Astrophysics and Space Science - Startseite...Astrophysics and Cosmology Stanford University CA...

Josep M. Paredes • Olaf Reimer • Diego F. TorresEditors

Astrophysics and Space Science

The Multi-Messenger Approach to High-EnergyGamma-Ray Sources

Third Workshop on the Nature of Unidentified High-Energy Sources

Reprinted from Astrophysics and Space ScienceVolume 309, Nos. 1–4, 2007

Josep M. ParedesDep. AstronomiaFac. FisiciaMarti i Franques 108028, BarcelonaSpain

Diego F. TorresInstitució de L’espai (IEEC-CSIC)Facultat de CienciesUniversitat Autònoma de BarcelonaTorre C5 Parell2a planta08193 BarcelonaSpain

Olaf ReimerW.W. Hansen Experimental PhysicsLaboratory & Kavli Institute for ParticleAstrophysics and CosmologyStanford UniversityCA 94305-4085USA

Cover illustration: The Crab Nebula as seen by the Hubble Space Telescope, an artistic view of a microquasar, and a cascade of a high-energy particlefrom a quasar jet entering the Earth atmosphere. The instruments are imaging atmospheric Cherenkov telescopes, submarine neutrino detectors, and thegamma-ray satellite GLAST.

Astronomy Subjects Classification (2007): SCP22006 Astronomy, Astrophysics and Cosmology; SCP22014 Astronomy; SCP22022 Astrophysics

Library of Congress Control Number: 2007930648

ISBN: 978-1-4020-6117-2 e-ISBN: 978-1-4020-6118-9

Printed on acid-free paper.

© 2007 Springer Science+Business Media, LLC. All rights reserved. This work may not be translated or copied in whole or in part without the writtenpermission of the publisher (Springer Science+Business Media, LLC., 233 Spring Street, New York, NY 10013, USA), except for brief excerpts inconnection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computersoftware, or by similar or dissimilar methodology now known or hereafter developed is forbidden.The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as anexpression of opinion as to whether or not they are subject to proprietary rights.

9 8 7 6 5 4 3 2 1

springer.com

Contents

Josep M. Paredes, Olaf Reimer and Diego F. Torres / Preface 1

Malcolm Longair / Prologue 3–4

Session 1: Global Properties of Gamma-Ray Sources

Roland Walter / New INTEGRAL sources and TeV emission 5–9

Stefan Funk / Status of identification of VHE γ -ray sources 11–16

Diego F. Torres, Shu Zhang, Olaf Reimer, Xavier Barcons, Amalia Corral, Valentí Bosch-Ramon, JosepM. Paredes, Gustavo E. Romero, Jin Qu, Werner Collmar, Volker Schönfelder and Yousaf Butt /INTEGRAL/XMM views on the MeV source GRO J1411-64 17–21

Shu Zhang and Werner Collmar / Evidence for a new MeV source observed by the COMPTEL experimentaboard CGRO 23–27

R. Mukherjee, E.V. Gotthelf and J.P. Halpern / Transient X-ray sources in the field of the unidentifiedgamma-ray source TeV J2032+4130 in Cygnus 29–33

A.W. Strong / Source population synthesis and the Galactic diffuse gamma-ray emission 35–41

Jennifer M. Siegal-Gaskins, Vasiliki Pavlidou, Angela V. Olinto, Carolyn Brown and Brian D. Fields /Population studies of the unidentified EGRET sources 43–49

Analía N. Cillis, Olaf Reimer and Diego F. Torres / Gamma-ray source stacking analysis at low galacticlatitudes 51–55

Olaf Reimer and Diego F. Torres / Identification of high energy gamma-ray sources and source populationsin the era of deep all-sky coverage 57–62

Session 2: Extragalactic Sources

Paolo Padovani / The blazar sequence: validity and predictions 63–71

Takuro Narumoto and Tomonori Totani / Gamma-ray luminosity function of blazars and the cosmicgamma-ray background: evidence for the luminosity-dependent density evolution 73–79

Vasiliki Pavlidou, Jennifer M. Siegal-Gaskins, Carolyn Brown, Brian D. Fields and Angela V. Olinto /Unidentified EGRET sources and the extragalactic gamma-ray background 81–87

Carlotta Pittori, Elisabetta Cavazzuti, Sergio Colafrancesco and Paolo Giommi / Blazar duty-cycle at γ -rayfrequencies: constraints from extragalactic background radiation and prospects for AGILE andGLAST 89–94

Markus Böttcher / Modeling the emission processes in blazars 95–104

Julian Sitarek and Wlodek Bednarek / Internal absorption of gamma-rays in relativistic blobs of activegalactic nuclei 105–109

D. Bose, V.R. Chitnis, P.R. Vishwanath, P. Majumdar, M.A. Rahman, B.B. Singh, A.C. Gupta and B.S.Acharya / Observations of AGNs using PACT 111–117

Frank M. Rieger, Valentí Bosch-Ramon and Peter Duffy / Fermi acceleration in astrophysical jets 119–125

Charles D. Dermer / High-energy cosmology 127–137

Matthias Beilicke / HESS observations of extragalactic objects 139–145

D. Pérez-Ramírez, J.R. Sánchez-Sutil, A. Muñoz-Arjonilla, J. Martí, J.L. Garrido and P. Luque-Escamilla /FIRST radio counterpart candidates to ULXs: a catalogue 147–150

Shoko Miyake, Shohei Yanagita and Tatsuo Yoshida / Numerical studies on the structure of the cosmic rayelectron halo in starburst galaxies 151–155

Evgeny V. Derishev / Synchrotron emission in the fast cooling regime: which spectra can be explained? 157–161

Sandip K. Chakrabarti and Samir Mandal / Spectral properties of shocked accretion flows—a self-consistentstudy 163–166

Paul Dempsey and Peter Duffy / Particle acceleration at multiple internal relativistic shocks 167–171

Saša Simic, Luka C. Popovic, Michael I. Andersen and Lise Christensen / Mechanism of light curvevariability in the gamma ray bursts 173–177

Session 3: Pulsars, PWN and Isolated Neutron Stars

W. Bednarek / High energy processes in pulsar wind nebulae 179–187

D. Horns, F. Aharonian, A.I.D. Hoffmann and A. Santangelo / Nucleonic gamma-ray production in pulsarwind nebulae 189–195

Yves A. Gallant / Associations of very high energy gamma-ray sources discovered by H.E.S.S. with pulsarwind nebulae 197–202

Olaf Reimer and Stefan Funk / Demystifying an unidentified EGRET source by VHE gamma-rayobservations 203–207

N. La Palombara, R.P. Mignani, E. Hatziminaoglou, M. Schirmer, G.F. Bignami and P. Caraveo /Multiwavelength observations of the two unidentified EGRET sources 3EG J0616-3310 and 3EGJ1249-8330 209–213

A.I.D. Hoffmann, D. Horns and A. Santangelo / INTEGRAL observations of TeV plerions 215–219

Alice K. Harding, Isabelle A. Grenier and Peter L. Gonthier / The Geminga fraction 221–230

Łukasz Bratek and Marcin Kolonko / An algorithm for solving the pulsar equation 231–234

Massimiliano Razzano / Pulsar simulation tools for GLAST 235–239

Massimiliano Razzano and Alice K. Harding / Pulsar sensitivity studies of the GLAST large area telescope 241–244

Peter L. Gonthier, Sarah A. Story, Brian D. Clow and Alice K. Harding / Population statistics study of radioand gamma-ray pulsars in the Galactic plane 245–251

Andrii Neronov and Maria Chernyakova / Radio-to-TeV γ -ray emission from PSR B1259–63 253–259

Dmitry Khangulyan, Slavomir Hnatic and Felix Aharonian / Modulation of the high energy γ -ray flux fromPSR B1259-63/SS2883 due to the orbital variation of the maximum energy of accelerated electrons 261–265

Session 4: Microquasars and Galactic Black Hole Sources

I.F. Mirabel / Gamma-ray binaries 267–270

Frank M. Rieger / Supermassive binary black holes among cosmic gamma-ray sources 271–275

Mathieu de Naurois / H.E.S.S. observations of LS 5039 277–284

Javier Rico / Results from MAGIC’s first observation cycle on galactic sources 285–291

P. Goldoni, M. Ribó, T. Di Salvo, J.M. Paredes, V. Bosch-Ramon and M. Rupen / INTEGRAL serendipitousdetection of the gamma-ray microquasar LS 5039 293–297

A. Smith, R.W. Atkins, S. Bradbury, O. Celik, Y.C.K. Chow, P. Cogan, C. Dowdall, S.J. Fegan, P. Fortin,D. Gall, G.H. Gillanders, J. Grube, K.J. Gutierrez, T.A. Hall, D. Hanna, J. Holder, D. Horan, S.B.Hughes, T.B. Humensky, I. Jung, P. Kaaret, G. Kenny, M. Kertzman, D.B. Kieda, A. Konopelko,H. Krawczynski, F. Krennrich, M.J. Lang, S. Le Bohec, G. Maier, J. Millis, P. Moriarty, R.A. Ong,J.S. Perkins, K. Ragan, G.H. Sembroski, J.A. Toner, L. Valcarcel, V.V. Vassiliev, R.G. Wagner, S.P.Wakely, T.C. Weekes, R.J. White and D.A. Williams / Whipple telescope observations of LS I +61303: 2004–2006 299–303

Samir Mandal and Sandip K. Chakrabarti / Spectral fit of Cygnus X-1 in high energy—a self-consistentstudy 305–308

J. Martí, D. Pérez-Ramírez, P. Luque-Escamilla, J.L. Garrido, J.M. Paredes, A. Muñoz-Arjonilla and J.R.Sánchez-Sutil / The search for hot spots associated with the Cygnus X-3 relativistic jet 309–313

Gloria Sala, Jochen Greiner, Eugenio Bottacini and Frank Haberl / The black-hole candidate XTE J1817-330as seen by XMM-Newton and INTEGRAL 315–319

Valentí Bosch-Ramon / Theoretical overview on high-energy emission in microquasars 321–331

M. Orellana and G.E. Romero / High-energy gamma-ray emission from the inner jet of LS I +61 303: thehadronic contribution revisited 333–338

P. Bordas, J.M. Paredes, V. Bosch-Ramon and M. Orellana / Secondary leptons synchrotron emission frommicroquasar jets 339–343

Session 5: Stars, SNRs and Molecular Clouds

Diego F. Torres and Eva Domingo-Santamaría / Collective effects of stellar winds and unidentifiedgamma-ray sources 345–350

Anita Reimer, Olaf Reimer and Martin Pohl / Gamma rays from colliding winds of massive stars 351–357

E. Orlando and A.W. Strong / Gamma rays from halos around stars and the Sun 359–363

Stefano Gabici, Felix A. Aharonian and Pasquale Blasi / Gamma rays from molecular clouds 365–371

Alberto Carramiñana / 3EG J2020+4017, the γ -Cygni source—before GLAST 373–378

M. Lemoine-Goumard, F. Aharonian, D. Berge, B. Degrange, D. Hauser, N. Komin, O. Reimer andU. Schwanke / Morphological and spectral studies of the shell-type supernova remnants RXJ1713.7–3946 and RX J0852.0–4622 with H.E.S.S. 379–384

E.G. Berezhko, L.T. Ksenofontov and H.J. Völk / Gamma-ray emission expected from Kepler’s SNR 385–388

H.J. Völk, E.G. Berezhko and L.T. Ksenofontov / New evidence for strong nonthermal effects in Tycho’ssupernova remnant 389–394

