Discovery of gamma-ray emission from the shell-type supernova remnant RCW 86 with H.E.S.S
aa r X i v : . [ a s t r o - ph ] O c t Draft version October 23, 2018
Preprint typeset using L A TEX style emulateapj v. 08/13/06
DISCOVERY OF GAMMA-RAY EMISSION FROM THE SHELL-TYPE SUPERNOVA REMNANT RCW 86WITH H.E.S.S.
F. Aharonian , A.G. Akhperjanian , U. Barres de Almeida , A.R. Bazer-Bachi , B. Behera ,M. Beilicke , W. Benbow , K. Bernl¨ohr , C. Boisson , A. Bochow , V. Borrel , I. Braun , E. Brion ,J. Brucker , R. B¨uhler , T. Bulik , I. B¨usching , T. Boutelier , S. Carrigan , P.M. Chadwick ,A. Charbonnier , R.C.G. Chaves , L.-M. Chounet , A.C. Clapson , G. Coignet , L. Costamante , M.Dalton , B. Degrange , H.J. Dickinson , A. Djannati-Ata¨ı , W. Domainko , L.O’C. Drury , F. Dubois ,G. Dubus , J. Dyks , K. Egberts , D. Emmanoulopoulos , P. Espigat , C. Farnier , F. Feinstein ,A. Fiasson , A. F¨orster , G. Fontaine , M. F¨ußling , S. Gabici , Y.A. Gallant , L. G´erard ,B. Giebels , J.F. Glicenstein , B. Gl¨uck , P. Goret , C. Hadjichristidis , D. Hauser , M. Hauser ,G. Heinzelmann , G. Henri , G. Hermann , J.A. Hinton , A. Hoffmann , W. Hofmann , M. Holleran ,S. Hoppe , D. Horns , A. Jacholkowska , O.C. de Jager , I. Jung , K. Katarzy´nski , S. Kaufmann ,E. Kendziorra , M. Kerschhaggl , D. Khangulyan B. Kh´elifi , D. Keogh , Nu. Komin , K. Kosack ,G. Lamanna , I.J. Latham , M. Lemoine-Goumard b , J.-P. Lenain , T. Lohse , V. Marandon , J.M. Martin ,O. Martineau-Huynh , A. Marcowith , C. Masterson , D. Maurin , T.J.L. McComb , C. Medina ,R. Moderski , E. Moulin , M. Naumann-Godo , M. de Naurois , D. Nedbal , D. Nekrassov , J. Niemiec ,S.J. Nolan , S. Ohm , J-F. Olive , E. de O˜na Wilhelmi , K.J. Orford , J.L. Osborne , M. Ostrowski ,M. Panter , G. Pedaletti , G. Pelletier , P.-O. Petrucci , S. Pita , G. P¨uhlhofer , M. Punch ,A. Quirrenbach , B.C. Raubenheimer , M. Raue , S.M. Rayner , M. Renaud , F. Rieger J. Ripken ,L. Rob , S. Rosier-Lees , G. Rowell , B. Rudak , J. Ruppel , V. Sahakian , A. Santangelo ,R. Schlickeiser , F.M. Sch¨ock , R. Schr¨oder , U. Schwanke , S. Schwarzburg , S. Schwemmer ,A. Shalchi , J.L. Skilton , H. Sol , D. Spangler , L. Stawarz , R. Steenkamp , C. Stegmann ,G. Superina , P.H. Tam , J.-P. Tavernet , R. Terrier , O. Tibolla , C. van Eldik , G. Vasileiadis ,C. Venter , J.P. Vialle , P. Vincent , J. Vink , M. Vivier , H.J. V¨olk , F. Volpe , S.J. Wagner ,M. Ward , A.A. Zdziarski , A. Zech Draft version October 23, 2018
ABSTRACTThe shell-type supernova remnant (SNR) RCW 86, possibly associated with the historical supernovaSN 185, with its relatively large size (about 40’ in diameter) and the presence of non-thermal X-raysis a promising target for γ -ray observations. The high sensitivity, good angular resolution of a few arcminutes and the large field of view of the High Energy Stereoscopic System (H.E.S.S.) make it ideallysuited for the study of the γ -ray morphology of such extended sources. H.E.S.S. observations haveindeed led to the discovery of the SNR RCW 86 in very high energy (VHE; E >
100 GeV) γ -rays.With 31 hours of observation time, the source is detected with a statistical significance of 8 . σ andis significantly more extended than the H.E.S.S. point spread function. Morphological studies havebeen performed and show that the γ -ray flux does not correlate perfectly with the X-ray emission.The flux from the remnant is ∼
