1: %\documentstyle[11pt, aaspp4,epsf,rotate]{article}
2: %\documentclass[preprint2]{aastex}
3: %\documentclass[emulateapj]
4: \documentclass[12pt, preprint]{aastex}
5: \usepackage{graphicx}
6: \sloppy
7: \nonfrenchspacing
8:
9: \def\zdot{$\dot{\rm z}$}
10: \def\gmin{\gamma_{\rm min}}
11: \def\gmax{\gamma_{\rm max}}
12: \def\gimin{\gamma_{\rm i, min}}
13: \def\gimax{\gamma_{\rm i, max}}
14: \def\gad{\gamma_{\rm ad}}
15: \def\fesy{f_{\epsilon}^{\rm sy}}
16: \def\feec{f_{\epsilon}^{\rm EC}}
17: \def\estar{\epsilon_{\ast}}
18: \def\eext{\epsilon_{\rm ext}}
19: \def\sigmat{\sigma_{\rm T}}
20:
21: \slugcomment{accepted for publication in ApJL}
22:
23: \begin{document}
24:
25: \title{A Time Dependent Leptonic Model for Microquasar Jets: Application
26: to LSI +61 303}
27:
28: \author{S. Gupta and M. B\"{o}ttcher}
29: \affil{Astrophysical Institute, Department of Physics and Astronomy,
30: Clippinger Hall 251B, Ohio University, Athens, OH 45701 -- 2979, USA}
31:
32: \begin{abstract}
33: The Galactic high-mass X-ray binary and jet source (microquasar) LSI~+61~303
34: has recently been detected at TeV $\gamma$-ray energies by the MAGIC telescope.
35: We have applied a time-dependent leptonic jet model to the broadband spectral
36: energy distribution and suggested (though not unambiguously detected)
37: orbital modulation of the very high energy $\gamma$-ray
38: emission of this source. Our model takes into account time dependent electron
39: injection and acceleration, and the adiabatic and radiative cooling of non-thermal
40: electrons. It includes synchrotron, synchrotron self-Compton and external
41: inverse Compton (with seed photons from the companion star),
42: as well as $\gamma\gamma$ absorption of $\gamma$-rays by starlight photons.
43: The model can successfully reproduce the available multiwavelength observational
44: data. Our best fit to the SED indicates that a magnetic field of $B_0 \sim 5 \times
45: 10^3$~G at $\sim 10^3 \, R_g$ is required, and electrons need to be accelerated out
46: to TeV energies ($\gamma_2 = 10^6$) with a nonthermal injection spectrum with a
47: spectral index of $q = 1.7$, indicating the operation of acceleration mechanisms
48: beyond the standard first-order Fermi mechanism at relativistic or non-relativistic
49: shocks. The orbital modulation of the VHE $\gamma$-ray emission can be explained solely
50: by the geometrical effect of changes in the relative orientation of the stellar
51: companion with respect to the compact object and jet as it impacts the position
52: and depth of the $\gamma\gamma$ absorption trough.
53: Such a scenario predicts a trend of spectral hardening during VHE $\gamma$-ray low orbital phases.
54: \end{abstract}
55:
56: \keywords{radiation mechanisms: non-thermal --- gamma-rays: theory ---
57: X-rays: binaries}
58:
59: \section{Introduction}
60: X-ray binaries with relativistic jets, or microquasars, have recently been
61: established as a new class of $\gamma$-ray emitting sources. Two sources
62: detected by EGRET on board the {\it Compton Gamma-Ray Observatory} were
63: found to be spatially consistent with the locations of microquasars, namely
64: LS 5039 and LSI +61 303 \citep{hartman99}. Observations of $\gtrsim$ 250 GeV
65: $\gamma$-rays from LS 5039 with the High Energy Stereoscopic System (HESS) have
66: recently shown this high-mass source to be a TeV emitter \citep{aharonian05},
67: also strengthening its earlier tentative identification with the EGRET source
68: 3EG J1824-1514 \citep{paredes2000}. The second source, LSI +61 303, associated
69: with the COS-B source 2CG 135+01 \citep{hermsen77, gt78} and the EGRET source
70: 3EG J0241+6103 \citep{kniffen97}, was recently observed with the Major Atmospheric
71: Gamma-ray Imaging Chrenkov (MAGIC) telescope by \cite{albert06}. They detected
72: variable $\gamma$-ray emission above 100 GeV over six orbital cycles,
73: suggesting a periodic modulation on the time scale of the orbital period of $29.496$~d,
74: though further observations are necessary to firmy establish the correlation
75: between the VHE variability and the orbital period.