Jorge A. Combi, Juan F. Albacete Colombo, Gustavo E. Romero and Paula Benaglia / Hard X-ray emissionfrom the SNR G337.2+0.1 395–399

T. Ogasawara, T. Yoshida, S. Yanagita and T. Kifune / A possible explanation of photon emission fromsupernova remnants by jitter radiation 401–405

Session 6: Multi-Messenger Connections to Gamma-Ray Astrophysics

Francis Halzen / Cosmic neutrinos from the sources of galactic and extragalactic cosmic rays 407–414

C. Distefano / Detection potential to point-like neutrino sources with the NEMO-km3 telescope 415–420

Markus Ackermann / Upper limits on neutrino fluxes from point-like sources with AMANDA-II 421–427

C. Stegmann, A. Kappes, J. Hinton and F. Aharonian / Potential neutrino signals in a northern hemisphereneutrino telescope from galactic gamma-ray sources 429–433

M. Amenomori, S. Ayabe, X.J. Bi, D. Chen, S.W. Cui, Danzengluobu, L.K. Ding, X.H. Ding, C.F. Feng,Zhaoyang Feng, Z.Y. Feng, X.Y. Gao, Q.X. Geng, H.W. Guo, H.H. He, M. He, K. Hibino, N. Hotta,Haibing Hu, H.B. Hu, J. Huang, Q. Huang, H.Y. Jia, F. Kajino, K. Kasahara, Y. Katayose, C. Kato,K. Kawata, Labaciren, G.M. Le, A.F. Li, J.Y. Li, H. Lu, S.L. Lu, X.R. Meng, K. Mizutani, J. Mu, K.Munakata, A. Nagai, H. Nanjo, M. Nishizawa, M. Ohnishi, I. Ohta, H. Onuma, T. Ouchi, S. Ozawa,J.R. Ren, T. Saito, T.Y. Saito, M. Sakata, T.K. Sako, T. Sasaki, M. Shibata, A. Shiomi, T. Shirai,H. Sugimoto, M. Takita, Y.H. Tan, N. Tateyama, S. Torii, H. Tsuchiya, S. Udo, B. Wang, H. Wang,X. Wang, Y.G. Wang, H.R. Wu, L. Xue, Y. Yamamoto, C.T. Yan, X.C. Yang, S. Yasue, Z.H. Ye,G.C. Yu, A.F. Yuan, T. Yuda, H.M. Zhang, J.L. Zhang, N.J. Zhang, X.Y. Zhang, Y. Zhang, Yi Zhang,Zhaxisangzhu and X.X. Zhou / Underground water Cherenkov muon detector array with the Tibetair shower array for gamma-ray astronomy in the 100 TeV region 435–439

Gabrielle Lelaizant / Study on the possible detection of Gamma Ray Bursts with the ANTARES neutrinotelescope 441–445

M.C. González-García, Michele Maltoni and Joan Rojo / Determination of the atmospheric neutrino fluxfrom experimental data 447–451

Veniamin Berezinsky / On origin of ultra high energy cosmic rays 453–463

Stefano Gabici and Felix A. Aharonian / Gamma ray signatures of ultra high energy cosmic ray accelerators:electromagnetic cascade versus synchrotron radiation of secondary electrons 465–469

K. Jedrzejczak, M. Kasztelan, L. Mankiewicz, M. Molak, K. Nawrocki, L.W. Piotrowski, M. Sokołowski,B. Szabelska, J. Szabelski, T. Wibig, A.W. Wolfendale and G. Wrochna / Search for correlations ofGRB and cosmic rays 471–475

Session 7: The Gamma-Ray Horizon

Manel Martinez / Cosmology and VHE gamma ray astrophysics: connections and perspectives 477–485

Luigi Costamante / A low density of the extragalactic background light revealed by the H.E.S.S. spectra ofthe BL Lac objects 1ES 1101-232 and H 2356-309 487–495

Daniel Mazin / Observations of extragalactic sources with the MAGIC telescope 497–503

Session 8: Dark Matter and Gamma-Ray Astrophysics

Gianfranco Bertone / Dark matter: the connection with gamma-ray astrophysics 505–515

Erica Bisesi / The impact of subhalos on the gamma-ray signal from dark matter annihilation 517–522

Session 9: Instruments and Facilities for Studying Gamma-Ray Sources

O. Reimer, P.F. Michelson, R.A. Cameron, S.W. Digel, D.J. Thompson and K.S. Wood / GLAST large areatelescope multiwavelength planning 523–526

Alberto Carramiñana, and The LMT-GTM collaboration / Unravelling unidentified γ -ray sources with thelarge millimeter telescope 527–530

M. Cwiok, W. Dominik, K. Małek, L. Mankiewicz, J. Mrowca-Ciułacz, K. Nawrocki, L.W. Piotrowski,P. Sitek, M. Sokołowski, G. Wrochna and A.F. Zarnecki / Search for GRB related prompt opticalemission and other fast varying objects with “Pi of the Sky” detector 531–535

G. Di Sciascio and T. Di Girolamo / GRBs search results with the ARGO-YBJ experiment operated in scalermode 537–540

Kinya Hibino, Toshisuke Kashiwagi, Shoji Okuno, Kaori Yajima, Yukio Uchihori, Hisashi Kitamura,Takeshi Takashima, Mamoru Yokota and Kenji Yoshida / The design of diamond Compton telescope 541–544

Astrophys Space Sci (2007) 309: 1DOI 10.1007/s10509-007-9518-4

Preface

Josep M. Paredes · Olaf Reimer · Diego F. Torres

Published online: 15 May 2007© Springer Science+Business Media B.V. 2007

More than one year ago we were pleased to announce theconference “The multi-messenger approach to high energygamma-ray sources” which was held in Barcelona, Spain,from Tuesday, July 4th to Friday, July 7th, 2006. This bookcollects its refereed proceedings. The conference was at-tended by more than a hundred scientists from a dozen coun-tries.

As motivation for such a meeting, and particular forthe multi-messenger approach introduced into the studyof high energy gamma-ray sources, we noted that for thefirst time in history, we are on the verge to simultane-ously observe the most energetic phenomena in the Uni-verse from radio to TeV photons, cosmic rays, and neutri-nos, with roughly similar sensitivity and angular resolution.The energy band between 20 and 200 GeV will be acces-sible by upcoming satellites (such as GLAST) and alreadyoperational ground-based telescopes (such as MAGIC andHESS). Ultra high energy cosmic ray detections are be-ing used to investigate whether they violate the GZK cut-off imposed by the cosmic background radiation, whetherwe can identify their origin, or whether we must enlargephysics by admitting new universal constituents or inter-actions. Neutrino astrophysics is reaching at the same timediscovery maturity, while new powerful equipments are cur-

rently being built to scrutinize the sky with these messen-gers.

Observationally driven high energy astrophysics thuscalls for detailed theoretical multi-messenger based model-ing of plausible sources of high energy radiation, for mul-tiwavelength observations, and for detailed population stud-ies. Upon these topics, the lectures and lively debates held atthe conference went along, pivoting on the prospects for theidentification and study of the classes and individual highenergy gamma-ray sources that are still undetected or hid-den as yet unidentified sources.

This workshop continued the series initiated by the meet-ing held at Tonantzintla in October 2000, followed by theconference at Hong Kong in May 2004. We have had ahighly focused, profitable meeting at the frontier of highenergy astrophysics, with a lot of discussion where expertsfrom different research fields gathered for a 4-days intenseexchange in the Mediterranean environment of the city ofBarcelona. We hope that in future, the pages of this book willbring along the scent of those days, and memories to thosewho were present, while accompanying all readers into newscientific challenges.

Barcelona, November 7th, 2006

Astrophys Space Sci (2007) 309: 3–4DOI 10.1007/s10509-007-9525-5

O R I G I NA L A RT I C L E

Prologue

Malcolm Longair

Received: 5 January 2007 / Accepted: 8 May 2007 / Published online: 24 May 2007© Springer Science+Business Media B.V. 2007

A recurring theme of the history of astronomy, astrophysicsand cosmology is that, every time a new waveband is openedup for astronomical observation, new and unexpected re-sults are found which not only change the perceptions ofastronomers, but also point the way to the next generationof challenges for astronomers and technologists. The newinsights need to be placed in the broader astronomical per-spective, which may need to change in response to these dis-coveries. The proceedings of this conference, “The Multi-Messenger Approach to High Energy Gamma-ray Sources,”provide startling evidence of this process in action in thefield of Gamma-ray astronomy.

The meeting concerned many different aspects of whatcan be called VERY high energy astrophysics in its broadestsense. Much of the discussion centred upon direct means ofinvestigating some of the most extreme phenomena in con-temporary astrophysics. Gamma-ray astrophysics was at theheart of the conference and this discipline has been reinvig-orated by space missions such as INTEGRAL and SWIFTand by the ground-based large Cherenkov Arrays such asHESS, all of which are now producing spectacular science.These studies are complemented by new studies of the high-est energy cosmic rays from the first results from the AugerArray in Argentina.

The reviews and papers presented at this excellent work-shop are mandatory reading for all those interested in highenergy astrophysical phenomena. It is unfair to pick out par-ticular items from the plethora of new results, but I can-not resist highlighting a few topics. The wonderful im-

M. Longair (�)Cavendish Laboratory, University of Cambridge,JJ Thomson Avenue, Cambridge CB3 0HE, UKe-mail: [email protected]

ages of supernova remnants in ultrahigh energy gamma-rays produced by the HESS Cherenkov Array in Namibiawill go instantly into all the text-books. For me, the mostplausible explanation is that these observations provide di-rect evidence for the acceleration of protons in the shockwaves associated with supernova remnants. The detectionof gamma-rays from blazars provides not only a challengeto identify the emission mechanism with certainty, but alsoprovides a powerful tool for setting significant upper lim-its to the intensity of the optical-infrared background radi-ation. The continued flood of information on gamma-raybursts is strongly constraining models for these extraordi-nary sources of enormous gamma-ray luminosities. New ev-idence on the spectrum of the highest energy cosmic raysand its interpretation suggests that the resolution of the long-standing debate over the interpretation of these data may bein sight.

The organisers of this meeting have emphasised theMulti-messenger Approach to gamma-ray astrophysics andthis wide-ranging vision was fully justified by the diverseapproaches taken in the reviews and contributed papers.The astrophysical questions raised by gamma-ray observa-tions were clarified and sharpened over the four days ofthe meeting and the most fruitful future directions werelaid out. When we talk about the astrophysics of the fu-ture, we mean pushing beyond what we are doing now togain understanding of matter and radiation under really ex-treme physical conditions, which are inaccessible in terres-trial laboratories. This workshop demonstrated how excit-ing these areas are with the prospect of new insights intosome of the most challenging problems of contemporary as-trophysics.

We owe a profound vote of thanks to the organisers, JosepM. Paredes, Olaf Reimer and Diego F. Torres, for bringingtogether a diverse group of specialists who thrashed out the

4 Astrophys Space Sci (2007) 309: 3–4

astrophysical significance of this wealth of new observation.I can thoroughly recommend careful study of the paperscontained in this volume in the confident belief that they willstretch the imaginations of readers and enable them to ap-

preciate the richness of these very high energy astrophysicalstudies.

Cambridge, October 8th, 2006



New INTEGRAL sources and TeV emission

Roland Walter

Received: 7 July 2006 / Accepted: 1 November 2006 / Published online: 12 April 2007© Springer Science+Business Media B.V. 2007

Abstract INTEGRAL is operational since more than threeyears and producing high quality data that allows to detectfainter new hard X-ray sources. The new sources, identifieduntil now, are mostly active galactic nuclei and absorbed ortransient high mass X-ray binaries. TeV emission could beexpected from the new high mass X-ray binaries accretingdense clumps of stellar wind. INTEGRAL sources with TeVcounterparts are discussed.