10% of the flux from the Crab nebula, with a similar photon index ofabout 2.5. Possible origins of the very high energy gamma-ray emission, via either Inverse Comptonscattering by electrons or the decay of neutral pions produced by proton interactions, are discussedon the basis of spectral features obtained both in the X-ray and γ -ray regimes. Subject headings: gamma-rays: observations – supernova remnants: general– supernova remnants:individual RCW 86 – H.E.S.S. * Correspondence and request for material should be sent [email protected] & [email protected] Max-Planck-Institut f¨ur Kernphysik, P.O. Box 103980, D69029 Heidelberg, Germany Yerevan Physics Institute, 2 Alikhanian Brothers St., 375036Yerevan, Armenia Centre d’Etude Spatiale des Rayonnements, CNRS/UPS,9 av. du Colonel Roche, BP 4346, F-31029 Toulouse Cedex 4,France Universit¨at Hamburg, Institut f¨ur Experimentalphysik, Luru-per Chaussee 149, D 22761 Hamburg, Germany Institut f¨ur Physik, Humboldt-Universit¨at zu Berlin, Newton-str. 15, D 12489 Berlin, Germany LUTH, Observatoire de Paris, CNRS, Universit´e ParisDiderot, 5 Place Jules Janssen, 92190 Meudon, France IRFU/DSM/CEA, CE Saclay, F-91191 Gif-sur-Yvette, Cedex,France University of Durham, Department of Physics, South Road,Durham DH1 3LE, U.K. Unit for Space Physics, North-West University, Potchefstroom 2520, South Africa Laboratoire Leprince-Ringuet, Ecole Polytechnique,CNRS/IN2P3, F-91128 Palaiseau, France Laboratoire d’Annecy-le-Vieux de Physique des Particules,CNRS/IN2P3, 9 Chemin de Bellevue - BP 110 F-74941 Annecy-le-Vieux Cedex, France Astroparticule et Cosmologie (APC), CNRS, Universite Paris7 Denis Diderot, 10, rue Alice Domon et Leonie Duquet, F-75205Paris Cedex 13, France Dublin Institute for Advanced Studies, 5 Merrion Square,Dublin 2, Ireland Landessternwarte, Universit¨at Heidelberg, K¨onigstuhl, D69117 Heidelberg, Germany Laboratoire de Physique Th´eorique et Astroparticules,CNRS/IN2P3, Universit´e Montpellier II, CC 70, Place Eug`eneBataillon, F-34095 Montpellier Cedex 5, France Universit¨at Erlangen-N¨urnberg, Physikalisches Institut,Erwin-Rommel-Str. 1, D 91058 Erlangen, Germany Laboratoire d’Astrophysique de Grenoble, INSU/CNRS,
F. A. Aharonian et al. (H.E.S.S. Collaboration) INTRODUCTION
Shell-type supernova remnants (SNR) are widely be-lieved to be the prime candidates for accelerating cosmicray protons and nuclei up to 10 eV. A promising way ofproving the existence of high energy hadrons acceleratedin SNR shells is the detection of very high energy (VHE;E >
100 GeV) γ -rays produced in nucleonic interactionswith ambient matter. VHE γ -ray emission has beendetected recently in several shell-type SNRs, especiallyfrom Cassiopeia A (Aharonian et al. (2001), Albert et al.(2007)), RX J1713.7-3946 (Aharonian et al. 2007a) andRX J0852.0-4622 (Aharonian et al. 2007b). These twolatest sources both show an extended morphology highlycorrelated with the structures seen in non-thermal X-rays. Although a hadronic origin is probable in the abovecases (e.g. Berezhko & V¨olk 2006), a leptonic origin cannot be ruled out (e.g. Porter et al. 2006).Another young shell-type SNR is RCW 86 (also known asG315.4-2.3 and MSH 14 − ). It has a complete shell inradio (Kesteven & Caswell 1987), optical (Smith 1997)and X-rays (Pisarski et al. 1984), with a nearly circu-lar shape of 40’ diameter. It received substantial atten-tion because of its possible association with SN 185, thefirst historical Galactic supernova (Clark & Stephenson1977). However, conclusive evidence for this connectionis still missing: using optical observations, Rosado et al.(1996) found an apparent kinematic distance of 2.8 kpcand an age of ∼
10 000 years, whereas recent observationsof the North-East part of the remnant with the Chan-dra and XMM-Newton satellites strengthen the case thatthe event recorded by the Chinese in 185 AD was a su-pernova and that RCW 86 is its remnant (Vink et al.2006). In this case, a distance to the SNR of ∼ Universit´e Joseph Fourier, BP 53, F-38041 Grenoble Cedex 9,France Institut f¨ur Astronomie und Astrophysik, Universit¨atT¨ubingen, Sand 1, D 72076 T¨ubingen, Germany LPNHE, Universit´e Pierre et Marie Curie Paris 6, Universit´eDenis Diderot Paris 7, CNRS/IN2P3, 4 Place Jussieu, F-75252,Paris Cedex 5, France Institute of Particle and Nuclear Physics, Charles University,V Holesovickach 2, 180 00 Prague 8, Czech Republic Institut f¨ur Theoretische Physik, Lehrstuhl IV: Weltraumund Astrophysik, Ruhr-Universit¨at