76: The strongest detections were not at the periastron, suggesting that the modulation
77: might not be related to a modulation of the accretion rate
78: \citep[as suggested
79: by, e.g.,][]{romero03,bosch04b},
80: but rather to geometrical effects. Geometrical
81: effects causing an orbital modulation of the high-energy emission include
82: the azimuthal-angle dependence of the $\gamma\gamma$ absorption of high-energy
83: emission by starlight photons \citep{boettcher05,dubus06} or of the Compton upscattering
84: of starlight photons by relativistic electrons in the microquasar jet \citep{dermer06}.
85: \cite{dubus06} has performed a detailed analysis of the $\gamma\gamma$
86: absorption of VHE photons in the photon field of the stellar companion, taking
87: into account the finite size of the star and the eccentricity of the orbit for the
88: case of VHE emission from the surface of a putative neutron star associated with
89: LSI +61 303. Note that he used a definition of the phase such that phase 0
90: corresponds to the periastron passage, so that his results should be shifted
91: by $\Delta\psi = 0.23$ (see below) when compared to the definition commonly
92: used throughout the literature on this object, including this {\it Letter}.
93: He found that the $\gamma\gamma$ absorption depth is expected to show a
94: pronounced maximum at phases just before the periastron passage, when
95: the compact object is located behind the stellar companion. The recent
96: VHE detections, along with the observation
97: of X-ray jet structures in several
98: microquasars using $Chandra$ and $XMM-Newton$ \citep[e.g.,]{corbel02, tomsick02},
99: have re-ignited interest in jet models for the high energy emission from microquasars,
100: analogous to the commonly adopted models for blazars \citep[for a recent review see,
101: e.g.][]{boettcher02}.
102:
103: In the leptonic model of microquasars, Very High Energy (VHE) emission might most
104: likely originate near the base of the mildly relativistic jet. The soft photons
105: from the companion star, the accretion disk as well as from jet synchrotron
106: radiation can be Compton upscattered by the ultra-relativistic electrons in
107: the jet. Steady-state leptonic jet models of the $\gamma$-ray emission from microquasars
108: have been presented, e.g., by \cite{bosch04a,bosch04b}, and \citep{dermer06}.
109: A time-dependent, broadband leptonic jet model was presented in \cite{gupta06},
110: (henceforth, Paper 1) where an analytical solution to the electron kinetic
111: equation was presented, restricting the analysis to Compton scattering in the
112: Thomson regime. This $Letter$ follows up on the analysis of Paper 1, now
113: incorporating a full Klein-Nishina treatment of the Compton scattering as
114: well as the angle dependence of the stellar radiation field. We then apply
115: the new model to broadband observations of LSI +61 303, iincluding the effect
116: of the orbital modulation on the $\gamma\gamma$ absorption in the $>$100GeV
117: range. In \S \ref{model}, we present a general outline of the model geometry
118: and the radiative processes involved. \S \ref{application} shows the application
119: to LSI +61 303, and we conclude with a brief summary in \S \ref{summary}.
120:
121:
122: \section{\label{model}Model Description}
123:
124:
125: The accretion flow onto the central compact object is ejecting a twin
126: pair of jets, assumed to be oriented perpendicular to the orbital plane,
127: which is inclined with respect to the line of sight by an angle $i$.
128: Two intrinsically identical disturbances, containing non-thermal
129: plasma (blobs) originate from the central source at the same time,
130: traveling in opposite directions along the jet at a constant speed
131: $v_j = \beta_j \, c$. Over a limited range in distance $x^{}_0 \le x^{}
132: \le x^{}_1$, relativistic electrons are accelerated and injected
133: in the emission region with an exponentially cut-off power-law distribution
134: in electron energies ($E_e = \gamma \, m_e c^2$) with low- and high-energy
135: cutoffs $\gamma_1$ and $\gamma_2$, respectively, in the co-moving frame:
136:
137: \begin{equation}
138: Q_{e}^{\rm inj}(\gamma;t) = Q_{0}^{inj}(t)\gamma^{-q} e^{-\gamma/\gamma_2},
139: \;\;\; \gamma_1\leq\gamma.