Keywords X- and gamma-ray telescopes andinstrumentation · X-rays: binaries · Pulsars · Mass loss andstellar winds

PACS 95.55.Ka · 97.80.Jp · 97.60.Gb · 97.10.Me

1 Introduction

Since the beginning of 2003, the international gamma-rayastrophysics laboratory (INTEGRAL) is surveying the skyat hard X-rays and gamma-rays, with a particular emphasison the plane and central regions of the Galaxy. The two mainscientific instruments consist of the imager (IBIS) (Ubertiniet al. 2003) and of the spectrometer SPI (Vedrenne et al.2003), providing respectively sub-arcmin source positioning

Based on observations with INTEGRAL, an ESA project withinstruments and science data centre funded by ESA member states(especially the PI countries: Denmark, France, Germany, Italy,Switzerland, Spain), Czech Republic and Poland, and with theparticipation of Russia and the USA.

R. Walter (�)Observatoire de Genève, INTEGRAL Science Data Centre,Chemin d’Ecogia 16, 1290, Versoix, Switzerlande-mail: [email protected]

and keV spectral resolution in a band ranging from 17 keV tofew MeV. Both instruments use coded-masks for the imag-ing. Substantial monitoring capabilities are also provided inthe X-rays (3–35 keV) and in the optical V band by the JEM-X (Lund et al. 2003) and OMC (Mas-Hesse et al. 2003) in-struments.

About 75% of the INTEGRAL observing program isdriven by selected open time observation proposals and tar-get of opportunity observations. The remaining of the ob-serving time is devoted to the so-called core program, thatmostly consists of regular scans of the galactic plane andcentral regions. During the two first years in operation manytransient and new sources were detected. With an averageof one IAU circular or Astronomer’s Telegram issued perweek following INTEGRAL observations (Courvoisier etal. 2003), one of the goal of the INTEGRAL core programnamely to monitor the hard X-ray sky is fully met.

Multiplexing large field of view images obtained formany short pointings allows to obtain Msec effective ex-posure times and sub mCrab sensitivity in many areas ofthe sky. Below 100 keV, the IBIS spatial resolution is a keyfeature to resolve the numerous point sources of hard X-rayemission detected in particular within the Galactic bulge andspiral arms.

2 Source detection and identification

All sky mosaic images have been constructed using all pub-lic INTEGRAL/IBIS data obtained from the beginning ofthe mission up to April 2005 (Fig. 2). Source candidateshave been extracted and filtered according to source signif-icance, available exposure time and source shape to mini-mize the number of false detection. Transient sources thatwere active on a short time scale are usually not significant


enough to appear in average mosaic images and need to beconsidered separately.

The number of source candidates detected above 25 keVamounts to about 375. Among them not more than 30 falsedetections are expected. For comparison, above 20 keV,HEAO-1 and SIGMA detected 70 sources down to 14 mCraband respectively 15 galactic sources to a sensitivity of30 mCrab.

Many detections correspond to sources known before IN-TEGRAL. Identification of a fraction of the new candidateshas been obtained by improving the source position fromarcmin to arcsec scales through the search for radio/soft X-ray counterparts in existing archives (Stephen et al. 2006) orthrough specific high resolution X-ray observations (Walteret al. 2006b). Optical/infrared spectroscopy of counterpartcandidates have then been obtained (Masetti et al. 2006).

About 70 out of 200 new hard X-ray sources have alreadybeen identified (Fig. 1). 40% of them are high-mass X-raybinaries (HMXB), 40% are active galactic nuclei (AGN) and

Fig. 1 Identification of the new INTEGRAL sources

the remaining are distributed as follow: 3 low-mass X-raybinaries (LMXB), 4 X-ray novae, 6 cataclismic variables,2 symbiotic stars, 1 msec pulsar and a few counterparts ofTeV sources (see Sect. 6 for more details).

Study of the new INTEGRAL detected sources providedseveral unexpected results:

– 25% of the Active Galactic Nuclei detected by INTE-GRAL are new detections. Those sources are not particu-larly absorbed but located behind the galactic plane (Bas-sani et al. 2006).

– 50% of the HMXB are new absorbed or transient systems(Walter et al. 2006b).

– Anomalous X-ray pulsars have very hard spectra in thesoft gamma-rays, signature of magnetar emission (Kuiperet al. 2006).

– Hard X-ray counterparts of several unidentified HESSsources (Ubertini et al. 2005; Malizia et al. 2005).

The distribution of INTEGRAL sources on the sky is asexpected with Active Galactic Nuclei following the expo-sure map, HMXB tracing the galactic plane, the Gould beltand the two Magellanic Clouds and finally LMXB tracingolder stellar population and in particular the bulge of theGalaxy.

3 High-mass X-ray binaries

A number of new bright persistent sources have been de-tected by INTEGRAL above 20 keV in the galactic plane.Such sources were either unknown before INTEGRAL orweakly detected in previous X-ray surveys. Follow-up ob-servations of a subset of those sources with XMM-Newton

Fig. 2 High resolution image of the inner galaxy above 25 keV by INTEGRAL/IBIS (the image covers 100◦ × 60◦)

Astrophys Space Sci (2007) 309: 5–9 7

Fig. 3 Known and new Be and super-giant HMXB systems (left andright column respectively) detected by INTEGRAL

revealed that 80% of those new persistent sources are highlyabsorbed. Most of them are accreting pulsars in HMXB sys-tems with long (100–1300 sec) spin periods characteristicof wind accretion. The orbital periods and infrared spectraindicate the presence of massive companions, most likelysuper-giant stars (Walter et al. 2006b).

A family of fast hard X-ray transients, discovered by IN-TEGRAL, flaring on few hours time scales (Sguera et al.2006), have also been associated with super-giant compan-ion stars (Negueruela et al. 2006).

Among the HMXB detected by INTEGRAL 25 wereknown previously and 26 are new systems. Because of theirtransient and long period nature only 15 Be systems havebeen detected out of the hundred known systems and 6 newones have been discovered. The 10 wind accreting super-giant persistent systems previously known in the Galaxyhave been detected by INTEGRAL. In addition 20 new su-pergiant systems have been discovered, increasing the num-ber of those systems by a factor of 3. 13 of them are obscuredand persistent and 7 are fast transients.

The distribution of HMXB detected by INTEGRALalong the galactic plane peaks in the Norma and Scu-tum/Sagittarius inner spiral arms regions. The sources areon average slightly brighter and more scattered along thegalactic plane in the Sagittarius region as expected if theNorma region sources are located further out from the Sun.This suggests that the bulk of the observed HMXB popu-lation is located in the outer parts of the inner arms at adistance of the order of 5 kpc from the Galactic Center.

4 Dense and clumpy stellar winds

In contrast with the new persistent HMXB discovered byINTEGRAL the previously known systems are most of thetime not strongly absorbed and bright in the X-ray band. Astheir average X-ray luminosity is not exceptional, the new

Fig. 4 X-ray spectrum of IGR J16318-4848 as observed byXMM-Newton and INTEGRAL. The strongly attenuated continuumat soft X-rays indicate an absorption column density as large as2 × 1024 cm−2. The region emitting the fluorescence Fe and Ni lines ismuch less absorbed

sources detected by INTEGRAL are very likely character-ized by peculiar wind geometrical configuration (e.g. denseequatorial disks or accretion wakes).

INTEGRAL observations indicate that the new absorbedsystems form the majority of the active super-giant HMXBdetected so far. The fluorescence lines are particularly strik-ing in those objects (Fig. 4). Together with the continuumspectral shape they point towards a transmission geometryin which the compact sources are embedded within a denseenvelope of cold matter. A likely model is one in which thecompact object orbits its companion within a dense stellarwind component. This model is confirmed by the eclipses(Hill et al. 2005; Zurita et al. 2006) and the evidences forvariation of the absorbing column densities that have beenfound in few of those sources (Rodriguez et al. 2006). Fastflaring activities also indicate inhomogeneities in the ac-creted wind (Walter 2006a).

Fast transient super-giant HMXB are characterized byshort flares lasting a few hours and separated by manyweeks. These flares are likely the signature of the interac-tion between the orbiting compact source and highly in-homogeneous stellar winds made of very dense clumps(Leyder et al. 2007). The timing characteristics of thoseflares allows to measure the physical characteristics of thoseclumps.

The main difference between transparent, obscured andfast transient super-giant HMXB systems is probably re-lated to the structure of the stellar wind. The clumpness ofthe wind could increase from transparent to obscured andeven further to fast transient systems. Hard X-ray variabilitystudies provide a direct way to probe the stellar wind char-acteristics (clumpness, density) and constrain clumpy stellarwind models (Leyder et al. 2007). This complements stud-ies based on high resolution spectroscopy (Oskinova et al.2006).


Table 1 TeV sources with possible INTEGRAL counterparts

Source Type Remarks

HESS J1745-290 SNR The peak of the hard X-ray emission detected close to the galactic center by INTE-GRAL is located 1 arcmin from Sgr A*. It very probably corresponds to the hardenergy tail of the X-ray diffuse emission observed within 6 arcmin of the galacticcenter. That hard X-ray emission could be interpreted as synchrotron emission as-sociated with the inverse Compton TeV emission detected by HESS (Neronov et al.2005)

Crab PWN The INTEGRAL detection is completely dominated by the pulsar emission

HESS J1514-591 PWN INTEGRAL is detecting the associated pulsar PSR B1509-58. The pulsar is toobright to allow for the detection of the jet-like PWN observed by Chandra and HESS

HESS J1837-069 ? The nature of this source is still unclear. As the INTEGRAL PSF is larger than theTeV extension measured by HESS, it is currently not possible to tell if the INTE-GRAL point-like source is a counterpart of the TeV emission (Malizia et al. 2005)

HESS J1813-178 SNR The point-like INTEGRAL source is compatible with the almost point-like HESSsource (Ubertini et al. 2005)

HESS J1420-607 PWN/SNR An excess of emission is observed by INTEGRAL at a slight offset from theTeV source. The INTEGRAL position coincides with the soft X-ray source RXSJ141935.3-604523 located about 4 arcmin at the west of the X-ray source PSRJ1420-6048 located close to the center of the TeV emission. The INTEGRAL/RXSand the HESS/ASCA sources coincide with two different wings of the KookaburraSNR

HESS J1616-508 PWN/SNR A clear and elongated excess of hard X-ray emission is detected by INTEGRAL atthe position of the TeV source. Both are offset from PSR J1617-5055 and close tothe SNRs G332.4-0.4 and G332.4+0.1

LS 5039 HMXB An excess of emission is observed by INTEGRAL at the position of LS 5039. Thesignificance is however too low for a formal detection

PSR B1259-63 HMXB+PWN The INTEGRAL and X-ray variability of PSR B1259-63 are compared to the HESSobservations in this volume by Neronov et al. (2006)

LSI +61 303 HMXB+PWN The TeV, INTEGRAL, X-ray and radio variability of LSI +61 303 are compared anddiscussed in Chernyakova et al. (2006)

HESS J1632-478 ? IGR J16320-4751 is an obscured HMXB featuring bright flares (Rodriguez et al.2006) probably related with inhomogeneous stellar winds. As such it could emitTeV (Sect. 5). However the HESS source seems extended, ruling out the association

5 Do neutron stars accreting dense stellar wind clumpsemit TeV?

The possible detection of HMXB in the TeV range in the80’s (see Protheroe 1986 for a review) led to the idea thatprotons trapped in the outer and closed regions of a neu-tron star magnetosphere could be accelerated up to γ ≈ 108

by multiple scattering of Alfvén waves close to the accre-tion column (Katz and Smith 1988). Protons could be ac-celerated to high energies only if the synchrotron loss timeis larger than the travel time needed to bounce back oftenenough to gain energy. Within milli-seconds, when the giro-radius becomes larger than the magnetospheric region, thehigh energy protons escape the system. For the highest en-ergies this occurs close to the Alfvén radius RA ≈ 108 cm.The luminosity of the high energy proton leaving the systemwas estimated as the total energy trapped over the escapetime and could reach 1036 erg/s (Smith et al. 1992).

