedforthe325MHzdata,and37facetsinthe610 tions for missing flux were done for individual filaments at
MHzband.Inthe1390MHzband,31facetswerecreatedfor 610 MHz which we describe in §4.3.
each pointing. First, high-resolution images of the compact The SNR was detected in the 325 and 610 MHz bands
sourcesweremadeforself-calibration usingvisibilities from (see Fig. 1); however, we do not detect it in the 1390 MHz
baselinesgreaterthan1km,i.e.excludingtheGMRTcentral bandbecauseofthemissingfluxduetolackofshortspacings
squareantennas.Theimagesweremadeusinguv range2to and thesensitivity limitation.
25 kλ, 3 to 30 kλ and 5 to 105 kλ in the325, 610 and 1390
MHz bands, respectively. Only point sources were cleaned,
whereas, the SNR region and other extended sources were
3.2 System Temperature (T ) correction
notcleanedduringhigh-resolutionimaging.Afewroundsof sys
phaseonlyself-calibrationandtworoundsofamplitudeand The SNR G353.6−0.7 is close to the Galactic plane, where
phase self-calibration were run. All the clean components theskytemperature(T )ishighduetothecontributionof
sky
weresubtractedfromtheuv datausingtaskUVSUB,which the Galactic diffuse emission. Since observations were done
takestheuvdatasetandcleancomponentfilesasinputsand withtheALCONinitsdefaultmodeandALCcontrolsthe
subtracts the clean components from the uv data set. This antennagains automatically, theantennagains are reduced
procedure was followed essentially to remove the compact withrespecttothecalibratorsandthefluxdensitymeasure-
sources in thefield since theSNRhas extendedstructure. mentswillbelessthantheirtruevalue.Astheantennaself-
Finally, low-resolution maps were made using small uv powers are proportional to the system temperatures (T ),
sys
range (only central square data). This corresponded to a acorrectioncanbefoundbymeasuringtheratioofantenna
uv range of 0 to 2 kλ, 0 to 3 kλ and 0 to 5 kλ for the self-powerstowardsthetargetsourceandfluxcalibrator,un-
325,610and1390MHzbandsdata,respectively.Theshort- der the assumption that the contribution of electronics to-
est baselines we have for these observations are B ∼ 50, wards the system temperature will not change significantly
min
100 and 500λ for the 325, 610 and 1390 MHz bands re- betweenthedaysofobservationsatthesamefrequency.The
spectively. In the 1390 MHz band, a significant fraction of actual fluxdensity of thetarget source can be recovered by
short baselines was flagged due to RFI. This makes the in- applyingthisT correction.
sys
terferometer sensitive toangular scales less than0.6λ/B Test observations for the T correction were taken in
min sys
(Basu et al.2012),providedthereareenoughmeasurements December 2015, and the data was analysed using GMRT
atshortbaselines.Allfacetsineachpointingwerecombined specificofflinepackages.Afterremovingcorrupteddata,the
MNRAS000,1–9(2016)
4 Nayana et al.
ratio of self-powers was found for each antenna. The mean andarewellbelowthemaximumangularscalesthatcanbe
andthermsoftheratioswerecomputedtoestimatetheav- mapped with GMRT in both the 325 and 610 MHz bands.
erage T correction. A detailed explanation of T correc- Weestimatedamissingfluxoflessthan2percentforthese
sys sys
tioncanbefoundinGMRTtechnicalreportbyRoy(2006)2. filamentsinboththe325and610MHzbandsinourobserva-
Anotherway ofobtainingT correction istoestimate tions.Thusthesestructuresarenotaffectedfromsignificant
sys
T atbothtargetsourceandfluxcalibratorpositionsfrom missing fluxproblems.
sky
theall-skytemperaturemap.Haslam408MHzfullskytem- To determine the spectral indices, both 325 and
perature map by Haslam et al. (1982) and 150 MHz tem- 610 MHz maps were convolved to the coarser resolution
perature measurements byLandecker& Wielebinski (1970) (150×105arcsec2)usingtheAIPStaskCONVL.Bothmaps
wereusedtoobtainT atbothpositions.TheT at325, were then aligned to same geometry using the AIPS task
sky sky
610 and 1390 MHz were calculated by assuming a spectral OHGEO. The flux density of the same region over both
index of −2.55 for Galactic diffuse emission (Roger et al. mapswas determinedusingtask BLSUMandrescaled with
1999). System temperature has contributions from the sky, theT correctionfactortogettheactualfluxdensity.The
sys
ground and receiver. The ground temperature (T ) and re- totalerrorσ inthefluxdensitymeasurementisacombi-
g total
ceivertemperature(T )weretakenfromaGMRTtechnical nation of errors from calibration and rms noise of the map.