140: \end{equation}
141: The normalization factor $Q_e^{\rm inj}$ is approximately related to the
142: power injected into relativistic electrons, $L_{\rm inj}$ through
143:
144: \begin{equation}
145: Q_0 \approx \cases{
146: \frac{L_{\rm inj}(2-q)}{m_ec^2(\gamma_{max}^{2-q}-\gamma_{min}^{2-q})} &
147: \mbox{if $q \not= 2$};
148: \cr
149: \frac{L_{inj}}{m_ec^2 ln(\gamma_{max}/\gamma_{min})} & \mbox{if $q=2$}.
150: \cr}
151: \end{equation}
152:
153: The injection luminosity $L_{\rm inj}$ and the functional dependence
154: of Eq. (1) are held constant between $x_0$ and $x_1$.
155: The blob's (transverse) radius, $R_{\perp}$, scales with distance from the
156: central engine as $R_{\perp} = R_{\perp}^0 \, (x / x_0)^{\alpha}$, i.e.,
157: $\alpha = 0$ corresponds to perfect collimation, and $\alpha = 1$ describes
158: a conical jet. Following the arguments given in \cite{aa97}, we choose a
159: magnetic-field dependence on distance from the central black hole as
160: $B (x) = B_0 \, (R_{\perp}/R_{\perp}^0)^{-2} = B_0 \, (x / x_0)^{-2 \alpha}$.
161:
162: \begin{figure}[t]
163: \includegraphics[height=12cm]{f1.eps}
164: \caption{Sketch of the geometry of the binary orbit of LSI +61 303, approximating
165: the massive (Be) star to be stationary. The labels on the compact-object orbit
166: indicate the orbital phase $\psi$ as well as the azimuthal angle with respect to
167: the apastron, $\chi$.}
168: \label{geometry}
169: \end{figure}
170:
171: The geometry of the orbit of the LSI~+61~303 system is illustrated in
172: Fig. \ref{geometry}. In order to incorporate the orbital modulation of
173: the Compton scattering, we replaced the circular orbit of the binary
174: system in Paper 1 by an elliptical one, so that the distance of the star
175: to the compact object now varies as
176: \begin{equation}
177: r(\chi) = \frac{a(1-e^2)} {1-e\cos(\chi)}
178: \label{r_chi}
179: \end{equation}
180: where $a$ is the semi-major axis, $e$ is the eccentricity of the orbit, and
181: $\chi$ is the phase angle \citep[which is {\it not} linearly related to the
182: orbital phase $\psi$, in contrast to the incorrect expression in][]{romero05}.
183: The $\gamma\gamma$ opacity as a function of $\gamma$-ray photon energy and
184: orbital phase has been calculated following \cite{boettcher05}.
185:
186: To treat the time-dependent electron dynamics and the radiation transfer in the
187: emitting blob, we solve the continuity equation for the relativistic electrons,
188: \begin{equation}
189: {\partial n_e(\gamma,t) \over \partial t}=-\frac {\partial}{\partial \gamma}\Big [
190: \Big ( \frac{d\gamma}{dt} \Big)_{loss} n_e(\gamma,t) \Big ] + Q_e^{\rm inj} (\gamma,t)
191: \end{equation}
192:
193: where, $(d\gamma/dt)_{loss}$ is the
194: (radiative and adiabatic)
195: energy loss rate for the electrons.
196: The electron cooling rates are calculated using subroutines used in the jet
197: radiation transfer code of \cite{bms97} and \cite{boettcher2000}. The discretized
198: electron continuity equation can be written in the form of a tridiagonal matrix
199: \citep{chiaberge99}, and solved using the standard routine of \cite{press92}.
200:
201: Radiation mechanisms included in our simulations are synchrotron emission,
202: Compton upscattering of synchrotron photons, namely synchrotron self-Compton
203: (SSC) emission, and Compton upscattering of external photons.
204: The dominant external soft photon source is the companion star
205: (for details, refer to Paper 1) which
206: is approximated as a point
207: source, emitting a blackbody with dimensionless temperature
208: $\Theta_*=kT_*/m_{e}c^2$.