Significant γ -ray production will take place if the highenergy protons interact with dense enough accreted mate-

rial outside of the Alfvén radius. The size of the stellarwind clumps assumed to be responsible for the hard X-rayflares observed in fast transient super-giant HMXB couldbe estimated as 1010 cm from the duration of the flares.At ∼109 cm from the neutron star, the pulsar magneticfield will be of the same order than the companion stellarwind magnetic field (i.e. ∼100G (Donati et al. 2006)) suchthat Bohm diffusion could take place. The diffusion timescale td ≈ 150 s × (R/1010 cm)2 × (1 TeV/E)× (B/100G)

becomes larger than the proton interaction timescale tp =470 s × (1023 cm−2/NH ) × (R/1010 cm) if

NH ≥ 5 × 1023 cm−2 × (E/1 TeV) × (100G/B)

× (1010 cm/R)

We conclude that HMXB accreting dense stellar windclump could be transient TeV sources if the column densityis large enough. This applies to dense clumps in fast tran-sient HMXB and to persistent obscured HMXB if the accre-tion wakes are dense enough. The strength of the TeV emis-


sion also depends on the magnetic strength and structure inthe stellar wind. As column densities larger than 1023 cm−2

have been observed in specific objects during flares, onecould expect to observe TeV flares from fast transient super-giant HMXB on a timescale of a few hours and possibly inpersistent highly absorbed systems as well.

6 INTEGRAL counterparts of TeV sources

Several of the sources detected by HESS and MAGIC inthe galactic plane do have counterparts detected by INTE-GRAL (see Table 1) including some of the unidentifiedHESS sources. In the case of point-like and variable TeVsources the correspondence between the TeV and INTE-GRAL sources is almost sure. For extended sources the sit-uation is more delicate as HESS and INTEGRAL may seedifferent particle acceleration sites emitted in different re-gions of the same supernova remnant (SNR) or by a pulsarand/or its pulsar wind nebula (PWN). Clearly more spatialresolution and sensitivity are needed from the hard X-raysto the TeV range to allow precise mapping of particle accel-eration sites in the Galaxy.

Currently none of the obscured or fast X-ray transientHMXB discovered by INTEGRAL has been detected in theTeV (with the possible exception of IGR J16320-478), how-ever one must note that TeV emission is expected only dur-ing the accretion of dense clumps which happens only whenthe source is active at hard X-rays, i.e. for a small fractionof the time.

References

Bassani, L., et al.: INTEGRAL IBIS extragalactic survey: active galac-tic nuclei selected at 20–100 keV. Astrophys. J. 636, L65 (2006)

Chernyakova, M., et al.: XMM-Newton observations of PSR B1259-63 near the 2004 periastron passage. Mon. Not. Roy. Astron. Soc.367, 1201 (2006)

Courvoisier, T., et al.: The INTEGRAL science data centre (ISDC).Astron. Astrophys. 411, L53 (2003)

Donati, J.-F., et al.: The surprising magnetic topology of tau Sco: fossilremnant or dynamo output? Mon. Not. Roy. Astron. Soc. 370, 629(2006)

Katz, J.A., Smith, I.A.: Particle acceleration in accreting magne-tospheres. Astrophys. J. 326, 733 (1988)

Kuiper, L., et al.: Discovery of luminous pulsed hard X-ray emissionfrom anomalous X-ray pulsars 1RXS J1708-4009, 4U 0142+61,and 1E 2259+586 by INTEGRAL and RXTE. Astrophys. J. 645,556 (2006)

Leyder, J.-C., et al.: Hard X-ray flares in IGR J084084503 unveilclumpy stellar winds. Astron. Astrophys. (2007, in press)

Lund, N., et al.: JEM-X: The X-ray monitor aboard INTEGRAL. As-tron. Astrophys. 411, L231 (2003)

Hill, A., et al.: The 1–50 keV spectral and timing analysis of IGRJ18027-2016: an eclipsing, high mass X-ray binary. Astron. As-trophys. 439, 255 (2005)

Johnson, P.: Rapid stochastic acceleration of protons to energies above100 TeV in the accretion column of Hercules X-1. Astropart. Phys.3, 53 (1995)

Malizia, A., et al.: The INTEGRAL/IBIS source AX J1838.0-0655:a soft X-Ray-to-TeV gamma-ray broadband emitter. Astrophys. J.630, L157 (2005)

Mas-Hesse, M., et al.: OMC: an optical monitoring camera for IN-TEGRAL. Instrument description and performance. Astron. As-trophys. 411, L261 (2003)

Masetti, N., et al.: Unveiling the nature of INTEGRAL objects throughoptical spectroscopy. IV. A study of six new hard X-ray sources.Astron. Astrophys. 455, 11 (2006)

Negueruela, I., et al.: Supergiant fast X-ray transients: a new class ofhigh mass X-ray binaries unveiled by INTEGRAL. In: Wilson, A.(ed.) The X-ray Universe 2005. ESA SP-604, vol. 1, p. 165.Noordwijk (2006)

Neronov, A., et al.: Hard X-ray diffuse emission from the Galactic Cen-ter seen by INTEGRAL. astro-ph/0506437 (2005)

Neronov, A., et al.: Radio to TeV gamma-ray emission from PSRB1259-63. Astrophys. Space Sci. doi:10.1007/s10509-007-9454-3(2007)

Protheroe, R.: Gamma-ray astronomy at the highest energies. Proc. As-tron. Soc. Aust. 6, 280 (1986)

Oskinova, L., et al.: High-resolution X-ray spectroscopy of bright O-type stars. Mon. Not. Roy. Astron. Soc. 372, 313 (2006)

Rodriguez, J., et al.: INTEGRAL and XMM-Newton observations ofthe X-ray pulsar IGR J16320-4751/AX J1631.9-4752. Mon. Not.Roy. Astron. Soc. 366, 274 (2006)

Sguera, V., et al.: Unveiling supergiant fast X-ray transient sources withINTEGRAL. Astrophys. J. 646, 452 (2006)

Smith, I.A., et al.: Proton acceleration in neutron star magnetospheres.Astrophys. J. 388, 148 (1992)

Stephen, J., et al.: Using the ROSAT catalogues to find counterpartsfor the second IBIS/ISGRI survey sources. Astron. Astrophys. 445,869 (2006)

Ubertini, P., et al.: IBIS: the imager on-board INTEGRAL. Astron. As-trophys. 411, L131 (2003)

Ubertini, P., et al.: INTEGRAL IGR J18135-1751 = HESS J1813-178:a new cosmic high-energy accelerator from keV to TeV energies.Astrophys. J. 629, L109 (2005)

Vedrenne, G., et al.: SPI: the spectrometer aboard INTEGRAL. Astron.Astrophys. 411, L63 (2003)

Walter, R., et al.: IGR J16318-4848 & Co., a new population of hid-den high-mass X-ray binaries in the Norma Arm of the Galaxy. In:Battrick, B. (ed.) The INTEGRAL Universe. ESA SP-552, p. 417.Noordwijk (2004)

Walter, R.: XMM-Newton and INTEGRAL observations of new ab-sorbed supergiant high-mass X-ray binaries. In: Wilson, A. (ed.)The X-ray Universe 2005. ESA SP-604, vol. 1, p. 161. Noordwijk(2006)

Walter, R., et al.: XMM-Newton and INTEGRAL observations of newabsorbed supergiant high-mass X-ray binaries. Astron. Astrophys.453, 133 (2006)

Zurita, J., et al.: IGR J17252-3616: an accreting pulsar observed by IN-TEGRAL and XMM-Newton. Astron. Astrophys. 448, 261 (2006)

http://dx.doi.org/10.1007/s10509-007-9454-3



Status of identification of VHE γ -ray sources

Stefan Funk


Abstract With the recent advances made by Cherenkovtelescopes such as H.E.S.S. the field of very high-energy(VHE) γ -ray astronomy has recently entered a new era inwhich for the first time populations of Galactic sources suchas e.g. Pulsar wind nebulae (PWNe) or Supernova remnants(SNRs) can be studied. However, while some of the newsources can be associated by positional coincidence as wellas by consistent multi-wavelength data to a known counter-part at other wavelengths, most of the sources remain notfinally identified. In the following, the population of Galac-tic H.E.S.S. sources will be used to demonstrate the status ofthe identifications, to classify them into categories accordingto this status and to point out outstanding problems.

Keywords RX J1713.7–3946 · HESS J1825–137 · HESSJ1813–178 · Gamma-rays · H.E.S.S. · Source identification

PACS 95.55.Ka · 95.85.Pw · 98.38.Mz

1 Introduction

A systematic survey of the inner part of the Galaxy per-formed by the H.E.S.S. Cherenkov telescope system has re-vealed a number of previously unknown sources of VHEgamma-rays above 100 GeV (Aharonian et al. 2005a,2006a). While in terms of a population approach the sources

For the H.E.S.S. collaboration

S. Funk (�)Kavli Institute for Particle Astrophysics and Cosmology,Stanford University, 2575 Sand Hill Road, PO Box 0029,Stanford, CA 94025, USAe-mail: [email protected]

Table 1 Categories into which the gamma-ray sources will be classi-fied in the following sections

Matching position/ Viable emission Consistent

morphology mechanism MWL picture

A yes yes yes

B no yes yes

C yes yes no

D no no no

can be described by common properties like generally ratherhard energy spectra (photon index ∼2.3) or a rather narrowdistribution in Galactic latitude (rms of ∼0.3°) the coun-terpart identification calls for an individual study of theseobjects. An unambiguous counterpart identification of these(initially) unidentified H.E.S.S. sources requires (i) spa-tial and ideally also morphological coincidence, (ii) a vi-able gamma-ray emission mechanism for the object, and(iii) a consistent multi-wavelength behaviour matching thesuggested identification and the particle distribution withinthe source. The H.E.S.S. sources can be classified accord-ing to their confidence in identification with known as-trophysical objects following the three requirements givenabove. Table 1 summarises the categories. Category A com-prises sources for which the positional and/or morphologi-cal match (in case of an extended source) with a counterpartsource is excellent and the emission processes can be mod-elled to provide a consistent picture describing the multi-frequency data. For these sources the association is beyonddoubt. For Category B sources the emission mechanismscan be consistently modelled, however these sources show aless convincing positional and/or morphological match withthe potential counterpart. Category C sources on the otherhand have a good positional counterpart, they show however


a non-consistent multi-wavelength picture, being it becauseof insufficient data at other wavebands, being it because of anot fully understood emission mechanism. For Category Dsources no counterpart candidate exists, these are the clas-sical unidentified sources. In the following I will describeexamples for sources belonging to each of the 4 categories.The description will focus on Galactic gamma-ray sources,since for extragalactic objects the counterpart identificationin the VHE gamma-ray regime has (so far) turned out to berather unproblematic.