rec
report3. Hence the uncertainties in flux density measurements are
The T correction obtained for each frequency estimated as
sys
from GMRT test observations, from the Haslam map
σ = σ2 +σ2 (1)
(Haslam et al. 1982) and from the 150 MHz map total q rms cal
(Landecker& Wielebinski 1970) are given in Table 2. The where σ is the map rms noise and σ is the calibration
rms cal
T correctionfromtheHaslammapis54.6percentgreater error.Weassumedthatthecalibrationerrorsare10percent
sys
than that obtained from GMRT test observations at 325 of the measured flux density.The errors on spectral indices
MHz and 27.6 percent greater at 610 MHz. The correction are computedusing standard error propagation formulas.
from the 150 MHz map is 39.7 per cent greater than that
of the GMRT correction factor at 325 MHz and 17.0 per
centgreater at610 MHz.At1390 MHz,allthreecorrection 4 RESULTS AND DISCUSSION
factors are similar. A similar trend in T corrections with
sys
4.1 Morphology of SNR G353.5−0.7
GMRTobservationsandtheHaslammapwasalsoreported
byMarcote et al. (2015). Our observations with the GMRT have revealed the radio
The Tsky measurements using GMRT at 240 MHz was morphology of the SNR at the 325 and 610 MHz frequen-
determined by Sirothia (2009). They report differences in cies. Fig. 1 (top panels) clearly shows a shell-like structure
theTsky measurements obtained from theHaslam map and withfilaments.Theextentofshellstructureis∼30arcmin,
GMRTmeasurements,andthermsofpercentagedifferences consistent with that measured by Tian et al. (2008) from
is56percent.Thisdiscrepancymightbeduetotheassump- the SGPS map at 1.4 GHz (bottom left panel of Fig. 1,
tion of a constant spectral index of −2.55 across the whole Haverkornet al.2006).The325MHzmaphasanrmsof4.8
sky, while determining Tsky using the Haslam map. This mJybeam−1 witharesolution of135×97 arcsec2.Sincethe
assumption need not hold in all regions of the sky, espe- SNR is extended, low-resolution maps were made in order
ciallytowardstheGalacticcentrewheresignificantcontribu- togetbettersignaltonoiseratiopersynthesizedbeam.The
tionfromthermalemission isexpected.Moreover,variation radio emission is brightest in the filament seen towards the
in Trec with elevation and ambient temperatures (Sirothia south-east,andis detected at 10σ significance. The faintest
2009), will also affect the Tsys measurements. Estimation filament is towards the north and is detected at 3σ signifi-
from the Haslam map does not account for these issues. cance. The rms of the 610 MHz map is 4 mJy beam−1 for
Hence Tsys correction from GMRT test observation seems a resolution of 152×105 arcsec2. The diffuse faint emission
to be more reliable than the one obtained from the Haslam towardsthenorthisnotdetectedinthe610MHzband.This
map.Inthispaper,weusethecorrectionfactorfromGMRT is due to sensitivity limitation and poor sampling at short
observations toscale theobserved fluxdensity values. baselines. The smaller field of view at the 610 MHz (43 ar-
cmin)alsolimitsbettermappingofthe30arcminstructure
inthesky.Filaments1and2(toprightpanelofFig.1)seen
3.3 Spectral Analysis inthe610MHzGMRTmaparealsoseenintheMOST843
MHzmap(bottomrightpanelofFig.1, Green et al.1999).
Analysis of spectral index (α, where the radio flux density
S at frequency ν is related to α as S = να) provides in- The brightest emission comes from the south-east filament
ν ν
(Filament 1),which isalso thecasein theGMRT325 MHz
formation about thenatureof theradio emission. Sincethe
image.
full shell structure was not detected in the 610 MHz band,
TheSNRisnotdetectedinthe1390MHzGMRTmap.
we determined spectral indices for four filaments (see fila-
Themaprmsis1.8mJybeam−1.Theshortestbaselineavail-
ments marked 1, 2, 3 and 4 in the top right of Fig. 1 and
able with the GMRT at 1390 MHz is ∼ 500λ. This makes
Table3) whichwere detectedinboththe610 and325 MHz
the interferometer sensitive to a maximum angular scale of
bands. The sizes of the filaments are less than 11 arcmin
4arcmin.ThusitisnotpossibletoimagethefullSNRshell
at 1390 MHz using the GMRT. But the bright filamentary
2 http://ncralib1.ncra.tifr.res.in:8080/jspui/handle/2301/315 structures of small angular size can be detected. However,
3 http://gmrt.ncra.tifr.res.in/gmrt−hpage/Users/doc/manual we did not detect any filaments due to sensitivity limita-
/Manual−2013/manual−20Sep2013.pdf tion and lack of enough short baselines. The flux densities
MNRAS000,1–9(2016)
Low Frequency Radio counterpart of HESS J1731-347 5
Table 2.Tsys correctionfromdifferentmethods.