209:
210:
211: \section{\label{application} Application to LSI +61 303}
212:
213: \subsection{\emph {Parameter Selection}}
214:
215: The high mass XRB, LSI +61 303, at a distance of 2 kpc \citep{hutchings81, frail91}
216: has been observed from radio \citep{paredes96} to TeV $\gamma$-rays
217: \citep{albert06}, providing a wealth of broadband data to test various emission
218: models. However, most of the models until now dealt more with the radio, X-ray, or
219: $\gamma$-ray variability of the source, than with its broadband properties and its
220: very-high-energy emission. Recently, however, \cite{chernyakova06} reproduced the
221: broadband spectrum of the system, including the recent MAGIC detection, using a model
222: of a rotation powered pulsar. The (slight) orbital modulation of the X-ray and
223: $\gamma$-ray emission in the {\it XMM-Newton} and {\it INTEGRAL} energy ranges was
224: explained based on a variable injection rate of high energy electrons. The VHE
225: $\gamma$-ray emission was attributed to hadronic processes, initiated by high-energy
226: protons in the pulsar wind, and its orbital modulation was not explained in that model.
227: In contrast, in this {\it Letter}, we are specifically addressing the broadband
228: spectral characteristics {\it and} the orbital modulation of the VHE $\gamma$-ray
229: emission.
230:
231: \cite{casares05} have found a mass function of $f(M) \approx 0.011 \, M_{\odot}$.
232: Since the inclination angle $i$ between the line of sight and the normal to the
233: orbital plane (which, in our model is identical to the jet axis) is poorly
234: constrained, and also the mass of the B0 V type stellar companion (with a
235: dense equatorial wind) is not well known and in the range of $M_{\ast} \sim$
236: 10 -- $15 \, M_{\odot}$, a rather wide range of masses for the compact object
237: is consistent with the observational constraints, even including a $1.4 \,
238: M_\odot$ neutron star \citep{hutchings81}. In our model, we follow the suggestion
239: of \cite{casares05} of $i = 30^o$ and adopt a companion mass of $\sim 12M_\odot$,
240: which would yield $M_X \sim 2.67 \, M_{\odot}$. The orbit has an eccentricity of
241: $e=0.72\pm0.15$, the orbital period is $\sim$~26.496~d \citep{gregory02}, and the orbital semi
242: major axis is $a = 5 \times 10^{12}$cm. MERLIN observations by \cite{massi04}
243: imply a jet Lorentz factor of $\Gamma_{jet} = 1.25$.
244:
245: In our fitting procedure, we are varying the following parameters: the low
246: and high energy cut-off of the electron injection spectrum $\gamma_1$ and
247: $\gamma_2$, respectively, the injection spectral index $q$, the initial distance
248: of the injection zone from the compact object $x_0$, and the initial magnetic
249: field $B_0$. The straight power-law shape of the X-ray spectrum indicates that any
250: disk blackbody component from an accretion disk around the compact object
251: should have a luminosity of $L_D \lesssim 10^{34}$~ergs~s$^{-1}$ and be
252: at least $\sim 4$ orders of magnitude lower than the luminosity of the
253: companion star. For such low disk luminosities, external Compton scattering
254: of accretion disk photons will be negligible as well.
255: The evolution of
256: the electron distribution in the emitting region is followed over a period of
257: $\sim$~5 days ($\ll P_{orb}$). Since most of the high-energy emission emanates
258: from the system within the first few seconds of electron evolution, the orbital
259: phase is taken to be a constant for each run of our model. Also, the centre of
260: mass of the system lies close to the massive ($12 \, M_\odot$) stellar
261: companion, so that the compact object essentially orbits the stellar companion,
262: which is assumed to be stationary.
263:
264: For an initial spectral fit, we concentrate on the orbital phase during which
265: MAGIC significantly detected the source. Specifically, we chose $\psi = 0.5$,
266: corresponding to a phase angle $\chi = -20^o$. This resulted in a best fit
267: for which all physical parameters are summarized in Table ~\ref{par_table}.