2 Category A—sources with an established counterpart

Two classes of sources can be distinguished for which acounterpart to the VHE gamma-ray source has been es-tablished: (a) point sources with a convincing positionalmatch and (b) extended sources with a convincing positionaland morphological match. For these objects with a firmcounterpart, having established the positional coincidence,the aim for these objects is to fully understand the detailsof the multi-frequency photon spectrum and to investigatethe emission mechanisms generating this photon spectrum.One important question in the VHE gamma-ray regime isfor example whether the gamma-ray emission is generatedby Inverse Compton scattering of ultra-relativistic electronson photon fields like the Cosmic microwave background(CMBR) or by pion-decay produced in proton-proton inter-actions, that is whether the gamma-ray emission has leptonicor hadronic origin. These two scenarios can not be directlydistinguished from the gamma-ray data alone, but have tobe separated by modelling the parent population of parti-cles responsible for the emission. For any source identifica-tion it should be mentioned that the good angular resolutionof VHE Cherenkov instruments (typically of the order of0.1° per event) as well as the very low level of the diffusegamma-ray background at energies above 100 GeV helpsagainst source confusion. Source confusion was a problemthat EGRET (Hartman et al. 1999) strongly had to face, es-pecially in observations in the Galactic plane where boththe density of sources and the level of the diffuse gamma-ray background was higher. The upcoming GLAST satellitemeasuring in the regime between <20 MeV to ∼300 GeVwill have the advantage of an improved angular resolution(∼0.6° at 1 GeV) over EGRET but will also be suscepti-ble to any systematic uncertainties in modelling the diffusegamma-ray flux from cosmic ray interactions with interstel-lar material.

The best-established example for a point-source with aconvincing positional match is the Crab Nebula (Aharonianet al. 2006f), which is frequently used as a calibration sourcein VHE γ -ray astronomy. In order to establish a positionalcorrelation with a gamma-ray point-source, it has to be

Fig. 1 ASCA X-ray image of RX J1713.7–3946, overlaid withsmoothed and acceptance-corrected H.E.S.S. gamma-ray image con-tours linearly spaced at the level of 30, 60 and 90 counts. The ASCAimage was smoothed to match the H.E.S.S. point-spread function forease of comparison

tested whether the nominal position of the counterpart can-didate is within the statistical and systematic error limits ofthe reconstructed position of the gamma-ray emission re-gion (for the Crab Nebula, the statistical error on the re-constructed position of the gamma-ray emission is 5′′, thesystematic error of the order of 20′′). Other objects of thisclass, where a positional counterpart to a point-like gamma-ray emission has been established are the newly discoveredgamma-ray microquasars LS 5039 (Aharonian et al. 2005h,2006e) and LS I+61 303 (Albert et al. 2006) or the com-posite PWN G0.9+0.1 (Aharonian et al. 2005b). A furtherstrengthening of the proposed association can be establishedif additionally (as in the case of LS 5039) a correlated pe-riodicity or variability between the gamma-ray source andthe counterpart source can be detected (LS 5039 shows amodulation in the gamma-ray data that matches the orbitalfrequency of the binary system as discussed by deNaurois etal. in this proceedings).

The best established example for an extended source witha convincing morphological match is the Supernova remnant(SNR) RX J1713.7–3946 (Aharonian et al. 2004, 2006b)(and also see Lemoine-Gourmard et al. in this proceedings)showing a striking correlation of the gamma-ray emissionto X-rays as e.g. measured by the ASCA satellite (corre-lation coefficient: ∼80%). From the morphological match,the association of the gamma-ray source with the Supernovaremnant is beyond doubt and can be regarded as a firm as-sociation. Another object of this class is the Supernova rem-nant RX J0852.0–4622 (Vela Jr.) showing also a high de-


gree of correlation between the gamma-ray and X-ray emis-sion (Aharonian et al. 2005c). Other objects of this classare the PWNe MSH–15–52 (Aharonian et al. 2005d) andVela X (Aharonian et al. 2006c).

In terms of modelling the multi-frequency emission fromthese objects where a firm counterpart has been established,the Crab Nebula can again serve as an excellent examplehow the outstanding coverage in wavebands from radio toVHE γ -rays can help to consistently describe the emissionmechanism in this object and to derive important parame-ters like the strength of the magnetic field responsible for thesynchrotron emission. For the microquasar LS 5039 as wellas for the extended gamma-ray emission from the Supernovaremnants RX J1713.7–3946 and RX J0852.0–4622, there isnot yet a unique solution to describe the multi-frequencydata and the gamma-ray emission can be explained both interms of a leptonic emission mechanism generated by In-verse Compton scattering as well as in terms of a hadronicscenario where the gamma-ray emission is generated by thedecay of neutral pions.

3 Category B—sources with a less convincing positionalor morphological match

The best example for members of this class are the newlydiscovered gamma-ray sources, which seem to belong tothe so-called offset Pulsar wind nebulae. These objects, forwhich Vela X (Aharonian et al. 2006c) is the archetypalexample show an extended emission around an energeticpulsar. The offset morphology is thought to arise from ananisotropy in the interstellar material surrounding the pul-sar, that prevents the symmetric expansion of the PWN inone direction and shifts the emission to the other direction(see e.g. (Blondin et al. 2001) for a hydro-dynamical simula-tion and discussion of this effect). The gamma-ray emissionin these objects is generated by Inverse Compton scatteringof relativistic electrons accelerated in the termination shockof the PWN. The gamma-ray sources that could possibly beexplained in this framework are typically extended and theiremission region overlaps with energetic pulsars (energeticenough to explain the gamma-ray flux by their spindownpower) and that very importantly also show evidence foran X-ray PWN. Apart from Vela X and MSH–15–52, thegamma-ray emission of the PWN HESS J1825–137 (Aha-ronian et al. 2005e, 2006d), possibly powered by the ener-getic pulsar PSR J1826–1334 can be used to illustrate theproperties of this class of objects. This object has been ob-served by H.E.S.S. in a very deep observation of ∼67 hours,due to its proximity to the microquasar LS 5039 (at a dis-tance of ∼1°). It was known to be a PWN candidate also inVHE gamma-rays since through XMM-Newton X-ray ob-servations of PSR J1826–1334 (Gaensler et al. 2003) es-tablished an offset X-ray PWN extending asymmetrically

Fig. 2 a XMM-Newton X-ray image in the energy range between 2and 12 keV of the small region (7′ ×7′) surrounding PSR J1826–1334.It can be seen, that the X-ray emission shows an asymmetric diffuseemission extending to the south of the pulsar. b H.E.S.S. gamma-rayexcess image of the region surrounding HESS J1825–137 and the en-ergetic pulsar PSR J1826–1334 (white triangle). The image has beensmoothed with a Gaussian of radius 2.5′ and has been corrected forthe changing acceptance across the field. The inset in the bottom leftcorner shows the PSF of the data set (smoothed in the same way as theexcess map). The dashed black and white contours are linearly spacedand denote the 5σ , 10σ and 15σ significance levels. The best fit posi-tion of HESS J1825–137 is marked with a black square, the best exten-sion and position angle by a black ellipse. Also shown (dotted white)is the 95% positional confidence contour of the unidentified EGRETsource 3EG J1826–1302. The bright point-source to the south of HESSJ1825–137 is the microquasar LS 5039. c Three-colour image, showingthe gamma-ray emission as shown in panel b, with different colours,denoting the gamma-ray emission in different energy bands, symbolis-ing the changing morphology even in the gamma-ray band alone. Moredetails can be found in the text and in (Aharonian et al. 2006d)

∼5′′ to the south of the pulsar. The asymmetric nature ofthe PWN as well as CO data that show a dense molecularcloud to the north of PSR J1826–1334 (Lemière 2005) sup-port the picture described above in which dense material tothe north shifts the PWN emission to the south.

The H.E.S.S. detected gamma-ray emission similarlyshows an asymmetric emission extending to the south ofthe pulsar, however on a completely different scale than theX-ray emission (the X-ray emission extends 5′′, whereas the


gamma-ray emission extends ∼1° to the south). This at firstglance prevents a direct identification as a counterpart, sincethe morphology can not be matched between X-rays andgamma-rays. However, modelling the emission mechanismand taking into account the different loss timescales of thegamma-ray and X-ray emitting electrons the different scaleof the emission regions can be plausibly explained in thefollowing way: for a typical magnetic field of 10 µG (as alsosuggested from the X-ray data), 1 keV X-rays are generatedby ∼50 TeV electrons, whereas 100 GeV gamma-rays aregenerated by ∼1 TeV electrons. The gamma-rays are there-fore generated by lower energy electrons than the X-raysand the different scales of the emission regions could be dueto faster loss times of the higher energetic synchrotron emit-ting electrons. This picture is further supported by the fact,that also in the gamma-ray regime on its own, a decreasein size with increasing energy can be established as shownin (Aharonian et al. 2006d) and illustrated in Fig. 2.

If this picture is correct and the X-ray and gamma-raysources are associated, then the different morphologies inthese two wavebands (i.e. the different angular scales) canbe explained in a consistent picture. Therefore these twosources can possibly be associated, even though there is nodirect morphological match. It should be noted, that in or-der to establish the association as a firm counterpart, moremulti-frequency data are needed that confirm this picture.Also it should be mentioned, that this source generates anew template for a large number of other extended gamma-ray sources found close to energetic pulsars. However, in thecase of HESS J1825–137 there are two striking argumentsthat further substantiate the association: (a) the asymmetricX-ray PWN found by XMM and (b) the changing morphol-ogy found by H.E.S.S. in the gamma-ray data. If these twoproperties had not been found, the association should not beconsidered more than a chance positional coincidence. Nev-ertheless as mentioned, there is a whole new class of objectsthat are now unidentified and that are close to energetic pul-sars, that could potential belong to this class of objects.

4 Category C—sources for which the multi-frequencydata does not (yet) provide a consistent picture

The sources belonging to this class of objects have a goodpositional counterpart, but the multi-frequency data does not(yet) provide a consistent picture of the emission mecha-nism. One object of this class is HESS J1813–178. Thisslightly extended gamma-ray source was originally discov-ered in the H.E.S.S. Galactic plane survey and originallydescribed as unidentified. Shortly afterwards several paperswere published, describing the positional coincidence withan ASCA X-ray source (2–8 keV) (Brogan et al. 2005), anIntegral hard X-ray source (20–100 keV) (Ubertini et al.

RA (hours)

).ge

d( ceD

-17.9

-17.88

-17.86

-17.84

-17.82

-17.8

-17.78

s30m13h18s40m13h18s50m13h18

Fig. 3 Gamma-ray image of HESS J1813–178 overlaid with VLA20 cm radio data in which the shell-type structure of the positionalcounterpart to the gamma-ray source can be seen. The best fit positionof the gamma-ray excess is marked with a magenta cross

2005), both having the same angular resolution and there-fore like H.E.S.S. unable to resolve the object.

Finally VLA archival radio data (90 cm) were re-ported (Brogan et al. 2005) showing a 2.5′ diameter shell-like object coincident with the X-ray sources and with HESSJ1813–178 (see Fig. 3). This observation led to the conclu-sion that the positionally coincident object was a Supernovaremnant and that the gamma-ray emission was either causedby the shell or by a centrally located PWN. However, themulti-frequency data does not distinguish between the twoscenarios, due to the lack of spatial resolution of the X-rayand gamma-ray instruments. A preliminary analysis of a re-cent 30 ks XMM-Newton observation of the region pointto a PWN origin of the emission, which would in turn al-low to model the radio to gamma-ray data in the picture ofa composite SNR and therefore finally possibly identifyingHESS J1813–178 as a gamma-ray PWN. The example ofHESS J1813–178 shows that high-quality multi-frequencyare a prerequisite in the identification of a gamma-ray sourceand an ongoing programme is connected to the study of theunidentified gamma-ray sources with high-resolution X-raydetectors like Chandra, XMM-Newton and Suzaku. Otherobjects for which the multi-frequency data at this momentdo not allow firm conclusions about potential counterpartsare e.g. HESS J1640–465 and HESS J1834–087.

5 Category D—sources with no counterpart

For this class of sources there is as of yet no counterpart atother wavelength determined. The common belief is, that


this is due to insufficient (in terms of sensitivity) multi-wavelength data. In principle all new gamma-ray detectionsare first classified in this category, before they can be movedto another class with the help of MWL data. The mostprominent examples of this class are the HEGRA sourcein the Cygnus region (TeV J2032+4131) (Aharonian et al.2005f) for which even in deep 50 ks Chandra observationno obvious X-ray counterpart was found. A similar caseis the unidentified H.E.S.S. source HESS J1303–631 (Aha-ronian et al. 2005g), serendipitously discovered in observa-tions of the binary pulsar PSR B1259–63. Also here a Chan-dra observation did not reveal any obvious positional coun-terpart (Mukherjee and Halpern 2005). In the search for X-ray counterparts XMM-Newton seems to be best suited forthe Galactic gamma-ray sources found by H.E.S.S. becausethe high sensitivity towards extended structures seems to beadvantageous for the gamma-ray sources that typically haveextensions of the order of 0.1°.

A large fraction of the new H.E.S.S. sources can be cat-egorised into this class. Examples are: HESS J1702–420,HESS J1708–410 or HESS J1745–303. As previously men-tioned, there is an ongoing effort to investigate these sourceswith various instruments from radio to gamma-rays.

6 Summary and conclusion

Table 2 summarises the H.E.S.S. gamma-ray sources intothe proposed categories as given in Table 1. As can be seenfrom this table, for less than half a firm counterpart can beidentified, putting these sources into class A. The majorityof the sources has to be classified as unidentified, since nogood counterpart at other wavebands exists at all. It is evi-dent, that the gathering of multi-waveband data will help inidentifying these objects. It should however also be noted,that a positional coincidence alone does not help. A con-sistent modelling of the emission mechanisms at work ingenerating the gamma-rays must be invoked, before a firmassociation can be established. For the upcoming GLASTsatellite, the situation will be further complicated by sourceconfusion due to the larger point-spread function and alsoby the stronger diffuse gamma-ray background from decaysof neutral pions in the Galactic plane.

Acknowledgements The support of the Namibian authorities and ofthe University of Namibia in facilitating the construction and operationof H.E.S.S. is gratefully acknowledged, as is the support by the Ger-man Ministry for Education and Research (BMBF), the Max PlanckSociety, the French Ministry for Research, the CNRS-IN2P3 and theAstroparticle Interdisciplinary Programme of the CNRS, the UK Par-ticle Physics and Astronomy Research Council (PPARC), the IPNP ofthe Charles University, the South African Department of Science andTechnology and National Research Foundation, and by the Universityof Namibia. We appreciate the excellent work of the technical supportstaff in Berlin, Durham, Hamburg, Heidelberg, Palaiseau, Paris, Saclay,and in Namibia in the construction and operation of the equipment.

Table 2 Summary of the classes for the H.E.S.S. Galactic gamma-raysources

Source Class Comment

J1713–397 A SNR RX J1713.7–3946

J0852–463 A SNR RX J0852.0–4622

J0835–455 A PWN Vela X

J1302–638 A PWN PSR B1259–63

J1420–607 A PWN PSR J1420–6048

J1514–591 A PWN MSH 15–52

J1747–281 A PWN G0.9+0.1

J1804–216 B PWN or SNR?

J1825–137 B PWN different scale

J1640–465 C SNR?

J1813–178 C Composite SNR?

J1834–087 C SNR?

J1303–631 D

J1614–518 D

J1632–478 D

J1634–472 D

J1702–420 D

J1708–410 D

J1745–290 D Galactic centre source

J1745–303 D

J1837–069 D

References

Aharonian, F., et al. (H.E.S.S. Collaboration): High-energy particleacceleration in the shell of a supernova remnant. Nature 432, 75(2004)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): A new population ofvery high energy γ -ray sources in the Milky Way. Science 307,1938 (2005a)

Aharonian, F., et al. (H.E.S.S. Collaboration): Very high energy gammarays from the composite SNR G0.9+0.1. Astron. Astrophys. 432,25 (2005b)

Aharonian, F., et al. (H.E.S.S. Collaboration): Detection of TeV γ -rayemission from the shell-type supernova remnant RX J0852.0-4622with H.E.S.S. Astron. Astrophys. 437, L7 (2005c)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): Discovery of ex-tended VHE gamma-ray emission from the asymmetric pulsarwind nebula in MSH 15-52 with HESS. Astron. Astrophys. 435,L17 (2005d)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): A possible associationof the new VHE γ -ray source HESS J1825-137 with the pulsarwind nebula G 18.0-0.7. Astron. Astrophys. 442, L25 (2005e)

Aharonian, F., et al.: The unidentified TeV source (TeV J2032+4130)and surrounding field: final HEGRA IACT-system results. Astron.Astrophys. 431, 197 (2005f)

Aharonian, F., et al. (H.E.S.S. Collaboration): Serendipitous discoveryof the unidentified extended TeV γ -ray source HESS J1303-631.Astron. Astrophys. 439, 1013 (2005g)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): Discovery of veryhigh energy gamma-rays associated with an X-ray binary. Science309, 746 (2005h)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): The H.E.S.S surveyof the Inner Galaxy in very high-energy gamma-rays. Astrophys.J. 636, 777 (2006a)


Aharonian, F.A., et al. (H.E.S.S. Collaboration): A detailed spectraland morphological study of the γ -ray supernova remnant RXJ1713.7-3946 with H.E.S.S. Astron. Astrophys. 449, 223 (2006b)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): First detection ofa VHE gamma-ray spectral maximum from a Cosmic source:H.E.S.S, discovery of the Vela X nebula. Astron. Astrophys. 448,L43 (2006c)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): Energy dependent γ -ray morphology in the prulsar wind nebula HESS J1825-137. As-tron. Astrophys. 460, 365 (2006d)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): Discovery of orbitalmodulation in the very high energy gamma-ray emission from theX-ray binary LS 5039. Astron. Astrophys. 460, 743 (2006e)

Aharonian, F.A., et al. (H.E.S.S. Collaboration): Observations of theCrab Nebula with H.E.S.S. Astron. Astrophys. 457, 899 (2006f)

Albert, J., et al.: Variable very high energy gamma-ray emission fromthe microquasar LS I +61 303. Science 312, 1771 (2006)

Blondin, J.M., Chevalier, R.A., Frierson, D.M.: Pulsar wind nebulae inevolved supernova remnants. Astrophys. J. 563, 806 (2001)

Brogan, C.L., Gaensler, B.M., Gelfand, J.D., Lazendic, J.S., Lazio,T.J., Kassim, N.E., McClure-Griffiths, N.M.: Discovery of a ra-dio Supernova remnant and non-thermal X-rays coincident withthe TeV source HESS J1813-178. Astrophys. J. Lett. 629, L105(2005)

Gaensler, B.M., Schulz, N.S., Kaspi, V.M., Pivovaroff, M.J., Becker,W.E.: XMM-Newton observations of PSR B1823-13: an asymmet-ric synchrotron nebula around a vela-like pulsar. Astrophys. J. 588,441 (2003)

Hartman, R.C., et al.: The third EGRET catalog of high-energygamma-ray sources. Astrophys. J. Suppl. Ser. 123, 79 (1999)

Lemière, A. (H.E.S.S. Collaboration): In: Proceedings of the 29thICRC, Pune, India, vol. 4, p. 105 (2005)

Mukherjee, R., Halpern, J.P.: Chandra observation of the unidentifiedTeV gamma-ray source HESS J1303-631 in the galactic plane. As-trophys. J. 629, 1017 (2005)

Ubertini, P., et al.: Integral IGR J18135-1751=HESS J1813-178: Anew cosmic high energy accelerator from KeV to TeV. Astrophys.J. Lett. 629, L109 (2005)



INTEGRAL/XMM views on the MeV sourceGRO J1411-64

Diego F. Torres · Shu Zhang · Olaf Reimer · Xavier Barcons · Amalia Corral ·Valentí Bosch-Ramon · Josep M. Paredes · Gustavo E. Romero · Jin Qu ·Werner Collmar · Volker Schönfelder · Yousaf Butt


Abstract The COMPTEL unidentified source GRO J 1411-64 was observed by INTEGRAL and XMM-Newton in2005. The Circinus Galaxy is the only source detectedwithin the 4σ location error of GRO J1411-64, but in hereexcluded as the possible counterpart. At soft X-rays, 22 re-

D.F. Torres (�)Institució de Recerca i Estudis Avançats (ICREA) & Institut deCiències de l’Espai (IEEC-CSIC), Facultat de Ciencies,Universitat Autònoma de Barcelona, Torre C5 Parell, 2a planta,08193 Barcelona, Spaine-mail: [email protected]

S. Zhang · J.L. QuLaboratory for Particle Astrophysics, Institute of High EnergyPhysics, Beijing 100049, China

O. ReimerW.W. Hansen Experimental Physics Laboratory, StanfordUniversity, Stanford, CA 94305, USA

X. Barcons · A. CorralInstituto de Física de Cantabria (CSIC-UC), 39005 Santander,Spain

V. Bosch-Ramon · J.M. ParedesUniversitat de Barcelona, Av. Diagonal 647, 08028 Barcelona,Spain

G.E. RomeroInstituto Argentino de Radioastronomia, CC5, 1894, Villa Elisa,Argentina

W. Collmar · V. SchönfelderMax-Planck-Institut für extraterrestrische Physik, P.O. Box 1603,85740 Garching, Germany

Y. ButtHarvard-Smithsonian Center for Astrophysics, 60 Garden St.,Cambridge, MA 02138, USA

liable and statistically significant sources (likelihood >10)were extracted and analyzed from XMM-Newton data. Onlyone of these sources, XMMU J141255.6-635932, is spec-trally compatible with GRO J1411-64 although the fact thesoft X-ray observations do not cover the full extent of theCOMPTEL source position uncertainty make an associa-tion hard to quantify and thus risky. At the best location ofthe source, detections at hard X-rays show only upper lim-its, which, together with MeV results obtained by COMP-TEL suggest the existence of a peak in power output lo-cated somewhere between 300–700 keV for the so-calledlow state. Such a spectrum resembles those in blazars or mi-croquasars, and might suggest at work by the models ac-cordingly. However, an analysis using a microquasar modelconsisting on a magnetized conical jet filled with relativisticelectrons, shows that it is hard to comply with all observa-tional constrains. This fact and the non-detection at hard X-rays introduce an a-posteriori question mark upon the phys-ical reality of this source, what is discussed here.

Keywords γ -Rays · Unidentified γ -ray sources

1 Introduction

GRO J1411-64 is the strongest variable unidentified MeVsource located near the Galactic plane. It was discovered byCOMPTEL/CGRO during 1995 March–July (viewing pe-riods 414-424), during which the source went on a burstevent at MeV energies (Zhang et al. 2002). The sourcewas detected at ∼7σ in the 1–3 MeV band by combiningthe 7 viewing periods (VPs, the periods of observations inCGRO), according to which the best location was measuredat (l,b) = (311.5◦, −2.5◦) and the source was referred asGRO J1411-64. The flare duration was several months and


the rather steep spectral shape obtained while the source wasflaring would predict a bright, hard X-ray source, if there isno break in the spectrum, which is explored here. In whatfollows, we present the results of the INTEGRAL observa-tions of this source, as well as of XMM-Newton observationof its best location and, following, the comment concerningthe possible nature of this source.

2 Observation and data analysis

GRO J1411-64 was observed by INTEGRAL during 2004December 30–2005 January 6. In total, 102 science windows(scws) were carried out to have 210 ks of effective exposure.Data reduction was performed using the version 5.0 of thestandard Offline Science Analysis (OSA) software, and thespectra were fitted with XSPEC of FTOOLS 5.3.1.

The best localization of COMPTEL source GRO J1411-64 was observed with XMM-Newton during revolution 960on the 7th of March of 2005 (Obs. ID: 0204010101). Thedata were pipeline-processed with the XMM-Newton Sci-ence Analysis Software (SAS) version 6.1. After removal ofbackground flares, a total of 15.8, 15.8 and 14.6 ks of gooddata survived for MOS1, MOS2 and pn respectively.

3 Results

3.1 Hard X-rays

No hint of signal was found for new hard X-ray sourceswithin the location uncertainty of GRO J1411-64 from indi-vidual scws of the INTEGRAL instruments. To improve thestatistics, mosaic maps were obtained for IBIS/ISGRI andJEMX by combining all data. The images of IBIS/ISGRIwere produced in the energies 20–100 keV, see Fig. 1 for themap in the 20–40 keV band as an example. The circle holdsthe 4-σ error region of GRO J1411-64 obtained by COMP-TEL during its flare in 1995 (Zhang et al. 2002). From thepossible counterparts of GRO J1411-64 discussed in Zhanget al.’s paper (2002), only the Circinus Galaxy shows up inthis error region as seen by INTEGRAL. The most signif-icant detection of Circinus Galaxy is in the energies 20–40 keV, at a confidence level of 38σ . The mosaic map ofJEMX shows no significant source feature is visible fromwithin the 4-σ error region. For SPI, the Circinus Galaxy isat the 6σ level in the 20–40 keV range, and it is the onlysource detected within the location of GRO J1411-64, theregion of our search.

The light curve for the Circinus Galaxy, detected mainlyby IBIS/ISGRI, is rather constant. The Circinus Galaxy wasinvestigated in (Soldi et al. 2005). Models of cutoffpl pluswabs in XSPEC can fit the data well, with a reduced χ2

Fig. 1 Sky map of the GRO J1411-64 region as seen by IBIS/ISGRIin the 20–40 keV range, by combining all data obtained in the obser-vations performed during 2004 December 30 to 2005 January 6

Fig. 2 Light curve of GRO J1411-64 as observed by COMPTEL at0.75–1 MeV band. Each bin is averaged over one CGRO Phase, withthe typical time scale of one year. The error bar is 1σ and upper limit2σ . These data points include the 7 viewing periods when the sourcewas flaring

of 1.1 (7 dof). The resulting parameters are consistent withthose in Soldi et al.

GRO J1411-64 shows likely persistent emission in 0.75–1 MeV band during its low state (Fig. 2), where the sourcewas detected at ∼4σ by COMPTEL (Fig. 3). The corre-sponding spectrum of the low state can be represented by


Fig. 3 The skymap of GRO J1411-64 as observed by COMPTEL in0.75–1 MeV during 1991–1996, not including the flare period of 4months in 1995. The star represents the best-guessed source location.The contour lines start at a detection significance level of 3σ with stepsof 0.5σ

a power law shape with spectral index 2.5+0.6−0.4 (see Fig. 4).

Circinus Galaxy can be safely ruled out as the counterpartdue to its spectral extrapolation well below the ones at MeVenergies. The ISGRI/SPI upper limits combined to spectraof both flare/low states shows the existence of a maximumin the power output at hard X-rays, which might remind usto consider the microblazar as the possible source nature.

3.2 Soft X-rays

A total of 31 X-ray sources were formally detected by theSAS source detection algorithm in the EPIC data. Nine ofthese were excluded due to detector defects and other arti-facts, in a careful inspection. The resulting 22 reliable andstatistically significant sources (likelihood >10) are shownin Fig. 5. Among them, the unfolded spectrum (largely inde-pendent of the model fitted) for XMMU J141255.6-635932,along with the best fit model, the COMPTEL detectionsand the INTEGRAL upper limits is plotted in Fig. 6. Thehard excess exhibited by the XMM-Newton data is appar-ent in that figure, and might be suggestive of a large Comp-ton bump that would peak in the several ∼100 keV region,fitting well with the COMPTEL detections. However, thefact that the XMM-Newton image does not cover the fullCOMPTEL source location and the non-detection by IN-TEGRAL of any reliable counterpart, would make the as-sumption that XMMU J141255.6-635932 is the counterpartto GRO J1411-64, although spectrally consistent, only ten-tative and risky.

Fig. 4 Combined energy spectrum of GRO J1411-64. Filled (open)circles represent flare (low) states at MeV energies, solid line for theflare state, dashed line for the low state, and the 2σ upper limits ob-tained from IBIS/ISGRI (triangles) and SPI (squares). The solid curveat low energies is the energy spectrum of Circinus Galaxy derived byfitting the IBIS/ISGRI data. The solid and dashed lines at high energiesare the fitted spectra from COMPTEL

Fig. 5 XMM-Newton EPIC image (combining all 3 cameras), wheredetected sources and previously catalogued sources have been labeled.The centroid of the COMPTEL source GRO 1411-64 is marked witha square box, the error contour being larger than the image itself. Seedetails in Torres et al. (2006)

4 Conclusion and summary

The observations, subsequent analysis and theoretical in-vestigations pursued shed light upon the nature of GRO


Fig. 6 XMM-Newton unfolded spectrum of the X-ray source XMMUJ141255.6-635932. The model shown is only the thermal componentin the X-ray spectrum. The COMPTEL detections and the INTEGRALupper limits are also shown at high energies, with horizontal bars de-noting 2σ upper limits

Fig. 7 A microquasar model on the light of observational constraints.See Torres et al. (2006) for details

J1411-64. The combined INTEGRAL, XMM-Newton andCOMPTEL observations reveal no obvious counterpart athigh energies (hard X-rays and gamma-rays). Nevertheless,the unique peak of the power output at these energies resem-bles the SED seen in microquasars, and suggests at work bythe models accordingly. However, an analysis using a micro-quasar model consisting on a magnetized conical jet filledwith relativistic electrons which radiate through synchrotronand inverse Compton scattering with star, disk, corona andsynchrotron photons shows that it is hard to comply with allobservational constrains (Fig. 7). The best fit parameters seeTable 1. This fact and the non-detection at hard X-rays in-troduce an a-posteriori question mark upon the physical re-

Table 1 Parameter values for GRO J1411-64. At the top of table, pa-rameter values for a typical microquasar system and jet geometry aregiven (Bosch-Ramon et al. 2006). We have considered different val-ues within the range open for the free parameters finding that it is notpossible to obtain a simple microquasar model that could fit the SED.In particular, in Fig. 7 we show a test case with the free parametersfixed to the values presented in this table, at the bottom. We note that,since the computed SED in Fig. 7 is dominated in the gamma-ray bandby SSC emission, the model results would also apply for a low massmicroquasar

Parameter Values

Stellar bolometric luminosity [erg s−1] 1038

Apex dis. to the comp. obj. [cm] 5 × 107

Initial jet radius [cm] 5 × 106

Orbital radius [cm] 3 × 1012

Viewing angle to the axis of the jet [◦] 45

Jet Lorentz factor 1.2

Jet leptonic kinetic luminosity [erg s−1] 3 × 1035

Maximum electron Lorentz factor (jet frame) 5×102

Maximum magnetic field [G] 8000

Electron power-law index 1.5

Total corona luminosity [erg s−1] 3 × 1033

ality of this source. See more details in (Torres et al. 2006).GLAST observations would help improving the location ofthe MeV source if radiation at higher energies is not com-pletely suppressed, and would open the door for more effi-cient multiwavelength searches of the counterpart. However,it is true that the nature of this COMPTEL source might notbe constrained further if this detection was a one-time onlytransient phenomena. GLAST will only be able to help if acandidate counterpart is caught in the act (flaring/quiescentstate of an AGN or a more rare galactic object). Having athand GLAST observations, in any case, will make our cur-rently reported investigation to naturally fit into the testingof any hypothesis on the nature of GRO J1411-64.

Acknowledgements We thank Dr. M.T. Ceballos for her help withthe XMM-Newton data. DFT has been supported by Ministerio deEducación y Ciencia (Spain) under grant AYA-2006-0530, as wellas additional support from the Guggenheim Foundation. S. Zhangwas subsidized by the Special Funds for Major State Basic ResearchProjects and by the National Natural Science Foundation of China. XBand AC were financially supported for this research by the Ministe-rio de Educación y Ciencia (Spain), under project ESP2003-00812.VB-R and JMP have been supported by Ministerio de Educacióny Ciencia (Spain) under grant AYA-2004-07171-C02-01, as well asadditional support from the European Regional Development Fund(ERDF/FEDER). VB-R has been additionally supported by the DGI ofthe Ministerio de (Spain) under the fellowship BES-2002-2699. GERwas supported by grants PIP 5375 y PICT 03-13291.


References

Bosch-Ramon, V., Paredes, J.M., Romero, G.E., Torres, D.F.: A micro-quasar model applied to unidentified gamma-ray sources. Astron.Astrophys. 446, 1081 (2006)

Soldi, S., Beckmann, V., et al.: INTEGRAL observations of six AGNin the Galactic Plane. Astron. Astrophys. 444, 431 (2005)

Torres, D.F., Zhang, S., et al.: INTEGRAL and XMM-Newton observa-tions towards the unidentified MeV source GRO J1411-64. Astron.Astrophys. 457, 257 (2006)

Zhang, S., Collmar, W., Schönfelder, V.: An unidentified variablegamma-ray source near the galactic plane detected by COMPTEL.Astron. Astrophys. 396, 923 (2002)



Evidence for a new MeV source observed by theCOMPTEL experiment aboard CGRO

Shu Zhang · Werner Collmar


Abstract We report first evidence for a new unidentifiedand variable MeV source, located near the galactic plane at(l,b) ∼ (284.5°, 2.5°). The source, GRO J1036-55, is foundat a significance level of ∼5.6σ by COMPTEL in its 3–10 MeV band. The energy spectrum indicates a spectralmaximum at 3–4.3 MeV with a steep slope at higher ener-gies. Since the COMPTEL 3–4.3 MeV data contain conta-mination by an instrumental background line, we performedseveral consistency checks, which all are consistent with anastrophysical nature of this emission feature.

Keywords γ -Rays · Unidentified γ -ray sources

1 Introduction

An important discovery of the Compton Gamma-Ray Ob-servatory (CGRO) is the detection of a large number ofunidentified γ -ray sources. The EGRET experiment aboardCGRO, measuring at energies above 100 MeV, detected 271γ -ray sources, of which 171 are unidentified (Hartman et al.1999). The Compton telescope COMPTEL aboard CGROwas measuring γ -rays between ∼0.75 and 30 MeV with anenergy-dependent energy and angular resolution of 5–8%(FWHM) and 1.7–4.4° (FWHM), respectively. Imaging inits circular field-of-view of ∼1 steradian was possible with

S. Zhang (�)High Energy Astrophysics Lab, Institute of High Energy Physics,P.O. Box 918-3, Beijing 100049, Chinae-mail: [email protected]

W. CollmarMax-Planck-Institut für extraterrestrische Physik, Garching,Germany

a location accuracy (flux dependent) of the order of 1–3°.For details on COMPTEL see Schönfelder et al. (1993).

During its mission of 9 years (April 1991 to June 2000)COMPTEL has now detected 11 unidentified γ -ray sources(e.g., Schönfelder et al. 2000). Six are located at low Galac-tic latitudes (|b| < 10°) and the rest at high Galactic latitudes(see Table 1). This paper presents first evidence for an addi-tional unidentified MeV source.

2 Observation and data analysis

For our analyses on the possible new MeV source, we se-lected all CGRO viewing periods (VPs) before the sec-ond reboost of CGRO in April 1997 (changing the COMP-TEL background environment) in which the new source waswithin 30° of the COMPTEL pointing direction (see Ta-ble 2). For reliability investigations, we analysed three ad-ditional VPs (see Table 3), where the source was outsidethe COMPTEL field of view, but which are either close intime to our flaring observation (VPs 530.0, 601.1) or con-tain the Crab as a calibration source (VP 413). All theseanalyses were consistently carried out by using the stan-dard COMPTEL maximum-likelihood analysis procedure,including the standard background generation technique,and models for subtraction of the Galactic and extra-galacticdiffuse γ -ray emissions. In addition, we investigated theCOMPTEL events of this emission feature directly, by gen-erating event distributions with respect to several measuredevent parameters.


Table 1 List of unidentified MeV sources detected by COMPTELduring its mission. The abbreviations have the following meanings:‘S’ means stable, ‘V’ variable, ‘P’ point, and ‘E’ extended source

Name l b Energy (MeV) Note

(degree) ( >3σ )

GRO J1823-12 18.5 −0.5 1–30 S, P

GRO J2227+61 106.6 3.1 1–3 V, P

GRO J0241+6119 135.7 1.1 1–30 V, P

GRO J1411-64 311.5 −2.5 0.75–3 V, P

GRO J1743-30 358.5 −0.5 1–30 S, P

Carina/Vela reg. 273 −6 3–10 S, E

GRO J1753+57 85.5 30.5 1–3 V, E

GRO J1040+48 165 57 0.75–3 V, P

GRO J1214+06 278.9 66.6 3–10 V, P

HVC M & A 145–195 35–65 0.75–3 S, E

HVC C 75–95 25–45 0.75–3 S, E

Table 2 List of CGRO VPs, for which the COMPTEL pointing di-rection was within 30° to GRO J1036-55. The VP numbers, their timeperiods in calendar date, pointing offset angles, and effective on-sourceexposures are given

VP Date Offset Effective

(dd/mm/yy) angle exposure

days

8.0 22 August 91–05 September 91 23° 2.84

14.0 14 November 91–28 November 91 3° 3.03

32.0 25 June 92–02 July 92 20° 0.92

208.0 02 February 93–09 February 93 29° 0.80

230.0 27 July 93–30 July 93 9° 0.56

230.5 30 July 93–03 August 93 6° 0.80

301.0 17 August 93–24 August 93 21° 0.77

314.0 03 January 94–16 January 94 20° 2.44

315.0 16 January 94–23 January 94 20° 1.17

316.0 23 January 94–01 February 94 30° 1.22

338.5 31 August 94–30 September 94 21° 3.62

402.0 18 October 94–25 October 94 27° 0.90

402.5 25 October 94–01 November 94 23° 0.98

414.0 21 March 95–29 March 95 16° 1.12

415.0 11 April 95–25 April 95 28° 1.88

522.0 11 June 96–14 June 96 4° 0.72

531.0 03 October 96–15 October 96 3° 2.32

3 Results

3.1 Detections

Evidence for a possible new MeV source was first found ina 3–10 MeV skymap (Fig. 1) of CGRO Phase 5, an obser-vational period of about one year, containing 36 individual

Table 3 COMPTEL observational periods in which GRO J1036-55was outside the field of view. The VP number, their time periods incalendar date, and pointing directions are given

VP Date Pointing direction

(dd/mm/yy) (l, b)

413.0 07 March 95–21 March 95 (191.8, −3.4)

530.0 06 September 96–03 October 96 (124.7, 6.4)

601.1 15 October 96–29 October 96 (70.1, −10.5)

Fig. 1 COMPTEL 3–10 MeV skymap for CGRO Phase 5 (including36 VPs). The MeV excess near (l/b) = (284.5°/2.5°) is obvious (�).The contour lines start at a likelihood ratio of 6 with steps of 3

VPs. Subsequent searches revealed that the emission mainlycomes from VP 531 (Fig. 2), a typical two-week VP. Thecenter of the emission is located at (l, b) ∼ (284.5°, 2.5°).In order to derive the MeV properties of this source, weinvestigated this sky region in detail by using all possibleCGRO VPs up to the second reboost of CGRO. We foundthat the source is only visible during VP 531 (3 to 15 Octo-ber 1996), i.e. showing a flare in this period. The likelihoodratio, −2 lnλ, for a source detection in the 3–10 MeV bandis 38.5, which corresponds to 5.6σ by assuming three de-grees of freedom. The probability of detecting randomly anunknown source at this significance level is 2 × 10−5, bytaking into account the trials for searching all CGRO VPsin four energy bands. During this flare the source reached aflux level of 350 mCrab in the 3–10 MeV band. However,the source is only marginally detected at lower COMPTELenergies, and not at all at higher (10–30 MeV) energies. Ananalysis of the simultaneous EGRET data (>100 MeV) of


Fig. 2 COMPTEL 3–10 MeV skymap of VP 531. The MeV excessnear (l/b) = (284.5°/2.5°) is obvious (�). The contour lines start at alikelihood ratio of 6 with steps of 3

Fig. 3 COMPTEL 3–4.3 MeV skymap of VP 531. The contour linesstart at a likelihood ratio of 6 with steps of 3

VP 531 did not yield any evidence for the source. By sub-dividing the 3–10 MeV band into smaller ones, we derive∼4.3σ and ∼2.5σ detections in the 3–4.3 MeV and 4.3–9 MeV bands, respectively (Figs. 3, 4). By combining all

Fig. 4 COMPTEL 4.3–9 MeV skymap of VP 531. The contour linesstart at a likelihood ratio of 6 with steps of 3

non-flare data (i.e., all VPs apart VP 531), we only find a∼3σ -feature in the 1–3 MeV at the source position.

3.2 Reliability investigations

It is known that activated 24Na in a cascade process will gen-erate two γ -rays at energies of 1.368 and 2.754 MeV. Suchinstrumental background photons, if emitted in specific di-rections, will be recorded by COMPTEL as a valid event ofenergy ∼4.1 MeV by having a ϕ value of ∼18° or ∼40°. Toconvince ourselves that this 3–10 MeV feature is not a back-ground artifact due to these background lines, we carried outseveral analyses.

(1) We analysed the VP 531 data in the 3–3.9 and 3.9–4.3 MeV bands, of which the latter one covers the 24Na cas-cade line. We derived detection significances of ∼3σ (3 dof)in each band, showing that the observed signal is not—atleast not solely—due to 24Na.

(2) We checked the significance and flux dependenciesof the emission feature on ϕ, and compared them to theones derived on the Crab in an observation (VP 413), whichhad similar properties (duration, source offset angle). In bothcases the distributions are similar.

(3) To check for a time-dependent instrumental back-ground feature, we analysed the 3–4.3 MeV data of VP 531and its two neighboring VPs, VP 530.0 (the 27 days before)and VP 601.1 (the 14 days after) not pointing to the newsource, in an instrumental coordinate system. In such a sys-tem, a certain instrumental effect should show up at the same


Fig. 5 3–10 MeV light curve of GRO J1036-55 up to the second re-boost of CGRO. Each data point represents an individual VP. The errorbars are 1σ

position in an ‘instrumental’ skymap. While GRO J1036-55provides a significant signal, the other two maps are emptyat this location.

(4) We generated distributions of the events associatedwith GRO J1036-55 with respect to several measured eventparameters (e.g. time-of-flight). These basic event distrib-utions do not obviously differ from the ones generated forthe Crab and 3C 273, by assuming similar conditions (e.g.,observation time).

Given the consistent results of our checks, we considerit unlikely, that the new MeV source feature is the result ofan instrumental effect. However, we can not rule out thispossibility finally.

3.3 Variability

The 3–10 MeV light curve indicates a MeV flare of GROJ1036-55 during VP 531 (Fig. 5). Fitting these 3–10 MeVfluxes with a constant flux level, results in a χ2 value of 55.This corresponds to a significance of 4.6σ that the sourceis variable, and—subsequently—to a probability of ∼3.6 ×10−6 that the source is non-variable.

Fig. 6 Combined COMPTEL/EGRET spectra of GRO J1036-55 forVP 531 and averaged over Phases 1–4. The error bars are 1σ and theupper limits 2σ

3.4 Energy spectra

We derived the COMPTEL spectra of GRO J1036-55 in thefour standard energy bands for VP 531 and for the sum ofall data for the CGRO Phases 1 to 4, which cover about thefirst 4.5 years of the mission. Figure 6 shows these spectracombined with the simultaneously derived EGRET fluxes(>100 MeV). The flare state in VP 531 and a soft spec-tral shape above 10 MeV is obvious. Figure 7 shows againthe COMPTEL/EGRET spectrum of VP 531, however withhigher energy resolution of the COMPTEL data, indicatinga spectral maximum at 3–4.3 MeV.

4 Summary and conclusion

We report first evidence (5.6σ ) for a new flaring MeVsources, which is located near the Galactic plane and showeda MeV flare in October 1996. Due to the fact, that the mainemission arises between 3 and 4.3 MeV, where COMPTELhas an instrumental background line, we performed severaltests on the data. All of them are consistent with the fact thatthis significant emission feature is due to an astrophysicalsource.


Fig. 7 COMPTEL/EGRET spectrum of GRO J1036-55 for VP 531with increased energy resolution in the COMPTEL band. The errorbars are 1σ and the upper limits 2σ

The MeV spectrum during the flare, although its shape isnot statistically significant determined, reminds on a predic-tion of Punsly et al. (2000). They calculated the broadbandspectrum of an accreting isolated Galactic Kerr-Newmanblack hole, by assuming a jet scenario including electron-positron annihilation effects. They predict an emission peakat MeV energies with a very steep spectrum at energiesabove the peak, i.e. similar to our flare spectrum. Prelimi-nary counterpart searches have not yielded an obvious coun-terpart.

Acknowledgements This research was supported by the Germangovernment through DLR grant 50 QV 9096 8. S. Zhang was subsi-dized by the Special Funds for Major State Basic Research Projectsand by the National Natural Science Foundation of China.

References

Hartman, R.C., Bertsch, D.L., Bloom, S.D., et al.: Astrophys. J. Suppl.Ser. 123, 79 (1999)

Schönfelder, V., Aarts, H., Bennett, K., et al.: Astrophys. J. Suppl. Ser.86, 657 (1993)

Schönfelder, V., Bennett, K., Bloemen, H., et al.: Astron. Astrophys.Suppl. Ser. 143, 145 (2000)

Punsly, B., Romero, G.E., Torres, D.F., et al.: Astron. Astrophys. 364,552 (2000)

Astrophysics and Space Science - Startseite...Astrophysics and Cosmology Stanford University CA...

Documents

Transcript of Astrophysics and Space Science - Startseite...Astrophysics and Cosmology Stanford University CA...