Frequency Correction Correctionusing408MHzmap Correctionusing150MHzmap
(MHz) (GMRT) (Haslametal.1982) (Landecker &Wielebinski,1970)
325 4.01 0.51 6.20 5.60
±
610 1.59 0.13 2.03 1.86
±
1390 1.08 0.04 1.10 1.10
±
Notes:Corection(GMRT)isthemeanofratiosofantennaself-powersmeasuredtowardsSNRandfluxcalibrator.Theuncertainitiesareat1σlevel.
Correction(Haslametal.1982)andcorrection(Landecker&Wielebinski,1970)arecorrectionsobtainedfromskytemperaturemaps.
Figure 1. Top left: The GMRT low resolution map of the field containing SNR G353.6 0.7 at 325 MHz. Map resolution is 135 97
arcsec2,rms=4.8mJybeam−1.Topright:weshowGMRTlowresolutionmapofthefieldc−ontainingSNRG353.6 0.7at610MHz.M×ap
resolutionis150 105arcsec2,rms=4mJybeam−1.Heretheareasmarkedby1,2,3,4arethefilamentsforwhic−hspectralindiceswere
×
estimated. Compact sources have been removed from both 325 and 610 MHz GMRT maps. For comparison, we show the SGPS map
(Haverkornetal.2006)ofthefieldcontainingSNRG353.6 0.7at1420MHzwithresolutionis100 100arcsec2,rms=10mJybeam−1
− ×
(bottom left ), as well as the MOST map (Greenetal. 1999) of the field containing SNR G353.6 0.7 at 843 MHz. with resolution is
49.71 43.00arcsec2,rms=3mJybeam−1 (bottomright).Thecirclesinallfourimagesareofsame−sizeandcentredattheCCOposition
×
(markedwithacross)indicatetheareaenclosingSNRG353.6 0.7.Thecolourschemeusedinthemapsiscubehelix(Green2011).
−
of point sources in the GMRT map are compared with the pectedsinceX-raysarefoundtobenon-thermal(Tian et al.
flux densities from NRAO VLA Sky Survey (NVSS) map 2010;Abramowski et al.2011;Bamba et al.2012).Thenon-
at 1.4 GHz and they matched within 10 – 15 per cent. The thermalX-rayand radioemission isexpected tocomefrom
brightest filament seen in both the 325 and the 610 MHz the synchrotron radiation from relativistic electrons accel-
GMRT maps is also seen as a 4σ feature in the NVSS map erated at the shock. However, our 325 MHz GMRT maps
which has an rmsof 0.7 mJy beam−1. clearlyshowradioemissiontobepresentinthisregionalso,
Tian et al. (2010) overlaid the X-ray map with the 1.4 eliminating this discrepancy.
GHz SGPS contour map, which revealed that the X-ray The bright compact source G353.464−0.690
emission extendsbeyond the radio shell towards the north- (Zoonematkermaniet al. 1990) inside the remnant which is
ern region (see Fig. 1 of Tian et al. 2010). This is not ex- seen in the 1.4 GHz SGPS image is not seen in the GMRT
MNRAS000,1–9(2016)
6 Nayana et al.
map.Thisisbecausewehavesubtractedpointsourcesdur- consistent with the non-thermal emission. The spectral in-
ing our analysis. The bright radio source J173028−344144 dex of the Hii region is found to be 0.63±0.14 which is
(Condon et al. 1998) seen towards thewest of theSNRis a consistent with thermal emission.
Hiiregion.Theextendedstructureseenfurtherwestisalso Here we note a steeper spectral index −1.11 ± 0.22
likely to be a Hii region since both of these are seen in the for the filament 2. This steepening is consistent with syn-
IRAS 60µm map (Neugebaueret al. 1984) and the WISE chrotron cooling which steepens the spectrum by a factor
22µm map (Wright et al. 2010). of ∆α = 0.5 from normal synchrotron spectral index. If
TheintegratedfluxdensityoftheSNRatthe325MHz this steepening is real, this could have important implica-
is 1.84 ± 0.15 Jy. For a typical SNR spectral index of α ∼ tions for the magnetic field in the SNR shell, and thus the
−0.5(Jones et al.1998),thefluxdensityoftheSNRat1420 VHE emission scenario (see §4.4). Our future observations
MHz is expected to be 0.88 Jy. Tian et al. (2008) derived with the upgraded GMRT with 200 MHz bandwidth and
1420 MHz flux density using combined ATCA and Parkes 2.5timeshighersensitivitycomparedtotheexistingGMRT
telescope data to be 2.2±0.9 Jy. This value is likely to be willbeverycrucialheretodeterminetheprecisespectralin-
an overestimation due to the contamination from thermal diceswithmuchimprovederrorbars.Thewidebandwidthof
emissionfromnearbyHiiregion.Thisdiscrepancymayalso upgradedGMRT providesexcellent uv coverage, thusmore
be due to the flux density estimation made by Tian et al. sensitivitytofaintextendedradioemission,whichiscrucial
(2008) by taking one-half of the SNR (half towards low in accurately determining the spectral indices of the fila-
Galactic latitude)andextrapolatingtotheotherhalftoes- ments.