268:
269: \begin{deluxetable}{lccc}
270: \tabletypesize{\scriptsize}
271: \tablecaption{Relevant parameter choices for our best fit to the broadband spectrum
272: of LSI~+61~303}
273: \tablewidth{0pt}
274: \tablehead{
275: \colhead{Parameter} & \colhead{Symbol} & \colhead{Value}
276: }
277: \startdata
278: Distance & $d$ & $6.17 \times 10^{21}$~cm \\
279: Jet inclination angle & $i$ & $30^o$ \\
280: Bulk Lorentz factor & $\Gamma_j$ & $1.25$ \\
281: Semi Major Axis & $a$ & $5 \times 10^{12}$~cm \\
282: Luminosity of companion star: & $L_{\ast}$ & $2\times 10^{38}$~ergs~s$^{-1}$ \\
283: Surface temperature of the companion star & $T_{\ast}$ & $2.25 \times 10^4$~K \\
284: Initial blob radius & $R_0$ & $10^3 \, R_g$ \\
285: Jet collimation parameter & $\alpha$ & $0.3$ \\
286: Electron injection spectrum, low-energy cutoff & $\gamma_{\rm min}$ & 10 \\
287: Electron injection spectrum, high-energy cutoff & $\gamma_{\rm max}$ & $10^6$ \\
288: Electron injection spectrum, spectral index & $q$ & 1.7 \\
289: Beginning of electron injection zone & $x_0$ & $10^3 \, R_g$ \\
290: End of electron injection zone & $x_1$ & $10^5 \, R_g$ \\
291: Magnetic field at $x_0$ & $B_0$ & $5 \times 10^3$~G \\
292: Injection luminosity & $L_{\rm inj}$ & $10^{35}$~ergs~s$^{-1}$ \\
293: Orbital Period & $P$ & $26.496$~days\\
294: Orbit eccentricity & $e$ & $0.72$ \\
295: \enddata
296: \label{par_table}
297: \end{deluxetable}
298:
299: \subsection{\emph {Results}}
300:
301: \begin{figure}[t]
302: \includegraphics[height=12cm]{f2.eps}
303: \caption{Spectral energy distribution of LSI +61 303. The plotted observational data
304: corresponds to $\psi \sim 0.5$ and are taken from: \cite{albert06} (MAGIC),
305: \cite{harrison00} (RXTE PCA + HEXTE), \cite{strickman98} (OSSE, NIR), \cite{goldoni95}
306: (ROSAT), \cite{leahy03} (EGRET), \cite{chernyakova06} (XMM, INTEGRAL). Also included
307: is the MAGIC $2 \, \sigma$ upper limit for phase $\psi \sim 0.2$ \citep{albert06}.
308: The solid curves show our fit results with $\gamma\gamma$ absorption calculated for
309: $\psi \sim 0.5$ (black) and $\psi \sim 0.16$ (red). A trend of spectral hardening
310: at TeV energies during VHE low phases is predicted.
311: }
312: \label{fit}
313: \end{figure}
314:
315: In Fig. \ref{fit}, we show the computed SEDs of LSI +61 303 using the parameters listed
316: in Table ~\ref{par_table}. The black solid line corresponds to our best fit to the SED
317: at the orbital phase of $\psi = 0.5$ $(\chi = -20^o)$. The figure also shows the individual
318: contributions of the synchrotron, stellar, and SSC emission, as well as the inverse
319: Compton (EC) scattering of starlight photons.
320:
321: The overall agreement of the simulated SED at $\psi = 0.5$ with the observed spectrum
322: is very good, given that the multiwavelength data is not simultaneous. Since the
323: radio emission most likely originates from regions of the jet further away from
324: the compact object than the evolution modeled here, our model underpredicts the
325: VLA data (not included in the figure). The NIR data \citep{strickman98} has been
326: de-reddened using $N_H = 8.4 \times 10^{21} $~cm$^{-2}$ \citep{taylor96}. The infrared
327: emission might be dominated by emission from the outer parts of the circumstellar
328: disk around the Be star which is not included in our model.
329:
330: In order to reproduce the keV -- MeV data, a magnetic field of $5 \times 10^3$~G is
331: required at a distance of $\sim 1.7 \times 10^{9}$~cm from the compact object. In
332: the X-ray band, most of the radiation is synchrotron emission from the jet
333: \citep[also see, e.g.][]{paredes06}, with negligible contributions from SSC
334: or EC. The SSC contribution dominates the spectrum at TeV energies.