Bochum, D 44780 Bochum,Germany University of Namibia, Private Bag 13301, Windhoek,Namibia Obserwatorium Astronomiczne, Uniwersytet Jagiello´nski,Krak´ow, Poland Nicolaus Copernicus Astronomical Center, ul. Bartycka 18,00-716 Warsaw, Poland School of Physics & Astronomy, University of Leeds, LeedsLS2 9JT, UK School of Chemistry & Physics, University of Adelaide,Adelaide 5005, Australia Toru´n Centre for Astronomy, Nicolaus Copernicus University,ul. Gagarina 11, 87-100 Toru´n, Poland Instytut Fizyki J¸adrowej PAN, ul. Radzikowskiego 152,31-342 Krak´ow, Poland European Associated Laboratory for Gamma-Ray Astron-omy, jointly supported by CNRS and MPG Astronomical Institute, Utrecht University, PO Box 80000,3508 TA Utrecht, The Netherlands a supported by CAPES Foundation, Ministry of Education ofBrazil b M. Lemoine-Goumard, Universit´e Bordeaux I, CNRS/IN2P3,Centre d’Etudes nucl´eaires de Bordeaux Gradignan, UMR 5797,Chemin du Solarium, 33175 Gradignan, France nario (Bocchino et al. 2000). The X-ray spectrum ob-tained with the Einstein satellite was first representedby a two-temperature plasma model (Winkler 1978).Then, RXTE (Petre et al. 1999) and ASCA observations(Bamba et al. (2000), Borkowski et al. (2001)), with awider spectral coverage, were used to resolve a non-thermal component in the X-ray spectrum which can bewell described by a soft power-law with a photon indexof ∼
3. The large-scale density gradient across RCW 86(Pisarski et al. (1984) and Claas et al. (1989)) possiblysuggests that the northern part could be the shockedhalf of a very low-density wind bubble plus dense shellfrom the progenitor star, and to this extent it could wellbe similar to RX J1713.7-3946 and RX J0852.0-4622.In its southern part, RCW 86 contains an HII region.Apparently, the gas density in this HII region is ratherhigh and spatially extended. Therefore, the SNR shockhas swept over an extended high density region in theSouth, with consequent high radio and thermal X-rayemissions (Bocchino et al. 2000). With a diameter ofabout 40’, RCW 86 is one of the very few non-thermal X-ray emitting SNRs resolvable in VHE γ -rays. H.E.S.S.,with its high sensitivity, its good angular resolution andits large field of view is ideally suited for morphologystudies of such an extended object.Evidence for γ -ray emission from RCW 86 was foundusing the CANGAROO-II instrument, but no firm de-tection was claimed (Watanabe et al. 2003). Here, wepresent data on RCW 86 obtained with the full H.E.S.S.array between 2004 and 2007. H.E.S.S. OBSERVATIONS AND ANALYSIS METHODS
H.E.S.S. is an array of four imaging Cherenkovtelescopes located 1800 m above sea level in theKhomas Highland in Namibia (Hinton 2004). Eachtelescope has a tesselated mirror with an area of107 m (Bernl¨ohr et al. 2003) and is equipped with acamera comprising 960 photomultipliers (Vincent et al.2003) covering a field of view of 5 ◦ in diameter. Due tothe effective rejection of hadronic air showers with thestereoscopic imaging technique, the H.E.S.S. telescopesystem can detect point sources near zenith at flux levelsof about 1% of the Crab nebula flux with a statisticalsignificance of 5 σ standard deviation in 25 hours of ob-servation (Aharonian et al. 2006).The shell-type SNR RCW 86 was observed between 2004and 2007 with the complete H.E.S.S. array. After stan-dard data quality selection and dead time correction, theresulting live time is 31 hours. The observations havebeen carried out at zenith angles ranging from 38 ◦ to53 ◦ . The data were taken using the wobble mode wherethe source is offset from the centre of the field of view, al-ternating between 28 minute runs in the positive and neg-ative declination or right ascension directions; the meanoffset angle of the data set used in this analysis is 0 . ◦ .The energy threshold of the system increases with zenithangle: for the observations presented here, the averagethreshold was 480 GeV.The data were calibrated using standard H.E.S.S. cal-ibration procedures, as discussed by Aharonian et al.