timatethecompletefluxdensity.However,ourvaluesmatch
within 1.5σ with that of Tian et al. (2008). The change in
4.4 Comparison with the VHE emission
flux density in the low-frequency end will change theshape
of the spectral energy distribution (SED) and the parame- SincethestructureoftheSNRisclearlydetectedinthe325
ters derived from multi-wavelength models. MHzradiomap,thespatialcorrelationoftheradioemission
with that of the VHE emission can be compared. We com-
paredthesitesofsynchrotronradioemission withtheVHE
4.2 Central compact object emission measured with the H. E. S. S. (Abramowski et al.
2011), as shown in Fig. 2. The spatial extent of the radio
A bright compact object , XMMS J173203−344518, was
emission is similar to that of the VHE emission except for
discovered in the X-ray band near the centre of the SNR
by Tian et al. (2010) from the XMM observations. The the region towards the west, where there is no radio emis-
CCO was also seen in the Suzaku and the Chandra data sion though bright VHE emission is seen. It is likely that
theradioemissioninthisregionissensitivitylimitedasthis
(Halpern & Gotthelf 2010a). A marginal pulsation period
region hashigherrms,duetocontributionfrom contamina-
of 1 s was found for the XMMS J173203−344518, which
tionfromnearbysources.The3σ upperlimitoffluxdensity
is between the periods of magnetars (2 − 12 s) and the
in this region at 325 MHz is 63 mJy which is comparable
CCOs (0.1−0.4 s) (Halpern & Gotthelf 2010a). However,
tothe fluxdensities of filaments 3 and 4 (see Table 3). The
Halpern & Gotthelf(2010b)failed toconfirmthispulsation
withthelaterChandraobservations.Thereisnoevidencefor morphology of theSNRis shell-likein both radioand VHE
emission, but the finer structure do not seem to be corre-
aradiocounterpartforthisobject(Tian et al.2010).Wedid
lated.Thebrightfilamentsinradio(south-eastfilamentand
notdetectradioemissionattheCCOpositionintheGMRT
eastern filament,see Fig.2)isfaint in VHEmap.Thepeak
observations at any frequencies. We provide deep limits on
in VHE emission is from the north-east region, where the
anypossibleradiocounterpartoftheCCO.The3σfluxden-
radio emission is faint (see Fig. 2).
sity upperlimits at the CCO position are 727, 780 and 184
The azimuthal variation in radio brightness can be ei-
µJy at the 325, 610 and 1390 MHz, respectively, from our
ther due to the gradient of ambient interstellar medium
GMRT high-resolution maps.
(ISM) density or due to the gradient of the magnetic field
strength at the shock. An encounter of SNR G353.6−0.7
with a dense molecular cloud can create a density gradi-
4.3 Spectral Index
ent in the ambient ISM, though so far there is no evidence
The global radio spectral index of the SNR cannot be de- for suchan interaction. The12CO spectra from thevicinity
terminedsincethecompleteshellisnotdetectedat the610 of SNR G353.6−0.7 obtained by Abramowski et al. (2011)
MHz. We determined the spectral indices of four filaments doesnotshowanykinematicfeaturesthataccountforapos-
oftheSNRlabelled1to4intheFig.1.Thephysicaldimen- sible interaction of the SNR with a molecular cloud as re-
sions in angular unitsand flux densities at the325 and 610 portedinmanyinteractingSNRs(e.g.Moriguchi et al.2005;
MHz of the filaments are listed in Table 3. The sizes of the Castelletti et al. 2013). The γ-ray azimuthal profile of SNR
filaments are less than 11 arcmin with less than 2 per cent G353.6−0.7 isroughlyflat(Abramowski et al.2011)except
missingfluxatboththe610andthe325MHzobservations. for the two γ-ray peaks towards the north-east and west,
Wehave also accounted for the backgroundcontribution to which issuggestiveof afairly uniform ambient ISMdensity
the SNR flux density. However, these are statistical esti- (Abramowskiet al. 2011). The shell structure of the SNR
mates, and the contribution of background flux density at recovered in the 325 MHz GMRT map is fairly symmetri-
theSNRposition is uncertain. cal. These three arguments suggest non-interaction of SNR
The spectral index values are −0.70 ± 0.19, −1.11 ± G353.6−0.7 with a molecular cloud. Hence it is reasonable
0.22, −0.5 ± 0.3 and −0.15 ± 0.32 for the filament 1, 2, to assume that the medium surroundingthe SNR is of rea-
3 and 4 respectively. The spectral index values are broadly sonably uniform density.
MNRAS000,1–9(2016)
Low Frequency Radio counterpart of HESS J1731-347 7
Table 3.Physicaldimensionsoffourfilamentsinangularunitsandfluxdensitiesat325MHzand610MHz.
Filament Length width Sint (325MHz) Sint (610MHz) Spectralindex
(arcmin) (arcmin) (mJy) (mJy)
Filament1 11 3 106.22 68.67 0.70 0.19
− ±
Filament2 6 2.8 121.66 58.94 1.11 0.22
− ±
Filament3 5.4 3.7 56.90 41.55 0.50 0.30
− ±
Filament4 5.9 3.4 42.99 39.45 0.15 0.32
− ±
Sintdenotestheintegratedfluxdensityofthefilamentsateachfrequencies.
In such a case, the variation in radio brightness can andreferencestherein).Thisprovidesastrongindicationof
be most likely due to the variation of magnetic field if the leptonicorigin of TeV γ-rays in SNRG353.6−0.7.
injection is isotropic (the efficiency of injection does not The presence of non-thermal X-ray emission further
depend on the angle between shock normal and magnetic supports the leptonic origin of VHEemission. Non-thermal
field).Radiobrightnessincreases intheregion ofhighmag- synchrotron X-ray emission implies the presence of TeV
netic field since synchrotron emissivity is proportional to electrons which can easily produce IC γ rays. The absence
B3/2 for a particledistribution N(E) scaling as E−2. Ifthe of any thermal X-ray emission further rules out hadronic
VHE emission is of leptonic origin, the electrons of ener- origin of VHE emission as thermal X-ray emission is pro-
gies E ∼ E emit VHE γ-rays. These electrons experi- portional to the square of the gas density, and hadronic
max
ence ample radiative losses in this region and the energy processes are likely to dominate in high density regions
loss rate is proportional to E2B2. Thus in the region of (Gabici & Aharonian2016).However,thehadronicscenario
high magnetic field, the number of electrons emitting In- is still possible if the SNR shock has not been able to heat
verse Compton (IC) γ-rays deplete faster. This results in a the gas to high temperature if the gas is clumpy, then the
low brightness of VHE emission where the radio brightness dense clump can produce TeV γ-rays via hadronic process
is still high (Petruket al. 2009a), thus potentially explain (Gabici & Aharonian 2014).
the anti-correlation in the radio and γ-ray emission. This Thisradio-VHEanti-correlationtrendisnotseeninany
explanationisvalidonlyinthecaseofsynchrotron-losslim- of the other four SNRs in the VHE shell class. In fact, cor-
ited acceleration scenario where Emax ∝ B−1/2 (Reynolds related emission of synchrotron radio and IC γ-ray emis-
2008).Inthecaseofage-limitedaccelarationscenario,Emax sion is seen in the bilateral SNR 1006 (Petruk et al. 2009b;
∝B (Reynolds2008)andtheanti-correlationcannotbeex- Aceroet al. 2010). This correlated emission is explained by
plained. However, the age of SNR G353.6−0.7 is not well thevariation ofmaximumenergyofelectronsovertheSNR
known.Theageestimates in theliteraturerangefrom 2000 to compensate for magnetic field variation (Petruk et al.
to 27000 years (Tian et al. 2008; Abramowskiet al. 2011; 2009a). Thiscan also happenifthedependenceof injection
Fukudaet al. 2014; Acero et al. 2015). and the electron maximum energy on the obliquity angle
Accordingtothismodel,weexplaintheanti-correlated (anglebetweenmagneticfieldandthenormaltotheshock)
emissionofradioandVHEforSNRG353.6-0.7inaleptonic is strong enough to dominate the magnetic field variations
scenario as follows: Evidence of variation of magnetic field (Petruket al. 2009a).
for SNR G353.6-0.7 comes from the analysis of spectral in- Visual inspection of Figure 2, suggests the VHE emis-
dices in §4.3. Here we see roughly a 0.5 steepening of the sion to be slightly beyond the radio shell. However, the ex-
spectral index of filament 2 compared to the typical SNR tentofradioemission couldbeduetosensitivitylimitation.
spectralindexof −0.5,consistent with synchrotroncooling, Anupgraded GMRT observation with 200 MHzbandwidth
indicating higher magnetic field in this filament. Thus the will provide 2.5 times the sensitivity and excellent uv cov-
magneticfieldstrengthisrelativelyhighintheregionofthe erage.Thusitwillbemoresensitivetothefainterextended
south-east and eastern filaments, and hence it is bright in featuresandwillbetracingthetrueextentoftheradioemis-
radio. Due to high synchrotron cooling, there are fewer IC sion.
electrons, and hence the VHE emission is faint here. The
magnetic field strength is low towards the northern region
of the SNR, and this results in less energy loss of electrons
4.5 Radio counterpart of nearby TeV source
and bright IC emission. Towards the west, magnetic field
HESS J1729−345
strengthislikelytobesimilarorlessthanthatofthenorth-
ern region, and the synchrotron emission is below the sen- There is another TeV source, HESS J1729−345, at posi-
sitivity limits where VHE emission is bright. One needs to tionα =17h29m35s,δ =−34◦32′22′′,∼30arcmin
J2000 J2000
measure local magnetic field using Zeeman splitting of OH away from the centre of HESS J1731−347 with no iden-
maser spots (Brogan et al. 2000). Magnetic field can also tified counterpart in other wavebands (Abramowski et al.
be estimated from the rim-width of thin X-ray filaments 2011). This source is located near a Hii region
by high-resolution X-ray observations (Parizot et al. 2006; (G353.381−0.114)andamoleculargasclumpinspatialpro-
Park et al. 2009).Theroughlya0.5steepeningof thespec- jection (Abramowskiet al.2011).Abramowski et al.(2011)
tral index of filament 2 coincides with theregion where the estimate near and far kinematic distances towards this
VHEemissionisweakest,supportingtheleptonicscenarioin molecular clump at ∼ 6 and ∼ 10 kpc respectively from
whichtheVHEproductionissuppressedduetosynchrotron 12COmolecularlineobservations.IfHESSJ1729−345isas-
cooling in high magnetic field (Gabici & Aharonian 2016, sociatedwithHESSJ1731−347,whichisonly3.2kpcaway,
MNRAS000,1–9(2016)
8 Nayana et al.
Figure3.MOSTmap(Greenetal.1999)ofthefieldcontaining
G353.6 0.7at843MHzoverlaidwithHESScontourswithlevels
Figure 2.TeVγ-rayexcessmap(Abramowskietal.2011)(res- 6,8,10,−12σ(σ=10counts).TeVmap(Abramowskietal.2011)
olution 144 144 arcsec2;shown inthe bottom leftcorner of the
image) over×laid with GMRT 325 MHz map (resolution 135 97 .
×
arcsec2;showninthebottomrightcorneroftheimage)contours.
Thescaleoftheγ-rayexcessmapisexcesscountspersmoothing 5 SUMMARY AND CONCLUSIONS
gaussian width (σ = 0.04 deg). The contours are √2nσ mJy
beam−1 (n= 2,2,3,4,5,6;σ=1.5mJybeam−1).T±hecirclein- WecarriedouttheGMRTobservationsofHESSJ1731−347
dicates the are−a enclosing SNR G353.6 0.7 centred at the CCO a.k.a.SNRG353.6−0.7inthe325,610and1390MHzband.
−
position (marked with a cross). The colour scheme used in the We also observed a nearby source, HESS J1729−345 with
mapiscubehelix(Green2011) the GMRT in the 325 and 1390 MHz frequency bands to
look for possible radio counterparts.Ourconclusions are:
this molecular clump cannot be in the vicinity of HESS (i) We detect the low-frequency radio morphology of SNR
J1729−345 (Abramowski et al. 2011). G353.6−0.7 with the GMRT at the 325 and 610 MHz
The TeV emission from HESSJ1729−345 hasbeen ex- frequencies. The shell structure of the SNR with diffuse
plained by Cui et al. (2016) with a hadronic scenario, due filaments is revealed in the 325 MHz map, and SNR is
totheinteractionoftheSNRwiththesurroundingmolecu- partially seen in the610 MHzmap.
lar material (as observed from 12CO emission), via proton–
proton collision. In that case, γ-ray photons would be pro- (ii) TheintegratedfluxdensityoftheSNRatthe325MHz
duced by proton–proton interaction, and synchrotron radio frequency is 1.84±0.15 Jy. Assuming a typical spectral
emission is not expected unless the secondary leptons pro- index of −0.5 for SNRs, we derive the integrated flux
duced by the decay of charged pions radiate significantly. density at the 1.4 GHz as 0.88 Jy. We note that this value
However, in order for this emission to be important, it re- is 1.5σ lower than the previously reported flux density
quiresrather high densities and magnetic fields. 2.2±0.9 Jy at 1.4 GHz (Tian et al. 2008). The change in
HESS J1729−345 was observed in both the 325 and radio fluxdensity changes theSED slightly.
1390 MHz bands. Radio emission is not seen in spatial co-
incidence with HESS J1729−345 in the 1390 MHz GMRT (iii) The spectral indices of the different filaments are
map. In the 325 MHz map, the diffuse radio emission to- estimated from the 325 and 610 MHz GMRT maps. The
wardsthewestoverlapstheregionofHESSJ1729−345 (see valuesof spectral indicesvaryfrom −1.11 to −0.15 and are
Fig. 2). The morphology of the emission does not seem to consistent with thenon-thermalradio emission.
becorrespondingtoHESSJ1729−345, whichis∼7arcmin
gaussianstructure.However,apossibleradiocounterpartof (iv) Wecomparetheradiobrightnessprofilewith theVHE
HESSJ1729−345isseeninthe843MHzMOST(seeFig.3) emission. It is particularly striking that the peak in γ-ray
andthe1.4GHzSGPSmaps.Thenon-detectioninthe1390 emission comes from the filament which is faintest in the
MHz GMRT map can be due to missing flux. Our GMRT radio, and the brightest filaments in the radio correspond
map at 1390 MHz is sensitive to angular scales up to 4 ar- to faint emission in γ-rays. The faintest filament in γ-rays
cmin. Thus significant flux will be missing while imaging a also shows a steep spectral index of −1.11±0.22. This
7arcminstructure.Weobtainaspectralindexof3.0±0.2, may be due to the effect of non-uniform magnetic field
fromthe843MHzMOSTmapandthe1.4GHzSGPSmap, strength, which is suggestive from the possible evidence
which is consistent with thermal emission. Thus the radio of synchrotron cooling in that region of the SNR. In this
emission from this possible counterpart is not synchrotron framework, the anti-correlated emission can be explained if
in nature. We note that Cui et al. (2016) explained VHE the VHE emission is of leptonic origin for an isotropic in-
emissionfromHESSJ1729−345inahadronicscenario.The jection. This kindof an anti-correlated emission isreported
positivespectralindexofthepossibleradiocounterpartsup- for thefirst time in VHEshell SNRs.
ports thisexplanation.
MNRAS000,1–9(2016)
Low Frequency Radio counterpart of HESS J1731-347 9
(v) The visual extent of the VHE emission appears to Greisen, E. W. 2003, inAndr´eHeck., ed., InformationHandling
be slightly beyond the radio shell. However, our radio inAstronomy-HistoricalVistas.vol.285,Dordrecht:Kluwer
observationsaremostlikelysensitivitylimited.Futureradio AcademicPublishers,2003.,p.109
observations with better sensitivity will probe the true Halpern,J.P.,&Gotthelf,E.V.2010a, ApJ,710,941
Halpern,J.P.,&Gotthelf,E.V.2010b, ApJ,725,1384
extent of the SNR and the upgraded GMRT observations
Haslam,C.G.T.,Salter,C.J.,Stoffel,H.,&Wilson,W.E.1982,
with 2.5times highersensitivity andanorderof magnitude
A&AS,47,1
larger bandwidthwill proveto beuseful for this.
Haverkorn,M.,Gaensler,B.M.,McClure-Griffiths,N.M.,Dickey,
J.M.,&Green,A.J.2006,ApJS,167,230
(vi) We also report a possible radio counterpart of HESS H.E.S.S. Collaboration, Abramowski, A., Acero, F., et al. 2011,
J1729−345 fromthe843MHzMOSTmapandthe1.4GHz A&A,531,A81
SGPS map. The radio emission is thermal and it supports H.E.S.S.Collaboration,Abdalla,H.,Abdalla,H.,etal.2016a,
thehadronicorigin of γ-raysfrom HESSJ1729−345. arXiv:1609.08671
H. E. S. S. Collaboration, Abdalla, H., Abramowski, A., et al.
2016b,arXiv:1611.01863
H.E.S.S.Collaboration,Abramowski,A.,Aharonian,F.,etal.
ACKNOWLEDGEMENTS
2016c,arXiv:1601.04461
Jones,T.W.,Rudnick,L.,Jun,B.-I.,etal.1998,PASP,110,125
We thank the anonymous referee for constructive inputs.
Landecker, T. L.,& Wielebinski, R. 1970, Australian Journal of
We acknowledge Nissim Kanekar for his support at vari-
PhysicsAstrophysicalSupplement, 16,1
ousstagesofthiswork.P.C.acknowledgessupportfromthe
Marcote,B.,Ribo´,M.,Paredes,J.M.,&Ishwara-Chandra,C.H.
Department of Science and Technology via SwaranaJayanti
2015,MNRAS,451,59
Fellowship award (file no.DST/SJF/PSA-01/2014-15). We Moriguchi,Y.,Tamura,K.,Tawara,Y.,etal.2005,ApJ,631,947
thank the staff of the GMRT that made these observations Neugebauer,G.,Habing,H.J.,vanDuinen,R.,etal.1984,ApJ,
possible.GMRTisrunbytheNationalCentreforRadioAs- 278,L1
trophysicsof the Tata Instituteof FundamentalResearch. Orlando,S.,Bocchino,F.,Reale,F.,Peres,G.,&Petruk,O.2007,
A&A,470,927
Parizot, E., Marcowith, A., Ballet, J., & Gallant, Y. A. 2006,
A&A,453,387
REFERENCES
Park,S.,Kargaltsev,O.,Pavlov,G.G.,etal.2009,ApJ,695,431
Acero,F.,Aharonian,F.,Akhperjanian,A.G.,etal.2010,A&A, Petruk,O.,Beshley,V.,Bocchino,F.,&Orlando,S.2009a,MN-
516,A62 RAS,395,1467
Acero, F., Pu¨hlhofer, G., Klochkov, D., et al. 2009, Petruk, O., Dubner, G., Castelletti, G., et al. 2009b, MNRAS,
arXiv:0907.0642 393,1034
Acero,F.,Lemoine-Goumard,M.,Renaud,M.,etal.2015,A&A, Reynolds,S.P.2008,ARA&A,46,89
580,A74 Roger,R.S.,Costain,C.H.,Landecker,T.L.,&Swerdlyk,C.M.
Aharonian, F. A.,Akhperjanian, A.G., Aye, K.-M.,et al. 2004, 1999,A&AS,137,7
Nature,432,75 Sirothia,S.K.2009,MNRAS,398,853
Aharonian, F., Akhperjanian, A. G., Bazer-Bachi, A. R., et al. Swarup, G., Ananthakrishnan, S., Kapahi, V. K., et al. 1991,
2006,A&A,449,223 CurrentScience, Vol.60,95.
Aharonian, F., Akhperjanian, A. G., Bazer-Bachi, A. R., et al. Tian, W. W., Leahy, D. A., Haverkorn, M., & Jiang, B. 2008,
2007,ApJ,661,236 ApJ,679,L85
Aharonian, F., Akhperjanian, A. G., Barres de Almeida, U., et Tian,W.W.,Li,Z.,Leahy,D.A.,etal.2010,ApJ,712,790
al.2008,A&A,477,353 Wright,E.L.,Eisenhardt,P.R.M.,Mainzer,A.K.,etal.2010,
Aharonian, F., Akhperjanian, A. G., de Almeida, U. B., et al. AJ,140,1868
Yang,R.-Z.,Zhang,X.,Yuan,Q.,&Liu,S.2014,A&A,567,A23
2009,ApJ,692,1500
Zoonematkermani,S.,Helfand,D.J.,Becker,R.H.,White,R.L.,
Bamba,A.,Pu¨hlhofer,G.,Acero,F.,etal.2012,ApJ,756,149
Basu, A., Mitra, D., Wadadekar, Y., & Ishwara-Chandra, C. H. &Perley,R.A.1990, ApJS,74,181
2012,MNRAS,419,1136
Blandford,R.,&Eichler,D.1987,Phys.Rep.,154,1
Blasi,P.2013, A&ARv,21,70
Brogan,C.L.,Frail,D.A.,Goss,W.M.,&Troland,T.H.2000,
ApJ,537,875
Castelletti, G., Supan, L., Dubner, G., Joshi, B. C., & Surnis,
M.P.2013,A&A,557,L15
Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ,
115,1693
Cui,Y.,Pu¨hlhofer,G.,&Santangelo, A.2016,A&A,591,A68
Degrange, B., & Fontaine, G. 2015, Comptes Rendus Physique,
16,587
Fukuda,T.,Yoshiike,S.,Sano,H.,etal.2014,ApJ,788,94
Gabici,S.,&Aharonian,F.A.2014,MNRAS,445,L70
Gabici,S.,&Aharonian,F.2016,EuropeanPhysicalJournalWeb
ofConferences,121,04001
Green, A. J., Cram, L. E., Large, M. I., & Ye, T. 1999, ApJS,
122,207
Green,D.A.2011,BulletinoftheAstronomicalSocietyofIndia,
39,289
MNRAS000,1–9(2016)