335: Note that the individual SSC component in Fig. \ref{fit} does not include
336: the $\gamma\gamma$ absorption feature, which is only applied to the total
337: emanating spectrum.
338:
339: After our fit to the $\psi = 0.5$ phase spectrum, we performed a model simulation with
340: the same intrinsic jet parameters, but changed the orbital geometry to $\psi =
341: 0.16$ $(\chi = 70^o)$, corresponding to the MAGIC low state. Since the EC (star)
342: component was rather insignificant, this will impact essentially only the
343: $\gamma\gamma$ absorption. The corresponding model fit is illustrated by the
344: solid red line in Fig. \ref{fit}. In this case, the $\gamma\gamma$ absorption
345: trough is deeper because (a) VHE $\gamma$-rays pass by the star more closely
346: than their nearest approach in the $\psi = 0.5$ geometry, and (b) starlight
347: photons intercept VHE photons at a more favorable angle for $\gamma\gamma$
348: absorption, leading to a lower energy threshold for this process. In this
349: phase, our model predicts an integrated flux over 400~GeV of $F (E > 400 {\rm GeV})
350: \approx 2.2 \times 10^{-12}$~ph~cm$^{-2}$~s$^{-1}$. This is in perfect agreement
351: with the $2 \sigma$ upper limit of $F (E > 400 {\rm GeV}) < 3 \times
352: 10^{-12}$~ph~cm$^{-2}$~s$^{-1}$ given in \cite{albert06} and indicated by the
353: maroon arrow in Fig. \ref{fit}. Consequently, we find that the orbital
354: modulation of the VHE emission in LS~+61~303 can be explained solely by
355: the effect of the azimuthal asymmetry of $\gamma\gamma$ absorption
356: feature due to starlight photons.
357:
358:
359: \section{\label{summary} Summary and Conclusions}
360:
361: We have applied a leptonic jet model to the broadband spectral energy distribution
362: and orbital modulation of the VHE $\gamma$-ray emission of the microquasar LSI~+61~303,
363: taking into account time dependent electron injection and acceleration, and the
364: adiabatic and radiative cooling of non-thermal electrons. Our model includes synchrotron,
365: synchrotron self-Compton and external inverse Compton emission (with seed photons
366: predominantly from the companion star),
367: as well as $\gamma\gamma$ absorption of $\gamma$-rays
368: by starlight photons. Compton scattering is treated using the full Klein Nishina cross
369: section and the full angular dependence of the stellar radiation field.
370:
371: The model can successfully reproduce the available multiwavelength observational data,
372: including the most recent MAGIC detection in the TeV range \citep{albert06}. Our best
373: fit to the SED indicates that a magnetic field of $B_0 \sim 5 \times 10^3$~G at $\sim
374: 10^3 \, R_g$ is required, and electrons need to be accelerated out to TeV energies
375: ($\gamma_2 = 10^6$) with a nonthermal injection spectrum with a spectral index of
376: $q = 1.7$. Such an injection spectrum can not be achieved by the first-order Fermi
377: mechanism at relativistic or non-relativistic shocks \citep[e.g.,][]{gallant99,achterberg01}
378: and therefore suggests that 2$^{nd}$ order Fermi acceleration \citep{virtanen05} and/or
379: acceleration at shear boundary layers \citep{ostrowski00,stawarz02,rieger04} may play
380: a significant role in the acceleration of relativistic particles in microquasar jets.
381: In our model, the X-ray and $\lesssim 1$~GeV $\gamma$-ray emission is dominated by
382: synchrotron emission from the jet, while SSC emission dominates the VHE $\gamma$-ray
383: emission.
384:
385: The suggested orbital modulation of the VHE $\gamma$-ray emission can be explained solely
386: by the geometrical effect of changes in the relative orientation of the stellar
387: companion with respect to the compact object and jet as it impacts the position
388: and depth of the $\gamma\gamma$ absorption trough \citep{boettcher05}.
389: If this interpretation is correct, a general trend of spectral hardening
390: at TeV energies during VHE $\gamma$-ray low phases is predicted.
391: Although
392: additional effects of a varying accretion rate due to the highly eccentric orbit
393: of the binary system are likely to occur, they are not necessary to explain the
394: observed orbital modulation of the VHE emission of LSI~+61~303.