(2004). The data were analyzed using a Hillas parame-ter based method as described in Aharonian et al. (2005)with standard cuts, which include a minimum require-ment of 80 photo electrons in each camera image. Twoiscovery of gamma-ray emission from the SNR RCW 86 with H.E.S.S. 3different background estimation procedures were used,as described in Berge et al. (2007). For 2D image gen-eration and morphology studies, the ring background method was applied with a mean ring radius of 0 . ◦ . Asthis method uses an energy averaged radial acceptancecorrection, the reflected-region background method wasapplied for spectral studies. In this second backgroundsubtraction procedure, OFF events were selected fromthe same field of view and in the same runs as the ONevents by selecting the region symmetric to the ON re-gion with respect to the camera centre. As a cross-check,a second analysis chain, sharing only the raw data andusing the “Combined Model” analysis (de Naurois et al.2005), was also applied to the data. The two analysismethods yield consistent results. RESULTS
A clear VHE γ -ray signal of 8 . σ standard deviationand 1546 ±
183 excess γ -rays is detected from a circularregion of 0 . ◦ radius, centered on ( α J = 14 h m s , δ J = − ◦ ′ ′′ ). This integration region was cho-sen a priori on the basis of the X-ray data obtained withthe ROSAT satellite and fully encompasses the SNR. Fig-ure 1 shows the VHE γ -ray excess map of the 1 . ◦ × . ◦ region around RCW 86. The map has been smoothedwith a Gaussian kernel with a σ of 4 . ′ to suppress sta-tistical fluctuations on scales smaller than the H.E.S.S.point-spread function (PSF). The VHE γ -ray excess fromRCW 86 is significantly extended beyond the PSF of theinstrument, which is illustrated in the bottom left cor-ner of Figure 1. Contours of constant significance aresuperimposed in white at the 4, 5 and 6 σ levels. Anexcess map has also been produced with the so-called“hard cuts” for better gamma hadron separation, whichincludes a stricter cut of 200 photo electrons on the im-age size compared to the “standard cuts”, and was foundto be compatible with Figure 1. The VHE emissionshown in Figure 1 is suggestive of a shell-like morphol-ogy. To test this hypothesis, the brightness profile of athick shell projected along the line of sight and foldedwith the H.E.S.S. point-spread function was fit to theunsmoothed excess map. As illustrated in Figure 1, thebest fit ( χ / ndf = 233.1/220) is obtained with an outerradius of 24 . ′ ± . ′ stat , a width of 12 . ′ ± . ′ stat anda centre of the shell at ( α J = 14 h m s ± . s stat , δ J = − ◦ ′ . ′′ ± . ′′ stat ).Figure 2 shows the radial profile of the VHE excessrelative to the fitted centre. The fit of the radialprofiles to the data points results in a chi-square perdegree of freedom of χ / ndf = 2.85/7 for a projectedshell (determined by outer ring radius, ring width andabsolute normalization) which is not significantly betterthan the fit of a projected uniformly-emitting spherecharacterized by a ring radius and a normalizationfactor ( χ / ndf = 5.43/8). Also visible in Figure 1 isan apparent deficit of γ -rays at the western part ofthe SNR. However, the azimuthal profile in Figure 3 isconsistent with a constant and reveals that this dip isnot significant ( χ / ndf = 1.47/5).Figure 4 shows the 3-6 keV X-ray map of RCW 86obtained using six observations of the remnant carriedout by the XMM-Newton satellite in 2006 (Vink et al.2006) and additional observations taken in 2007. The Fig. 1.— : H.E.S.S. γ -ray image of RCW 86. Themap was smoothed with a Gaussian function with a σ smooth = 4 . ′ to reduce the effect of statistical fluctua-tions. The linear color scale is in units of excess countsper arcmin . White contours correspond to 4, 5, 6 σ sig-nificance, obtained by counting gamma rays within 0.14 ◦ from each given location. The image inset in the bottomleft corner indicates the size of a point source as seenby H.E.S.S., for an equivalent analysis, smoothing andzenith angles. The centre of the fitted shell, as discussedin the text, is marked by a black cross. The two solidgreen circles correspond to the inner and outer radii ofthis shell. excess p e r a r c m i n H.E.S.S. excessShell FitSphere Fit [degr] q no r m . excess Fig. 2.— : Upper panel:
H.E.S.S. radial profiles aroundthe fitted centre of the SNR ( α J = 14 h m s , δ J = − ◦ Lowerpanel:
Radial profiles of the X-ray data (3-6 keV) fromXMM-Newton. These data are background subtractedand smoothed to match the H.E.S.S. angular resolution.Additionally, the obtained excess profile was normalized.energy range was selected to avoid as much as possiblecontamination from line emission from the, in general,cool plasma ( < excess p e r a r c m i n [degr] f no r m . excess Fig. 3.— : Upper panel:
H.E.S.S. azimuthal profileintegrated over a region of 0 . ◦ radius covering the SNRRCW 86. The azimuthal angle is calculated with re-spect to the fitted shell centre. 0 ◦ corresponds to theNorth part of the source and 90 ◦ to the East. The solidline shows the result of a fit of the data to a constantwhich yields a chi-square of 1.47 for 5 degrees of freedom. Lower panel:
Azimuthal profiles of the X-ray data (3-6keV) from XMM-Newton. These data are backgroundsubtracted and smoothed to match the H.E.S.S. angularresolution. Additionally, the obtained excess profile wasnormalized.Ar and Ca lines, but no such line emission is seenin the available Chandra, XMM-Newton (Vink et al.2006) or Suzaku spectra (Ueno et al. 2007). Thismap was obtained by first automatically cleaning theobservations of > σ excursions to the mean count rate,thus minimizing the background of the maps. Then,for each observation and for each of the three detectors(MOS1, MOS2, and PN), a background count rate inthe 3-6 keV band was determined using a relativelyempty region of the field of view. In the final stage,the background image was subtracted from the countrate image, and then corrected using the exposure mapsobtained with the standard XMM-Newton SAS 7.1.0software (which includes vignetting correction), in orderto obtain the background corrected map displayed inFigure 4. An overall positional agreement with theH.E.S.S. contours derived from Figure 1 as well as agood compatibility between the outer radius of the γ -ray emission (24 . ′ ± . ′ stat ) and the extensionof the X-ray emission can be observed. However, theemission peak apparent in the X-ray azimuthal profileis not visible in γ -rays (Figure 3). Furthermore, thedip in surface brightness at the center of the remnantseems more pronounced in the X-ray radial profiles(Figure 2). A more detailed comparison of the γ -rayand X-ray morphologies would require higher statisticsthan presently available, and hence will have to awaitfuture longer observations.For the spectral analysis, the source region (ON region)is defined by a circle of 0 . ◦ radius centered on the bestfit position of the shell, chosen to fully enclose the wholesource. The radius of the extraction region is illustratedin Figure 2. The spectrum obtained (see Figure 5) iswell described by a power-law with a photon index of2 . ± . stat ± . sys and a flux normalisation at 1 h m
45 44 43 42 41 40Right Ascension (J2000)-62 (cid:176) ¢ D e c li na t i on ( J ) Fig. 4.— : Excess contours of γ -ray emission (0.55,0.8, 1.05 γ -rays per arcmin Gaussian smoothed with σ smooth = 4 . ′ ) superimposed on the background sub-tracted XMM-Newton EPIC (MOS/PN) 3-6 keV X-rayimage of the remnant.TeV of (3 . ± . stat ± . sys ) × − cm − s − TeV − ( χ / ndf = 6.30/4). The integral flux in the energy range1 - 10 TeV is (2 . ± . stat ± . sys ) × − cm − s − ,which corresponds to ∼
10% of the integrated flux of theCrab nebula in the same energy interval. No significantimprovement is obtained by fitting a power-law with anexponential cut-off ( χ / ndf = 2.96/3). If the fit rangeis restricted to energies below 10 TeV, a photon indexof 2 . ± . stat ± . sys and a flux normalisation at 1TeV of (3 . ± . stat ± . sys ) × − cm − s − TeV − are determined ( χ / ndf = 0.68/2), compatible with thefit of the SNR in the whole energy range. Energy (TeV)1 10 ) - s - c m - d N / d E ( T e V -17 -16 -15 -14 -13 -12 -11 Energy (TeV) F / F D R e s i du a l s - -202 Fig. 5.— : Differential energy spectrum of RCW 86,extracted from a circular region of 0 . ◦ radius aroundthe position ( α J = 14 h m s , δ J = − ◦ σ statistical errors; the upperlimit (arrow) is estimated at the 2 σ level. The bottompanel shows the residuals to the power-law fit. Eventswith energies between 600 GeV and 60 TeV were used inthe determination of the spectrum.iscovery of gamma-ray emission from the SNR RCW 86 with H.E.S.S. 5 DISCUSSION
There are two commonly invoked mechanisms for VHE γ -ray production in young supernova remnants, inverseCompton (IC) scattering of high energy electrons off am-bient photons (leptonic scenario) and π meson produc-tion in inelastic interactions of accelerated protons withambient gas (hadronic scenario). In such a hadronic sce-nario, a comparison between the expected thermal X-rayemission and the actually measured thermal emission hasto await deeper observations in which one can better de-termined whether the TeV emission traces the denser,thermal X-ray emitting parts of the SNR, or is moreclosely correlated with the X-ray synchrotron emissionfrom the remnant.The measured γ -ray spectrum from RCW 86, restrictedto energies below 10 TeV, translates into an energyflux between 1 and 10 TeV of 8 . × − erg cm − s − .The X-ray spectrum of the whole remnant is mixed be-tween thermal and non-thermal emission. Assuming thatthe hard X-ray continuum originates from non-thermalsynchrotron emission as reported by Rho et al. (2002),Vink et al. (2006) and Ueno et al. (2007), the measure-ment made by Petre et al. (1999) using RXTE data pro-vides an estimate of the total amount of non-thermal fluxfrom RCW 86. They find that the spectrum is well fittedby a power-law of index ∼ − cm − s − keV − , which extrapolateddown to the 0.7 to 10 keV band leads to an integral fluxof 2 . × − erg cm − s − . In a leptonic scenario, as-suming that the γ -ray emission is entirely due to the ICprocess on cosmic microwave background photons, theratio of the synchrotron power and IC power radiated isoften used to constrain the magnetic field. For a power-law distribution of electron energies, Kγ − p , the generalequation relating the synchrotron power ( P S ) producedby electrons with Lorentz factors between γ ,X and γ ,X and the IC power ( P IC ) radiated between γ ,IC and γ ,IC can be expressed as follow: P S P IC = U B U ph ( γ − p ,X − γ − p ,X )( γ − p ,IC − γ − p ,IC ) (1)where U ph and U B are the energy density of the pho-ton field and the energy density of the magnetic field,respectively. It should be noted here that, for a fixedX-ray energy, γ ,X and γ ,X are inversely proportionalto the square root of the magnetic field. If X-rays and γ -rays probe the same region of the electron spectrum,one finds the standard relation between the synchrotronand IC power P S P IC = U B U ph . Assuming that the target pho-ton field is the cosmic microwave background, a magneticfield of 30 µ G can be estimated using Equation 1 and thesynchrotron photon index of ∼
3, independent of the dis-tance and age of the SNR. This estimate is compatiblewith that of Vink et al. (2006) based on thin filamentsresolved by Chandra (assuming a distance of 2.5 kpc)in which the authors also deduce a high speed of theblast wave ( ∼ − ); their estimated value wouldincrease to ∼ µ G for a distance of 1 kpc. However,it is still a factor of 2 lower than the maximum fieldstrength determined by V¨olk et al. (2005) using a lowershock velocity of 800 km s − as suggested by optical datain the Southern region of the SNR (Rosado et al. 1996).The difference between the field amplification estimated by Vink et al. (2006) and that of V¨olk et al. (2005) liesin the fact that V¨olk et al. obtained a higher result whenthey de-projected the measured filament width, as for anideal spherical shock, whereas Vink et al. did not. With-out de-projection the two results remarkably agree, eventhough they were obtained for the southern side and thenorthern side, respectively. A discussion of de-projectionfor RCW 86 is given in V¨olk et al. (2005). With similardata, Bamba et al. (2005) deduced a significantly lowermagnetic field strength of ∼ − µ G. However, theiranalysis is based on rather different assumptions on thenature of filament formation.In a hadronic scenario, one can estimate the total energyin accelerated protons W p in the range 10 −
100 TeVrequired to produce the γ -ray luminosity L γ observedby H.E.S.S. using the relation W p (10 −
100 TeV) ≈ τ γ × L γ (1 −
10 TeV), in which τ γ ≈ . × (cid:0) n − (cid:1) − sis the characteristic cooling time of protons through the π production channel (Kelner et al. 2006). The totalenergy injected in protons is calculated by extrapolatingthe proton spectrum down to 1 GeV. Because of this ex-trapolation over 4 decades in energy, the uncertainty ofthe estimate can be as large as a factor of 10. Assum-ing that the relatively steep slope of the proton spec-trum (as inferred from the observed γ -ray spectrum) isthe result of an energy cut-off (somewhere around sev-eral tens of TeV in proton energy), and that at lowerenergies the proton spectrum has a E − type spectrumrepresentative of those predicted by the diffusive shockacceleration theory, the total energy budget in all protonsfor the distance of 2.5 kpc and the ambient gas densitybetween 0 . − and 0 . − (Bocchino et al. 2000),would be (2 − × erg. This estimate is in rea-sonable agreement with theoretical expectations that asignificant fraction of the explosion energy of 10 erg isreleased in relativistic protons. On the other hand, ifthe power-law spectrum of protons continues to GeV en-ergies with the spectral index Γ = 2 . erg for adistance of 2.5 kpc. This would exclude the hadronicorigin of TeV γ -rays, unless the SNR is nearby ( ∼ γ -rays are produced in very dense regions.Indeed, Pisarski et al. (1984) and Claas et al. (1989) re-ported that there is a large density contrast across theremnant, e.g. in the South, where the density could beas high as 10 cm − ; with such a dense medium, a largerdistance for the remnant could still be compatible withthe observed γ -ray flux. CONCLUSIONS
H.E.S.S. observations have led to the discovery of theshell-type SNR RCW 86 in VHE γ -rays. The γ -ray signalis significantly more extended than the H.E.S.S. point-spread function. The possibility of a shell-like morphol-ogy was addressed, but cannot be settled on the basisof the limited statistics available at the moment. Theflux from the remnant is ∼
10% of that from the Crabnebula, with a photon index of about 2.5. The questionof the nature of the particles producing the γ -ray signalobserved by H.E.S.S. is also discussed.In a leptonic scenario, assuming that the γ -ray emissionis entirely due to the IC process on cosmic microwavebackground photons and that the synchrotron and IC F. A. Aharonian et al. (H.E.S.S. Collaboration)photons are produced by the same electrons, the ratio ofthe γ -ray energy flux and the X-ray flux determines themagnetic field to be close to 30 µ G.In the hadronic scenario, the lack of information aboutthe low-energy γ -ray spectrum results in large uncertain-ties on the total energy budget in protons. If below sev-eral tens of TeV, the proton spectrum has a E − typespectrum, the total energy in protons would be in rea-sonable agreement with theoretical expectations. On theother hand, if we assume that the proton spectrum con-tinues down to GeV energies with the observed spectralindex Γ = 2 .
4, energetics would rule out a hadronic ori-gin for the TeV γ -rays unless the SNR is nearby, or if the γ -rays are produced in a very dense medium as reportedin the southern part of the remnant.The support of the Namibian authorities and of the University of Namibia in facilitating the constructionand operation of H.E.S.S. is gratefully acknowledged,as is the support by the German Ministry for Edu-cation and Research (BMBF), the Max Planck Soci-ety, the French Ministry for Research, the CNRS-IN2P3and the Astroparticle Interdisciplinary Programme of theCNRS, the U.K. Science and Technology Facilities Coun-cil (STFC), the IPNP of the Charles University, the Pol-ish Ministry of Science and Higher Education, the SouthAfrican Department of Science and Technology and Na-tional Research Foundation, and by the University ofNamibia. We appreciate the excellent work of the tech-nical support staff in Berlin, Durham, Hamburg, Hei-delberg, Palaiseau, Paris, Saclay, and in Namibia in theconstruction and operation of the equipment. REFERENCESAharonian, F., 2001, A&A, 112, 307Aharonian, F., (
H.E.S.S. Collaboration ) 2004, APh, 22, 109Aharonian, F., et al. (
H.E.S.S. Collaboration ) 2005, A&A, 430, 865Aharonian, F., et al. (
H.E.S.S. Collaboration ) 2006, A&A, 457, 899Aharonian, F., et al. (
H.E.S.S. Collaboration ) 2007a, A&A, 464,235Aharonian, F., et al. (
H.E.S.S. Collaboration ) 2007b, A&A, 661,236Albert, J., et al. 2007, A&A, 474, 937Bamba, A., Koyama, K., & Tomida, H., 2000, PASJ, 52, 1157Bamba, A., Yamazaki, R., Yoshida, T., Terasawa, T., & Koyama,K., 2005, ApJ, 621, 793Berge, D., Funk, S., & Hinton, J., 2007, A&A, 466, 1219Berezhko, E. G., & V¨olk, H. J., 2006, A&A, 451, 981Bernl¨ohr, K., et al., 2003, APh, 20, 111Bocchino, F., Vink, J., Favata, F., Maggio, A., & Sciortino, S.,2000, A&A, 360, 671Borkowski, K. J., Arnaud, K. A., Dorman, B., Hughes, J. P.,Sarazin, C. L., & Smith, R. A., 2001, ApJ, 550, 334Claas, J.J., Kaastra, J. S., Smith, A., Peacock, A., & de Korte, P.A. J., 1989, ApJ, 337, 399Clark, D., & Stephenson, F., 1977, The Historical Supernovae(Oxford: Pergamon Press), 83Hinton, J. A., 2004, NewAR, 48, 331Kelner, S. R., Aharonian, F. A., & Bugayov, V. V., 2006, PhysicalReview D, 74, 3Kesteven, M. J., & Caswell, J. L., 1987, A&A, 183, 118de Naurois, M. et al. 2005, in Proceedings of the conference“Towards a Network of Atmospheric Cherenkov Detectors VII”,ed. B. Degrange & G. Fontaine (Palaiseau: Ecole Polytechnique),173 Petre, R., Allen, G. E., & Hwang, U., 1999, Astron. Nachr., 320,199Pisarski, P. L., Helfand, D. J., & Kahn, S. M., 1984, ApJ, 277, 710Porter, T. A., Moskalenko, I. V., & Strong, A. W. 2006, ApJ, 648,L29Rho, J., Dyer, K. K., Borkowski, K. J., & Reynolds, S. P., 2002,ApJ, 581, 1116Rosado, M., Ambrocio-Cruz, P., Le Coarer, E., & Marcelin, M.,1996, A&A, 315, 243Smith, R. C., 1997, AJ, 114, 2664Ueno, M., et al. 2007, PASJ, 59, 171Vincent, P., et al. 2003, in Proceedings of the 28th InternationalCosmic Ray Conference, T. Kajita et al., Eds. (UniversalAcademy Press, Tokyo, 2003), 2887Vink, J., Bleeker, J., Van Der Heyden, K., Bykov, A., Bamba, A.,& Yamazaki, R., 2006, ApJL, 648, 33V¨olk, H. J., Berezhko, E. G., & Ksenofontov, L. T., 2005, A&A,433, 229Watanabe, S., et al. (