395:
396: \acknowledgements{
397: We thank the anonymous referee for useful comments and a quick
398: review, and C. D. Dermer, G. Dubus, and V. Bosch-Ramon for helpful comments and
399: discussions.
400: This work was partially supported by NASA through INTEGRAL GO (Theory)
401: grant award no. NNG~05GK59G.}
402:
403:
404: \begin{thebibliography}{}
405:
406: \bibitem[Achterberg et al.(2001)]{achterberg01}Achterberg, A., et al., 2001,
407: MNRAS, 328, 393
408:
409: \bibitem[Aharonian et al.(2005)]{aharonian05}Aharonian F., et al., 2005, Science,
410: 309, 746
411:
412: \bibitem[Albert et al.(2006)]{albert06}Albert, J., et al., 2006, Science, Vol. 312,
413: Issue 5781, p. 1771
414:
415: \bibitem[Atoyan \& Aharonian(1997)]{aa97}Atoyan, A. M., \& Aharonian, F. A.,
416: 1997, ApJ, 490, L149
417:
418: \bibitem[B\"ottcher(2002)]{boettcher02}B\"ottcher, M., 2002, in proc.
419: ``The Gamma-Ray Universe'', XXII Moriond Astrophysics Meeting, eds.
420: A. Goldwurm, D. N. Neuman, \& J. T. T. V$\hat{\rm a}$n, p. 151
421:
422: \bibitem[B\"ottcher \& Bloom(2000)]{boettcher2000}B\"ottcher, M., \& Bloom, S. D.,
423: 2000, AJ, 119,469
424:
425: \bibitem[B\"ottcher \& Dermer(2005)]{boettcher05}B\"ottcher, M., \& Dermer,
426: C. D., 2005, ApJL, 634L, 81B
427:
428: \bibitem[B\"ottcher, Mause \& Schlickeiser(1997)]{bms97}B\"ottcher, M., Mause, H.,
429: \& Schlikeiser, R.,1997, A\&A,324,395
430:
431: \bibitem[Bosch-Ramon \& Paredes(2004a)]{bosch04a}Bosch-Ramon, V., \& Paredes,
432: J. M., 2004a, A\&A, 417, 1075
433:
434: \bibitem[Bosch-Ramon \& Paredes(2004b)]{bosch04b}Bosch-Ramon, V., \& Paredes,
435: J. M., 2004a, A\&A, 425 1069
436:
437: \bibitem[Casares et al.(2005)]{casares05}Casares, J., Ribas, I., Paredes, J. M., Mart\'\i,
438: J., and Allende Prieto, C., 2005, MNRAS, 360, 1105
439:
440: \bibitem[Chernyakova et al.(2006)]{chernyakova06}Chernyakova, M., Neronov,
441: A., \& Walter, R., 2006, MNRAS, submitted (astro-ph/0606070)
442:
443: \bibitem[Chiaberge \& Ghisellini(1999)]{chiaberge99}Chiaberge, M., \& Ghisellini,
444: G., 1999, MNRAS, 306, 551
445:
446: \bibitem[Corbel et al.(2002)]{corbel02}Corbel, S., Fender, R. P., Tzioumis, T.,
447: Tomsick, J. A., Orosz, J. A., Miller, j. M., Wijnands, R., \& Kaaret, P., 2002,
448: Science, 298, 196
449:
450: \bibitem[Dermer \& B\"ottcher(2006)]{dermer06}Dermer, C. D., \& B\"ottcher, M.,
451: 2006, ApJ, 643, 1081
452:
453: \bibitem[Dubus(2006)]{dubus06}Dubus, G., 2006, A\&A, 451, 9
454:
455: \bibitem[Rieger \& Duffy(2004)]{rieger04}Rieger, F. M., \& Duffy, P., 2004,
456: ApJ, 617, 155
457:
458: \bibitem[Frail \& Hjellming(1991)]{frail91}Frail, D. A., \& Hjellming, R. M.,1991,
459: AJ,101,2126
460:
461: \bibitem[Gallant et al.(1999)]{gallant99}Gallant, Y., et al., 1999, A\&AS, 138, 549
462:
463: \bibitem[Goldoni \& Mereghetti(1995)]{goldoni95}Goldoni, P., \& Mereghetti, S.,
464: 1995, A\&A, 299, 751
465:
466: \bibitem[Gregory(2002)]{gregory02}Gregory, P. C.,2002, ApJ, 575, 427
467:
468: \bibitem[Gregory \& Taylor(1978)]{gt78}Gregory, P. C., \& Taylor, A. R., 1978,
469: Nature, 272, 704
470:
471: \bibitem[Gupta et al.(2006)]{gupta06}Gupta, S., B\"ottcher, M., \& Dermer, C. D.,
472: 2006, ApJ, 644, 409
473:
474: \bibitem[Harrison et al.(2000)]{harrison00}Harrison, F. A., Ray, P. S.,
475: Leahy, D. A., Waltman, E. B., \& Pooley, G. G., 2000, ApJ, 528, 454
476:
477: \bibitem[Hartman et al.(1999)]{hartman99}Hartman, R. C., 1999, ApJS,
478: 123, 79
479:
480: \bibitem[Hermsen et al.(1977)]{hermsen77}Hermsen, W., et al., 1977, Nature,269,
481: 494
482:
483: \bibitem[Hutchings \& Crampton(1981)]{hutchings81}Hutchings, J. B., \& Crampton, D.,
484: 1981,PASP,93,486
485:
486: \bibitem[Kniffen et al.(1997)]{kniffen97}Kniffen, D. A., et al., 1997, ApJ, 486,
487: 126
488:
489: \bibitem[Leahy(2003)]{leahy03}Leahy, D., 2003, in proc. of 28$^{th}$ ICRC,
490: Tsukuba, Japan, 2003; Univ. Academy Press, Inc.; p. 2461
491:
492: \bibitem[Massi et al.(2004)]{massi04}Massi, M., Rib\`{o}, M., \& Paredes, J. M.,
493: 2004,A\&A,414,L1
494:
495: \bibitem[Ostrowski(2000)]{ostrowski00}Ostrowski, M., 2000, MNRAS, 312, 579
496:
497: \bibitem[Paredes et al.(1996)]{paredes96}Paredes, J. M., et al,1996,ASPC,93,243P
498:
499: \bibitem[Paredes et al.(2000)]{paredes2000}Paredes, J. M., Mart\`{i}, J., Rib\`{o},
500: M., \& Massi, M., 2000, Science, 288, 2340
501:
502: \bibitem[Paredes et al.(2006)]{paredes06}Paredes, J. M., Bosch-Ramon, V., \& Romero, G. E.,
503: 2006,A\&A,451,259
504:
505: \bibitem[Press et al.(1992)]{press92}Press, W. H., et al., 1992, Numerical Recipes
506: in C (Cambridge: Cambridge Univ. Press)
507:
508: \bibitem[Romero et al.(2003)]{romero03}Romero, G. E., Torres, D. F., Kaufman Bernard\'o,
509: M. M., Mirabel, I. F., A\&A, 410, L1
510:
511: \bibitem[Romero et al.(2005)]{romero05}Romero, G. E., Christiansen, H. R., \& Orellana,
512: M.,2005,ApJ,632,1093
513:
514: \bibitem[Stawarz \& Ostrowski(2002)]{stawarz02}Stawarz, L., \& Ostrowski, M., 2002,
515: ApJ, 578, 763
516:
517: \bibitem[Strickman et al.(1998)]{strickman98}Strickman, M. S., et al.,1998,ApJ,497,419
518:
519: \bibitem[Taylor et al.(1996)]{taylor96}Taylor, A. R., Young, G., Peracaula, M.,
520: Kenny, H. T., \& Gregory, P. C., 1996, A\&A, 305, 817
521:
522: \bibitem[Tomsick(2002)]{tomsick02}Tomsick, J. A., Corbel, S., Fender, R. P., Miller,
523: J. M., Orosz, J. A., Tzioumis, T., Wijnands, R., \& Kaaret, P., 2002, ApJ, 582, 933
524:
525: \bibitem[Virtanen \& Vainio(2005)]{virtanen05}Virtanen, J. J. P., \& Vainio, R., 2005,
526: ApJ, 621, 313
527:
528: \end{thebibliography}
529:
530: \end{document}
531: