1: \documentclass[useAMS,usenatbib,a4paper]{mn2e}
2: \usepackage{graphicx}
3: \usepackage{amsmath}
4: \usepackage{txfonts}
5:
6: \def\nodata{\dots}
7: \def\aap{A\&A}
8: \def\apjl{ApJ}
9: \def\aph{APh}
10: \def\apjs{ApJS}
11: \def\apj{ApJ}
12: \def\araa{ARA\&A}
13: \def\nat{Nat}
14: \def\sci{Sci}
15: \def\baas{BAAS}
16: \def\mnras{MNRAS}
17: \def\prl{PRL}
18:
19: \title[Synoptic VHE blazar studies]{Synoptic studies of seventeen blazars detected in very high-energy gamma-rays}
20: \author[R. M. Wagner]{R. M. Wagner\thanks{E-mail: robert.wagner@mppmu.mpg.de}
21: \\
22: Max-Planck-Institut f\"ur Physik, F\"ohringer Ring 6, D-80805
23: M\"unchen, Germany}
24:
25: \date{Accepted 2007 December 12. Received 2007 December 10; in original form 2007 September 13}
26:
27: \pubyear{2008}
28:
29: \begin{document}
30: \maketitle
31: \label{firstpage}
32:
33: \begin{abstract}
34: Since 2002, the number of detected blazars at gamma-ray energies
35: above 100 GeV has more than doubled. I study 17 blazars currently
36: known to emit $E>100$~GeV gamma rays. Their intrinsic energy
37: spectra are reconstructed by removing extragalactic background
38: light attenuation effects. Luminosity and spectral slope in the
39: $E>100$~GeV region are then compared and correlated among each
40: other, with X-ray, optical and radio data, and with the estimated
41: black hole (BH) masses of the respective host galaxies.
42:
43: According to expectations from synchrotron self-Compton emission
44: models, a correlation on the 3.6-$\sigma$ significance level between gamma-ray
45: and X-ray fluxes is found, while correlations between gamma-ray
46: and optical/radio fluxes are less pronounced. Further, a general
47: hardening of the blazar spectra in the $E>100$~GeV region with
48: increasing gamma-ray luminosity is observed, both for the full
49: 17-source sample and for those sources which have been detected at
50: distinct flux levels. This goes in line with a correlation of the
51: gamma-ray luminosity and the synchrotron peak frequency, which is
52: also seen. Tests for possible selection effects reveal a hardening
53: of the spectra with increasing redshift. The blazar gamma-ray
54: emission might depend on the mass of the central BH. The blazars
55: under study do, however, show no correlation of the BH masses with
56: the spectral index and the luminosity in the $E>100$~GeV region.
57:
58: I also consider temporal properties of the X-ray and $E>100$~GeV
59: gamma-ray flux. No general trends are found, except for the
60: observation that the blazars with the most massive BHs do not show
61: particularly high duty cycles. These blazars include Mkn\,501 and
62: PKS\,2155-304, for which recently very fast flares have been
63: reported. In general, VHE flare time-scales are not found to scale
64: with the BH mass.
65:
66: As a specific application of the luminosity study, a constraint
67: for the still undetermined redshift of the blazar PG\,1553+113 is
68: discussed.
69: \end{abstract}
70:
71: \begin{keywords}
72: galaxies: active -- BL Lacertae objects: individual
73: (1ES\,0229+200, 1ES\,0347-121, 1ES\,1011+496, 1ES\,1101-232,
74: 1ES\,1218+304, 1ES\,1959+650, 1ES\,2344+514, 3C\,279, BL~Lacertae,
75: H\,1426+428, H\,2356-309, Mkn\,180, Mkn\,421, Mkn\,501,
76: PG\,1553+113, PKS\,0548-322, PKS\,2005-489, PKS\,2155-304) -- black hole physics -- galaxies: jets.
77: \end{keywords}
78:
79: \section{Introduction}
80: %
81: All but one of the detected extragalactic very high energy (VHE,
82: defined by $E>100\,\mathrm{GeV}$) gamma ($\gamma$) ray sources so
83: far are blazars. Within the unified scheme
84: \citep[e.g.][]{UrryPadovani} of active galactic nuclei (AGN),
85: blazars comprise the rare and extreme subclasses of BL~Lac objects
86: and flat spectrum radio quasars (FSRQs). These are characterised
87: by high apparent luminosities, short variability time-scales, and
88: apparent superluminal motion of jet components. These observations
89: can be explained by highly relativistic, beamed plasma outflows
90: (jets) closely aligned to the observer's line of sight
91: \citep{1979ApJ...232...34B} powered by central supermassive black
92: holes accreting at sub-Eddington rates
93: \citep{1969Natur.223..690L,1978Natur.275..516R}. The prime
94: scientific interest in VHE $\gamma$-ray emitting blazars (in the
95: following, `VHE blazars') is twofold: (1) To understand the
96: particle acceleration and $\gamma$-ray production mechanisms,
97: assumed to take place in the jets and to be linked to the central
98: supermassive black hole (BH). Knowledge of the VHE
99: emission process will also contribute to the further understanding
100: of the accretion processes in AGN, jet formation processes, and
101: the jet structure. (2) To use the VHE $\gamma$-rays as a probe of
102: the extragalactic background light \citep[EBL;
103: e.g.][]{hauser,2005PhR...409..361K} spectrum in the wavelength
104: range between about $0.3$ to $30~\mu\rm m$. Determining the EBL
105: spectrum in this wavelength range may allow to constrain the star
106: formation rate (convolved with the initial mass function) in the
107: early Universe. In order to assess both issues, it is essential to
108: have a large sample of VHE $\gamma$-ray blazars at hand. Ideally
109: it should encompass a wide range in redshift for EBL studies and
110: at the same time include groups of sources at similar distances in
111: order to probe and compare properties of the individual sources
112: without possible systematic uncertainties caused by the EBL
113: de-absorption.
114: %
115: \begin{figure*}
116: \center{\includegraphics[width=0.65\linewidth]{fig01}
117: \caption{Currently known VHE $\gamma$-ray blazars along with the
118: identified (66 objects) and tentatively identified AGN (27
119: objects) in the 3rd EGRET catalogue of $\gamma$-ray sources (solid
120: grey dots: identified AGN; open grey dots: tentatively identified
121: AGN). EGRET data from \citet{hartman}.
122: The sources are shown in a galactic coordinate
123: system.\label{fig:comp:sp}}}
124: \end{figure*}
125:
126: The preconditions for such studies have much improved recently:
127: Before 2004, only a few nearby extragalactic sources had been
128: established as VHE $\gamma$-ray emitters \citep[e.g.][]{Moriicrc}
129: and provided hardly enough data to perform comparative studies.
130: Around 2004, the third generation of imaging air Cerenkov
131: telescopes (IACTs, e.g. MAGIC, \citealt{Baixeras,Cortina} and
132: H.E.S.S., \citealt{HESStech}), the most successful tools so far to
133: explore VHE $\gamma$-rays, started to deliver scientific results.
134: To date, the VHE blazar sample with available spectral information in the
135: VHE region comprises 17 BL Lac objects, %\footnote{See e.g. http://www.mppmu.mpg.de/$\sim$rwagner/sources/ for an up-to-date list}
136: among them one LBL object, BL~Lacertae. Furthermore, now the
137: redshifts of the known VHE blazars reach up to $z=0.212$ -- or even
138: to $z=0.536$, considering the recently announced discovery of the
139: first FSRQ in VHE $\gamma$-rays, 3C~279 \citep{279}. For
140: 3C~279, however, no VHE spectrum has been published yet, therefore
141: the VHE luminosity and spectral slope of 3C~279 are not yet
142: available for this study. M87, a FR~I radio galaxy also detected
143: in VHE $\gamma$-rays, is not included in the study, as its VHE
144: $\gamma$-ray production mechanism may differ from that in blazars
145: \citep{hess87n}. Fig.~\ref{fig:comp:sp} shows the sky positions of
146: the known VHE blazars in galactic coordinates along with the AGN
147: identified in the 3rd EGRET catalogue.
148:
149:
150: The electromagnetic continuum spectra of blazars extend over many
151: orders of magnitude from radio frequencies to sometimes multi-TeV
152: energies and are dominated by non-thermal emission that consists
153: in a $\nu F_\nu$ representation of two pronounced peaks. The
154: low-energy peak, located between the IR and hard X-rays, is
155: thought to arise from synchrotron emission of ultrarelativistic
156: electrons, accelerated by shocks moving along the jets at
157: relativistic bulk speed. Depending on the location of the
158: low-energy peak, BL~Lac objects are often referred to as
159: high-frequency peaked (HBL; in the UV to X-ray domain) or
160: low-frequency peaked (LBL; in the near-IR to optical) BL~Lac
161: objects \citep{fossati}, although the transition is smooth rather
162: than dichotomic. The origin of the high-energy peak at MeV to TeV
163: energies is still debated. It is commonly explained by inverse
164: Compton upscattering of low-energy photons by electrons. The seed
165: photons may originate from synchrotron radiation produced by the
166: same electron population \citep*[synchrotron-self Compton (SSC)
167: models; e.g.][]{maraschi,coppi2} or belong to ambient thermal
168: photon fields \citep*[external inverse Compton models;
169: e.g.][]{mk89,1994ApJ...421..153S,1994ApJS...90..945D}. In hadronic
170: models, which can also explain the observed features, interactions
171: of a highly relativistic jet outflow with ambient matter
172: \citep{1997ApJ...478L...5D,1993ApJ...402L..29B}, proton-induced
173: cascades \citep{MannheimProton}, synchrotron radiation by protons
174: \citep{AharonianHad,MueckeProtheroe}, or curvature radiation, are
175: responsible for the high energy photons. The AGN identified in
176: the EGRET data are predominantly powerful FSRQs and quasars with
177: SEDs peaking at rather low frequencies, and thus only few of these
178: (Mkn 421, PKS 2155-304, BL~Lacertae and 3C~279) were also
179: detected in the VHE range.
180:
181: Knowing the variability time-scales of the VHE $\gamma$-ray
182: emission and the form of the two-bump spectral energy distribution
183: (SED) enables the derivation of all input parameters of one-zone
184: SSC models \citep{Tavecchio1998}, which describe the observed
185: emission in BL~Lac objects reasonably well. Strictly simultaneous
186: and temporally-resolved measurements of the SED, however, are only
187: rarely possible and more often than not are also severely
188: restricted by the (temporal) instrumental resolution. Generally,
189: detailed spectral studies particularly in low-emission states are
190: rather demanding. In addition, the determination of the location
191: of the high-energy peak $\nu^\mathrm{IC}_\mathrm{peak}$ generally
192: would require complementary satellite detector coverage of the SED
193: between some hundred MeV and $\approx 50$~GeV. This region of the
194: SED, however, is difficult to access due to the low fluxes
195: expected (up to $\approx 100$ MeV) from extragalactic sources in
196: between the two bumps and due to insufficient instrumental
197: sensitivity for energies exceeding some GeVs.
198:
199: In this paper, for the first time studies of the VHE emission
200: properties of the complete set of all currently known VHE blazars
201: (cf. Tab.~\ref{tab:TeVSources}) are performed.
202: %
203: First, the detected VHE blazars are brought into context with the
204: AGN searches conducted by IACTs so far and the expected
205: $\gamma$-ray attenuation by the EBL in Sect.~\ref{sect:grhorsect}.
206: After a study of the black hole mass distribution of the VHE
207: blazars in Sect.~\ref{sect:comp:bhmavg}, I infer intrinsic
208: emission properties in the VHE $\gamma$-ray regime in
209: Sect.~\ref{sect:intrvhe}. Because measurements of the location and
210: shape of the high-energy bump are elusive at present for almost
211: all VHE blazars, the observed $\gamma$-ray luminosity and spectral
212: slope in the VHE region are used as auxiliary observables to
213: characterise the VHE $\gamma$-ray emission. The main part of the
214: paper (Sect.~\ref{sect:corstudies1}--\ref{sect:intrinsicspectra}) is devoted to the search for
215: correlations of these observables with the X-ray emission
216: properties, the optical and the radio luminosity, and with the
217: black hole mass estimations. In Sect.~\ref{sect:comp:501lumilim}
218: the VHE blazar luminosity distribution is used to address the
219: specific problem of the unknown redshift of PG\,1553+113 by
220: deriving an upper redshift for this VHE blazar. Finally,
221: Sect.~\ref{sect:dcsect} turns to the study of X-ray and VHE
222: $\gamma$-ray timing properties. Sect.~\ref{sect:conc} summarises
223: the main conclusions of the studies.
224:
225: \begin{table}
226: \caption{Extragalactic VHE $\gamma$-ray sources, listed in
227: chronological order of their discovery.} \label{tab:TeVSources}
228: \begin{tabular}{lccl}
229: \hline
230: Source & Type & Redshift $z$ & Discovery reference \\
231: \hline
232: Mkn\,421 & HBL & 0.030\phantom{0} & \citet{Punch1992} \\
233: Mkn\,501 & HBL & 0.034\phantom{0} & \citet{Quinn1996} \\
234: 1ES\,2344+514 & HBL & 0.044\phantom{0} & \citet{Catanese1998} \\
235: 1ES\,1959+650 & HBL & 0.047\phantom{0} & \citet{Nishiyama1999} \\
236: PKS\,2155-304 & HBL & 0.116\phantom{0} & \citet{Chadwick} \\
237: H\,1426+428 & HBL & 0.129\phantom{0} & \citet{Horan} \\
238: M87$^a$ & FR~I & 0.0044 & \citet{Tluczykont}\\
239: \hline
240: PKS\,2005-489 & HBL & 0.071\phantom{0}& \citet{HESS2005} \\
241: 1ES\,1218+304 & HBL & 0.182\phantom{0} & \citet{MAGIC1218} \\
242: H\,2356-309 & HBL & 0.165\phantom{0} & \citet{HESSAGNNature} \\
243: 1ES\,1101-232 & HBL & 0.186\phantom{0} & \citet{HESSAGNNature} \\
244: PG\,1553+113 & HBL & $^b$ & \citet{HESS1553}, \\
245: & & & \citet{MAGIC1553published} \\
246: Mkn\,180 & HBL & 0.045\phantom{0} & \citet{MAGIC180} \\
247: PKS\,0548-322 & HBL & 0.069\phantom{0} & \citet{0548icrc} \\
248: BL~Lacertae & LBL & 0.069\phantom{0} & \citet{MAGICBLLac} \\
249: 1ES\,1011+496 & HBL & 0.212\phantom{0} & \citet{1011} \\
250: 1ES\,0229+200 & HBL & 0.139\phantom{0} & \citet{0229} \\
251: 1ES\,0347-121 & HBL & 0.188\phantom{0} & \citet{0347} \\
252: 3C~279$^c$ & FSRQ& 0.536\phantom{0} & \citet{279} \\
253: \hline
254: \end{tabular}
255:
256: \medskip
257: The upper part of the table shows the confirmed sources prior to
258: the advent of new generation instruments like MAGIC and H.E.S.S.,
259: while the lower panel summarises the sources discovered after
260: 2002. HBL: High-frequency peaked BL~Lac object, LBL: Low-frequency
261: peaked BL~Lac object, FSRQ: Flat spectrum radio quasar, FR:
262: Fanaroff--Riley galaxy. $^a$ M87 is not included in the present
263: study. $^b$ This redshift is currently under discussion, cf. Sect.
264: \ref{sect:comp:501lumilim}. $^c$ No spectrum of 3C~279 has been
265: published yet.
266: \end{table}
267:
268: \section{Population studies and the $\gamma$-ray horizon}
269: \label{sect:grhorsect}
270:
271: \begin{figure*}
272: \center{
273: \includegraphics[width=0.495\linewidth]{fig02a}
274: \includegraphics[width=0.495\linewidth]{fig02b}
275: \caption{{(a)} Lower energy thresholds from searches for VHE
276: $\gamma$-ray emission from AGN. Dots: blazars, crosses: other AGN
277: types (starburst galaxies, radio galaxies, etc.). The curves
278: represent Fazio--Stecker relations (flux attenuation by a factor
279: $\mathrm{e}^{-1}$) for different EBL models given by
280: \citet{kneiske4}. Data from searches by HEGRA \citep{Tluczykont},
281: H.E.S.S. \citep{2005A&A...441..465A,HESS-UL-ICRC}, Whipple
282: \citep{1995ApJ...452..588K,Horan04}, VERITAS
283: \citep{VERITAS-UL-ICRC,veul}, CANGAROO
284: \citep{2002PASA...19...26N} and MAGIC \citep{MAGIC-UL-Paper}.
285: {(b)} Detected VHE $\gamma$-ray sources (see
286: Tab.~\ref{tab:parvhe} for references). Shown are the energy ranges
287: of the measured $\gamma$-ray spectra; high-energy cutoffs were
288: found only for Mkn\,421 and Mkn\,501, for all other sources arrows
289: represent spectra possibly continuing to higher energies. The
290: dashed lines represent PG\,1553+113 at the lower and upper limit
291: for its redshift, $z>0.09$ \citep{Sbar} and $z<0.74$
292: \citep{HESS1553,MAGIC1553published}.} \label{fig:comp:compa}}
293: \end{figure*}
294:
295: When travelling cosmological distances, VHE $\gamma$-rays interact
296: with the low-energy photons of the EBL \citep[see,
297: e.g.][]{nikishov,gould,hauser,2005PhR...409..361K}. The
298: predominant reaction $ \gamma_{\mbox{\scriptsize{\,VHE}}} +
299: \gamma_{\mbox{\scriptsize EBL}} \rightarrow
300: \mathrm{e}^{+}\,\mathrm{e}^{-}$ modifies source-intrinsic
301: $\gamma$-ray energy spectra. The cross-section of this process
302: peaks strongly at $E_\mathrm{CM}=1.8\times 2m_\mathrm{e} c^2$,
303: therefore a given VHE photon energy probes a narrow range of the
304: EBL spectrum. The part of the EBL to which VHE $\gamma$-rays are
305: sensitive comprises the (redshifted) relic emission of galaxies
306: and star-forming systems and the light absorbed and re-emitted by
307: dust. The EBL attenuation results in a maximum distance over which
308: photons with a particular energy can survive: The Fazio-Stecker
309: relation \citep*[FSR,][]{fs70,sjs92} describes the distance at
310: which the optical depth for a VHE photon of a given energy reaches
311: unity (attenuation by a factor $\mathrm{e}^{-1}$).
312: Thus the FSR defines the {\it cosmological $\gamma$-ray horizon}.
313: Fig.~\ref{fig:comp:compa}a shows the instrumental low-energy
314: thresholds from searches for VHE $\gamma$-ray emission from AGN.
315: Along with these, the FSR for different EBL models as given by
316: \citet{kneiske4} is plotted. The models differ in the IR density,
317: dust properties and the star formation rate in the early Universe.
318: The FSR divides the plot into a region from which no $\gamma$-rays
319: can reach the Earth and into another region, in which positive
320: detections are to be expected or a too weak source-intrinsic
321: emission made detections fail (due to insufficient instrumental
322: sensitivity). Obviously, with a decreasing instrumental energy
323: threshold, the visible Universe `opens up', providing access to a
324: larger source population. Current EBL models result in a
325: steepening of the intrinsic spectra from $\approx
326: 200\,\mathrm{GeV}$ on (power-law spectra are softened, but their
327: shape is approximately retained), while for lower energies the
328: effects are minimal. When an optical depth of one is reached, a
329: quasi-exponential cutoff in the observed spectra occurs.
330: Fig.~\ref{fig:comp:compa}b shows the energy ranges over which
331: blazars in VHE $\gamma$ radiation have been detected. Up to now,
332: only for the strong, close-by blazars Mkn~421 and Mkn~501
333: indications of the expected exponential high energy cutoff have
334: been observed thanks to high $\gamma$ statistics
335: \citep{2001A&A...366...62A,2001ApJ...560L..45K,MAGIC421published}.
336: The observed spectra of all other blazars can be accurately
337: described by power-laws or broken power-laws. While most of the
338: nearby VHE blazars cannot constrain the current EBL models, some
339: of the sources at $z>0.1$ start challenging these
340: \citep{HESSAGNNature,0347,279}, as no cutoffs have been observed
341: so far at the high-energy ends of their $\gamma$-ray spectra.
342:
343: \section{Black hole masses in blazars}
344: \label{sect:comp:bhmavg}
345: %
346: It is well established that all galaxies with a massive bulge
347: component host supermassive black holes in their centres
348: \citep{Richstone-98,Bender-03}. There are a couple of indirect
349: methods to infer the masses of the central BHs: One is to estimate
350: $M_\bullet$ using the correlation between $M_\bullet$ and the
351: central velocity dispersion $\sigma$ of the host galaxy
352: \citep[$M_\bullet-\sigma$ relation,][]{2000ApJ...539L...9F,Gebhardt} found
353: from stellar and gas kinematics and maser emission. I estimated
354: the black hole masses of VHE $\gamma$-ray emitting blazars using
355: the $M_\bullet-\sigma$ relation given by \citet{2002ApJ...574..740T}. This
356: approach assumes that AGN host galaxies are similar to non-active
357: galaxies. The velocity dispersions were collected from the
358: literature or are inferred from the fundamental plane
359: \citep{1987ApJ...313...59D}, a relation between $\sigma$, the
360: effective galaxy radius $R_\mathrm{e}$, and the corresponding
361: surface brightness $\langle\mu_\mathrm{e}\rangle$, which is valid
362: for elliptical galaxies, in particular also for AGN and radio
363: galaxies \citep{Bettoni}. Whenever more than one $\sigma$ value is
364: given in the literature, individual masses were derived for each
365: of the $\sigma$ values and averaged. The $\sigma$ values and the
366: resulting black hole masses for the VHE $\gamma$-ray blazars
367: studied here are given in Tab.~\ref{tab:bhmasses}.
368: %
369:
370: \begin{table*}
371: \begin{minipage}{147mm}
372: \caption{Measured velocity dispersions and resulting estimated
373: black hole masses for the VHE blazars.} \label{tab:bhmasses}
374: \begin{tabular}{lccccccc}
375: \hline Object & $\sigma\,[\mathrm{km}\,\mathrm{s}^{-1}]$ &
376: $\log(M_\bullet/\mathrm{M}_{\sun})$ &
377: $\sigma\,[\mathrm{km}\,\mathrm{s}^{-1}]$ &
378: $\log(M_\bullet/\mathrm{M}_{\sun})$ &
379: $\sigma\,[\mathrm{km}\,\mathrm{s}^{-1}]$ &
380: $\log(M_\bullet/\mathrm{M}_{\sun})$ &
381: $\log(M_\bullet/\mathrm{M}_{\sun})$ \\
382: &Reference 1 & & Reference 2 & & Reference 3 & & averaged \\
383: \hline
384: Mkn~421 & $219\pm11$ & $8.29\pm0.18$ & $236\pm10$ & $8.42\pm0.15$ & $324\pm18$ & $8.97\pm0.12$ & 8.56 \\
385: Mkn~501 & $372\pm18$ & $9.21\pm0.11$ & $291\pm13$ & $8.78\pm0.11$ & \nodata & \nodata & 9.00 \\
386: 1ES\,2344+514 & $294\pm24$ & $8.80\pm0.15$ & \nodata & \nodata & $389\pm20$ & $9.29\pm0.12$ & 9.04 \\
387: Mkn~180 & $209\pm11$ & $8.20\pm0.19$ & $244\pm10$ & $8.47\pm0.14$ & $251\pm16$ & $8.52\pm0.15$ & 8.40 \\
388: 1ES\,1959+650 & \nodata & \nodata & $195\pm15$ & $8.08\pm0.23$ & $219\pm15$ & $8.28\pm0.19$ & 8.18 \\
389: BL~Lacertae & \nodata & \nodata & \nodata & \nodata & $245\pm16$ & $8.48\pm0.15$ & 8.48 \\
390: PKS~0548-322 & $202\pm24$ & $8.14\pm0.24$ & \nodata & \nodata & \nodata & \nodata & 8.14 \\
391: PKS~2005-489 & \nodata & \nodata & \nodata & \nodata & $257\pm16$ & $8.57\pm0.14$ & \nodata \\
392: H\,1426+428 & \nodata & \nodata & \nodata & \nodata & $269\pm16$ & $8.65\pm0.13$ & \nodata \\
393: H\,2356-309 & \nodata & \nodata & \nodata & \nodata & $195\pm14$ & $8.08\pm0.23$ & \nodata \\
394: 1ES\,0229+200 & \nodata & \nodata & \nodata & \nodata & $363\pm19$ & $9.16\pm0.11$ & \nodata \\
395: 1ES\,1218+304 & \nodata & \nodata & \nodata & \nodata & $191\pm14$ & $8.04\pm0.24$ & \nodata \\
396: 1ES\,0347-121 & \nodata & \nodata & \nodata & \nodata & $214\pm15$ & $8.24\pm0.19$ & \nodata \\
397: 1ES\,1011+496 & \nodata & \nodata & \nodata & \nodata & $219\pm15$ & $8.28\pm0.19$ & \nodata \\
398: \hline
399: \end{tabular}
400:
401: \medskip
402: References: (1) \citet*{Barth2003}; (2)
403: \citet*{2002ApJ...569L..35F}; (3) \citet*{2002AA...389..742W}. The
404: velocity dispersions $\sigma$ were translated into estimated BH
405: masses using the $M_\bullet-\sigma$ relation from
406: \citet{2002ApJ...574..740T}. The $\sigma$ values taken from
407: \citet{2002AA...389..742W} were indirectly determined using the
408: fundamental plane of radio galaxies \citep{Bettoni}. BH masses
409: given in units of the solar mass, $\mathrm{M}_{\sun}$. If more
410: than one $M_\bullet$ value is given, the average $M_\bullet$ is
411: used. Due to the possible different systematic errors of the
412: individual data sets, the largest error was assumed as error of
413: the average $M_\bullet$.
414: %
415: \end{minipage}
416: \end{table*}
417: %
418: The determination of $M_\bullet$ suffers from rather large
419: systematic uncertainties due to the different methods used to
420: derive $\sigma$. The relation between $M_\bullet$ and bulge
421: luminosity $L_\mathrm{B}$ \citep{1995ARAA..33..581K} has generally
422: a larger scatter than the $M_\bullet-\sigma$ relation and was therefore
423: only used for PKS\,2155-304, because for this blazar no $\sigma$
424: or $\langle\mu_\mathrm{e}\rangle$ measurement is available. I used
425: the $R$-band luminosity given by \citet{1996MNRAS.283..241F} to
426: calculate $\log(M_\bullet/\mathrm{M}_{\sun})=8.98\pm0.44$ using
427: eq.~12 in \citet{2007MNRAS.379..711G}. \citet{2155flare} give an
428: estimate of $M_\bullet=(1\dots2)\times10^9 \mathrm{M}_{\sun}$. For
429: a comparative study using values inferred by different methods is
430: not advisable due to possible different systematics. Therefore,
431: the $M_\bullet$ value used for PKS\,2155-304 should be taken with
432: care in the following. The BH mass of 3C~279 was determined using
433: the virial BH mass estimate of \citet{McLure} and is given as
434: $\log(M_\bullet/\mathrm{M}_{\sun}) = 8.912$ by
435: \citet{2001MNRAS.327.1111G}.
436: For two of the blazars under study, PG\,1553+113 and
437: 1ES\,1101-232, no $M_\bullet$ estimations exist yet.
438:
439: Recent estimations of the BH masses for 452 AGN find them
440: distributed over a large range of
441: $(10^6-7\times10^{9})\,\mathrm{M}_{\sun}$ with no evidence for
442: dependencies on the radio loudness of the objects
443: \citep{2002ApJ...579..530W,2002ApJ...581L...5W}. A recent study of
444: the BH mass distribution of 66 BL~Lac objects
445: \citep{2005ApJ...631..762W} reports an $M_\bullet$ range of
446: $(10^7-4\times10^{9})\,\mathrm{M}_{\sun}$ and could also not find
447: a correlation of $M_\bullet$ with radio or X-ray luminosity. (In
448: Sect. \ref{sect:intrinsicspectra} the distribution of VHE blazars
449: in luminosity and $M_\bullet$ is discussed). As
450: Fig.~\ref{fig:comp:compaz} shows, there is no dependence of the BH
451: masses of the VHE blazars on their redshift, but they are rather
452: flatly distributed in their BH masses between $(10^8-10^{9.5})
453: \mathrm{M}_{\sun}$.
454: %
455: \begin{figure*}
456: \center{\includegraphics[width=0.65\linewidth]{fig03}
457: \caption{Redshift vs. $M_\bullet$ distribution for the known VHE
458: $\gamma$-ray emitting AGN. The superimposed histogram shows the
459: BH mass distribution of the 375 AGN given in
460: Fig.~\ref{fig:comp:compazwu} (linear vertical scale independent of
461: $z$). \label{fig:comp:compaz}}}
462: \end{figure*}
463: %
464: Although AGN harbour BHs with $M_\bullet > 10^6
465: \mathrm{M}_{\sun}$, up to now only blazars with rather massive
466: BHs, $M_\bullet \ga 10^8 \mathrm{M}_{\sun}$, have been discovered
467: in VHE $\gamma$-rays, raising the question whether a physics
468: reason is responsible for the non-detection of blazars with less
469: massive BHs in the mass range $(10^7-10^8)\,\mathrm{M}_{\sun}$.
470: There exist studies that find radio-loud AGN, and therefore also
471: blazars, to be associated with BHs with $M_\bullet \ga 10^9
472: \mathrm{M}_{\sun}$ \citep{laor2000} or at least on average with
473: more massive BHs than radio-quiet AGN \citep{mm06}. The latter
474: authors also report a threshold BH mass for the onset of radio
475: activity, with very little dependence of the radio output on the
476: BH mass once above the threshold mass. Whether such a mass
477: threshold is also at work for the VHE emission, remains subject
478: for further studies at this point.
479: %
480: The BH masses of the VHE blazars are compared to those of 375 AGN
481: collected by \citet{2002ApJ...579..530W} in
482: Fig.~\ref{fig:comp:compazwu}. The confinement of Seyfert galaxy
483: measurements to low redshifts presumably is due to a selection
484: effect: These are spiral galaxies and therefore expected to
485: harbour comparatively low-mass BHs. Distant Seyfert galaxies
486: ($z\ga1.0$) might just not be luminous enough to obtain
487: $M_\bullet$ measurements. Conversely, quasars are too rare as to
488: be found in small volumes and thus at small distances.
489: %
490: \begin{figure*}
491: \center{\includegraphics[width=0.65\linewidth]{fig04}
492: \caption{The redshift vs. $M_\bullet$ distribution for 375 AGN
493: collected by \citet{2002ApJ...579..530W} and the known VHE
494: $\gamma$-ray emitting AGN.\label{fig:comp:compazwu}}}
495: \end{figure*}
496:
497: \section{Intrinsic VHE $\gamma$-ray emission parameters}
498: \label{sect:intrvhe}
499: %
500: VHE $\gamma$-ray observations enable us to look deep into the
501: emission regions of blazar jets and thus convey information on the
502: responsible particle acceleration and cooling processes. Here I
503: study primarily the differential energy spectra in the VHE domain,
504: which are summarised in Tab.~\ref{tab:parvhe}.
505: %
506: \begin{table*}
507: \begin{minipage}{175mm}
508: \caption{Measured VHE blazar spectra, reconstructed intrinsic
509: spectral indices and luminosities.} \label{tab:parvhe}
510: \begin{tabular}{lllcccc}
511: \hline Object & Measured Energy Spectrum $\mathrm{d}F/\mathrm{d}E$
512: & Reference &
513: Intrinsic & $\nu_\gamma L_\gamma$ \\
514: & $[\mathrm{TeV}^{-1}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}]$
515: & & Slope $\Gamma$ &
516: $[\mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{sr}^{-1}]$\\
517: \hline
518: %
519: Mkn~421 & $(12.1\pm0.5)10^{-12} (E/1.0\,\mathrm{TeV})^{-3.09\pm0.07} $ & \citet{AharonianMkn4211999} & $2.85\pm0.58$ & $(9.68\pm0.40)\times10^{43}$ \\
520: Mkn~501 & $(8.4\pm0.5)10^{-12} (E/1.0\,\mathrm{TeV})^{-2.76\pm0.08} $ & \citet{Aharonian2001} & $2.49\pm0.84$ & $(6.89\pm0.41)\times10^{43}$ \\
521: 1ES\,2344+514 & $(1.2\pm0.2)10^{-11} (E/0.5\,\mathrm{TeV})^{-2.95\pm0.12} $ & \citet{MAGIC2344published} & $2.67\pm0.21$ & $(2.80\pm0.47)\times10^{43}$ \\
522: Mkn~180 & $(4.5\pm1.8)10^{-11} (E/0.3\,\mathrm{TeV})^{-3.3\pm0.7} $ & \citet{MAGIC180} & $3.06\pm0.50$ & $(2.04\pm0.81)\times10^{43}$ \\
523: 1ES\,1959+650 & $(3.4\pm0.5)10^{-12} (E/1.0\,\mathrm{TeV})^{-2.72\pm0.14} $ & \citet{NadiaPaper} & $2.37\pm0.29$ & $(5.95\pm0.88)\times10^{43}$ \\
524: BL Lacertae & $(1.9\pm0.5)10^{-11} (E/0.3\,\mathrm{TeV})^{-3.64\pm0.54} $ & \citet{MAGICBLLac} & $3.17\pm0.25$ & $(2.35\pm0.62)\times10^{43}$ \\
525: PKS\,0548-322 & $(1.9\pm0.4)10^{-13} (E/1.0\,\mathrm{TeV})^{-2.8\pm0.3} $ & \citet{0548icrc} & $2.38\pm0.28$ & $(9.66\pm2.05)\times10^{42}$ \\
526: PKS\,2005-489 & $(1.9\pm0.7)10^{-13} (E/1.0\,\mathrm{TeV})^{-4.0\pm0.4} $ & \citet{HESS2005} & $3.52\pm0.27$ & $(2.53\pm0.93)\times10^{43}$ \\
527: PKS\,2155-304 & $(1.96\pm0.12)10^{-12} (E/1.0\,\mathrm{TeV})^{-3.32\pm0.06}$ for $E< 700\,\mathrm{GeV}$, & \citet{Aharonian2005} & $2.43\pm0.64$ & $(6.32\pm0.39)\times10^{44}$ \\
528: & $(2.4^{+0.4}_{-0.3})10^{-12} (0.7\pm0.2)^{(3.79^{+0.46}_{-0.27}-3.15^{+0.10}_{-0.12})}$ & & & \\
529: & $\times (E/1.0\,\mathrm{TeV})^{-3.79^{+0.46}_{-0.27}}$ for $E> 700\,\mathrm{GeV}$ & & & \\
530: H\,1426+428 & $(2.9\pm1.1)10^{-11} (E/0.43\,\mathrm{TeV})^{-2.6\pm0.6} $ & \citet{2001ICRC....7.2622H}, & $1.58\pm0.23$ & $(7.42\pm2.81)\times10^{44}$ \\
531: & & \citet{2002AA...384L..23A} & & \\
532: 1ES\,0229+200 & $(2.34\pm0.37)10^{-14} (E/3.0\,\mathrm{TeV})^{-2.5\pm0.2} $ & \citet{0229} & $1.39\pm0.30$ & $(6.13\pm0.98)\times10^{43}$ \\
533: H\,2356-309 & $(3.08\pm0.75)10^{-13} (E/1.0\,\mathrm{TeV})^{-3.06\pm0.4} $ & \citet{HESSAGNNature} & $1.77\pm0.37$ & $(3.29\pm0.80)\times10^{44}$ \\
534: 1ES\,1218+304 & $(8.1\pm2.1)10^{-11} (E/0.25\,\mathrm{TeV})^{-3.0\pm0.4} $ & \citet{MAGIC1218} & $1.97\pm0.40$ & $(1.26\pm0.33)\times10^{45}$ \\
535: 1ES\,1101-232 & $(4.44\pm0.74)10^{-13} (E/1.0\,\mathrm{TeV})^{-2.88\pm0.14} $ & \citet{HESSAGNNature} & $1.33\pm0.37$ & $(3.19\pm0.53)\times10^{44}$ \\
536: 1ES\,0347-121 & $(4.52\pm0.85)10^{-13} (E/1.0\,\mathrm{TeV})^{-3.10\pm0.23} $ & \citet{0347} & $1.76\pm0.14$ & $(3.19\pm0.60)\times10^{44}$ \\
537: 1ES\,1011+496 & $(2.0\pm0.1)10^{-10} (E/0.2\,\mathrm{TeV})^{-4.0\pm0.5} $ & \citet{1011} & $2.56\pm0.29$ & $(15.4\pm0.77)\times10^{44}$ \\
538: PG\,1553+113$^a$ & $(1.8\pm0.3)10^{-10} (E/0.2\,\mathrm{TeV})^{-4.21\pm0.25} $ & \citet{MAGIC1553published} & $3.68\pm0.68$ & $(6.40\pm1.07)\times10^{43}$ \\
539: PG\,1553+113$^b$ & $(1.8\pm0.3)10^{-10} (E/0.2\,\mathrm{TeV})^{-4.21\pm0.25} $ & \citet{MAGIC1553published} & $2.34\pm0.46$ & $(3.14\pm0.52)\times10^{45}$ \\
540: %
541: \hline
542: Mkn~421$^c$ & $(23.40\pm0.73)10^{-11} (E/1.0\,\mathrm{TeV})^{-2.32\pm0.03} $ & \citet{Krennrich2002} & $2.09\pm0.30$ & $(1.08\pm0.03)\times10^{45}$ \\
543: Mkn~501$^c$ & $(2.50\pm0.16)10^{-10} (E/1.0\,\mathrm{TeV})^{-2.22\pm0.04} $ & \citet{AharonianMkn501-1999a} & $1.95\pm0.41$ & $(1.39\pm0.09)\times10^{45}$ \\
544: 1ES\,2344+514$^c$ & $(5.1\pm1.0)10^{-11} (E/1.0\,\mathrm{TeV})^{-2.54\pm0.17} $ & \citet{Schroedter} & $2.20\pm0.31$ & $(6.65\pm1.30)\times10^{44}$ \\
545: 1ES\,1959+650$^c$ & $(1.23\pm0.25)10^{-10} (E/1.0\,\mathrm{TeV})^{-2.78\pm0.12} $ & \citet{2005ApJ...621..181D} & $2.43\pm0.29$ & $(2.25\pm0.46)\times10^{45}$ \\
546: PKS\,2155-304$^c$ & $(2.06\pm0.16)10^{-10}
547: (E/1.0\,\mathrm{TeV})^{-2.71\pm0.06}$ for $E< 340\,\mathrm{GeV}$, & \citet{2155flare} & $2.28\pm0.40$ & $(2.74\pm0.17)\times10^{46}$ \\
548: & $(2.06\pm0.16)10^{-10} (0.430\pm0.022)^{(3.53\pm0.05)-(2.71\pm0.06)}$ & & & \\
549: & $\times (E/1.0\,\mathrm{TeV})^{-3.53\pm0.05}$ for $E> 340\,\mathrm{GeV}$ & & & \\
550: %
551: \hline
552: \end{tabular}
553:
554: \medskip
555: $\Gamma$ denotes the reconstructed (intrinsic) VHE spectral
556: power-law index at 500 GeV and $\nu_\gamma L_\gamma$ represents
557: the source luminosity at 500 GeV. Both values were calculated from
558: the measured spectra assuming a \citeauthor{kneiske4} `low-IR' EBL
559: density. $^a$ at an assumed $z=0.1$.~$^b$ at an assumed
560: $z=0.3$.~$^c$ spectrum measured during a flare state of the
561: respective blazar.
562: \end{minipage}
563: \end{table*}
564: %
565: For the sources Mkn~421, Mkn~501, PKS\,2155-304, 1ES\,1959+650,
566: and 1ES\,2344+514 observations of clearly distinct flux states
567: exist. Accordingly, for each of those sources two spectra, one
568: `low-state' and one `high-state' spectrum, are considered.
569: Low-state spectra are characterised by the absence of high ($\ga
570: 0.2$~Crab units\footnote{The Crab nebula exhibits a strong,
571: constant VHE $\gamma$-ray flux and is therefore often considered a
572: standard candle in VHE $\gamma$-ray astronomy}) flux levels and
573: short-term variability (probably beyond instrumental sensitivity
574: though), while flare spectra were obtained during outbursts of the
575: respective sources ({\em viz.} the 1995 December 20 flare of
576: 1ES\,2344+514, the 1997 flare of Mkn~501, the 2002 flare of
577: 1ES\,1959+650 and the 2006 July 28 flare of PKS\,2155-304). At
578: present for none of these sources, even for those with only one
579: flux state, can a true baseline flux state be claimed, although
580: low-flux states have been observed
581: \citep{NadiaPaper,MAGIC2344published}. Long-term monitoring
582: campaigns are currently performed to address this issue
583: \citep{magicmonitoring,veritasmonitoring,hessmonitoring}.
584:
585: The measured spectra suffer $\gamma\gamma$ absorption on photons
586: of the EBL as shown in Sect.~\ref{sect:grhorsect}. The intrinsic
587: source spectra are reconstructed employing \citep{DanielDiplom}
588: the EBL `low-IR' model given in \citet{kneiske4}, which assumes
589: the least possible infrared star formation rate as allowed by
590: galaxy counts and which is in reasonable agreement with other
591: models \citep*{primack421,steckerNEW}. Note that due to the fact
592: that the VHE $\gamma$-rays are attenuated exponentially with the
593: optical depth, an accurate knowledge of the EBL is crucial for the
594: individual interpretation of the intrinsic VHE $\gamma$-ray
595: spectra.
596:
597: In the following, I will use two observables to characterise the
598: VHE $\gamma$-ray emission: The $K$-corrected \citep{hogg-02}
599: luminosity at $500\,\mathrm{GeV}$, $\nu_\gamma L_{\gamma} = 4 \pi
600: d_\mathrm{L}^2 \cdot (500\,\mathrm{GeV})^2
601: F(500\,\mathrm{GeV}/(1+z))/(1+z)$ with the luminosity distance
602: $d_\mathrm{L}$ and the intrinsic photon index $\Gamma$ in the
603: region around 500~GeV, which is determined by fitting the
604: intrinsic spectra with pure power-laws of the form
605: $\mathrm{d}F/\mathrm{d}E = f_0 \cdot (E/E_0)^{-\Gamma}$. These two
606: parameters act as proxies for the peak position and the spectral
607: shape on the falling edge of the high-energy bump, which cannot,
608: as explained before, easily be determined from the existing VHE
609: data. For extraction of the luminosity and of the spectral slope,
610: the region around 500 GeV was chosen because all blazars under
611: study have measured spectra in this energy region. All
612: calculations and fits have been performed in energy ranges where
613: $\gamma$-ray spectra for the respective blazars have actually been
614: observed, so that no extrapolations in energy regions not covered
615: by the data were required. All in all, extragalactic source
616: observations included in this paper cover the energy range
617: $85\,\mathrm{GeV} \leq E \leq 11 \,\mathrm{TeV}$. For the
618: determination of the luminosity distances the cosmological
619: parameters given in \citet{WMAPpaperpublished} were used:
620: $\Omega_\mathrm{m}h^2 = 0.127^{+0.007}_{-0.013}$;
621: $\Omega_\mathrm{b}h^2 = 0.0223^{+0.0007}_{-0.0009}$ with the
622: Hubble constant $H_0=100\cdot
623: h\,\mathrm{km}\,\mathrm{s}^{-1}\,\mathrm{Mpc}^{-1} =
624: 73^{+3}_{-3}\,\mathrm{km}\,\mathrm{s}^{-1}\,\mathrm{Mpc}^{-1}$.
625:
626: Tab.~\ref{tab:parvhe} also shows the resulting source luminosities
627: in the VHE region and the spectral slope of the reconstructed
628: source-intrinsic spectra. While the luminosities range from
629: $\approx 10^{43}\,\mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{sr}^{-1}$
630: to $\approx
631: 3\times10^{45}\,\mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{sr}^{-1}$
632: ($\approx
633: 10^{45}\,\mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{sr}^{-1}$ to
634: $\approx
635: 3\times10^{46}\,\mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{sr}^{-1}$
636: for blazars in outburst), the photon indices of the reconstructed
637: intrinsic spectra vary between $\Gamma=1.4 - 3.3$, except for
638: 1ES\,1101-232, which probably has an intrinsic spectrum peaking
639: far beyond $E=1$\,TeV \citep{1101}. The VHE measurements lie close
640: to, but generally above the maximum of the high-energy bump (which
641: occurs at $\Gamma=2$). Spectra with $\Gamma<1.5$ are difficult to
642: explain in current acceleration models
643: (\citealt{2001RPPh...64..429M,HESSAGNNature}; but see
644: \citealt{SS3,SS4,sbs07} for models that would explain harder
645: spectra). The rather hard intrinsic slopes found for some sources
646: go in line with indications that the EBL absorption effects are
647: still smaller than currently modelled
648: \citep[e.g.][]{HESSAGNNature}: a lower EBL level would soften the
649: intrinsic spectra inferred here, i.e. increase the value of
650: $\Gamma$.
651:
652: Before instruments like MAGIC and H.E.S.S. became operational, the
653: average observed photon index in BL~Lac objects was
654: $\Gamma\approx2.3$. This raised the expectation that AGN would in
655: general exhibit rather hard spectra in the VHE range, which also
656: would be compatible with the average EGRET blazar spectrum (at
657: MeV-GeV energies), found to have a slope of
658: $\Gamma_\mathrm{EGRET}=2.27$ \citep{2007ApJ...666..128V}. In fact,
659: the average spectral slope at VHE has not changed much, although
660: the scatter of $\Gamma$ increased as a new population of objects
661: with intrinsically rather soft spectra (Mkn\,180, PKS\,2005-489)
662: has been tapped, and at the same time distant, hard-spectrum
663: sources were found.
664:
665: For PG\,1553+113 with its unknown redshift (see
666: Sect.~\ref{sect:comp:501lumilim} for details), two possible
667: distances ($z=0.1$, $z=0.3$) were assumed in this paper. The
668: resulting `intrinsic spectra', however, are only given for
669: illustrative purposes and are not used for any conclusions
670: throughout this study unless stated otherwise.
671:
672: \section{Correlation studies}
673: \label{sect:corstudies}
674: \subsection{Correlation of X-ray, optical, radio and VHE $\gamma$-ray luminosity}
675: \label{sect:corstudies1}
676: %
677: In SSC models, the X-ray and the VHE emission are closely
678: connected, owing to their common origin. While in some blazars
679: clear evidence for a corresponding correlation has been observed
680: \citep[Mkn 421,][]{krawczinski,bla421,MAGIC421published}, the
681: connection is only weak for other ones \citep[Mkn
682: 501,][]{MAGIC-501} or even non-existing during some flare states
683: \citep[the 1ES 1959+650 {\it orphan flare}
684: case,][]{2005ApJ...621..181D}. Fig.~\ref{fig:comp:compaxrc} shows
685: $\nu_\gamma L_\gamma$ versus the X-ray luminosity at 1 keV
686: ($\nu_\mathrm{X} L_\mathrm{X}$; from \citealt{costamante}). Note
687: that high thermal contributions at 1 keV are unlikely and would
688: imply a very high amount of gas and pressure. Unfortunately, the
689: X-ray and VHE data have not been taken simultaneously.
690: While VHE measurements during outbursts were not used in the
691: Figure, variations in the X-ray domain for the blazars under study
692: are, according to the compilation of X-ray fluxes in
693: \citet{2001AA...375..739D}, not larger than a factor 4.6 (most
694: extreme object: Mkn\,501) with an average of a factor 1.4 and a
695: variance of 2.2. According to expectations, a trend towards a
696: correlation is visible. When including all data points except for
697: those representing PG\,1553+113, I find a correlation coefficient
698: of $r=0.76^{+0.09}_{-0.14}$, which is within 3.6 standard
699: deviations different from zero. A linear fit to the data yields a
700: slope of $m=1.11\pm0.09$ ($\chi^2_\mathrm{red} = 61.8/14$).
701: %
702: \begin{figure*}
703: \center{\includegraphics[width=.65\linewidth]{fig05}
704: \caption{VHE $\gamma$-ray luminosity $\nu_\gamma L_\gamma$ vs.
705: X-ray luminosity at 1 keV, $\nu_\mathrm{X} L_\mathrm{X}$ for 17
706: VHE blazars. The two data points for PG\,1553+113 (open crosses)
707: are for assumed redshifts of $z=0.1$ ($\nu_\gamma
708: L_{\gamma}=6.4\times10^{43}\,\mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$)
709: and $z=0.3$ ($\nu_\gamma
710: L_{\gamma}=3.1\times10^{45}\,\mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$),
711: respectively, and are not used in the fit and for determining the
712: correlation coefficient (see text). It should be noted that VHE
713: data points have an additional systematic error of typically 35
714: per cent. The systematic error of the X-ray luminosities is
715: unfortunately unknown.\label{fig:comp:compaxrc}}}
716: \end{figure*}
717:
718: Recently, optical triggers lead to the successful discoveries of
719: the VHE blazars Mkn~180 \citep{MAGIC180} and 1ES\,1011+496
720: \citep{1011}. At times of lower optical emission the latter in
721: fact showed a lower flux and thus only yielded a marginal
722: detection \citep{MAGIC-UL-Paper}. This seems to imply the
723: existence of a VHE -- optical connection, which would be a
724: convenient proxy for finding new VHE blazars. However, in the past
725: no \citep{bla421,2006ApJ...641..740R,MAGIC1553published} or only
726: weak \citep{1996ApJ...472L...9B} evidence for optical or radio to
727: VHE correlations was found for individual observations of Mkn~421
728: and PG\,1553+113.
729:
730: The optical flux data at 5500~\AA\, used here were collected from
731: several source catalogues by \citet{costamante}, as were the radio
732: flux data at 5~GHz. I converted them into luminosities taking into
733: account the appropriate luminosity distances $d_\mathrm{L}$.
734: Fig.~\ref{fig:comp:VHE_vs_O} shows the corresponding correlations:
735: the data in the VHE -- optical plane feature a larger scatter than
736: that in the VHE -- X-ray data, while in the VHE -- radio plane no
737: clear trend is seen.
738:
739: \begin{figure*}
740: \center{\includegraphics[width=\linewidth]{fig06}
741: \caption{{(a)} Correlation of VHE luminosity $\nu_\gamma
742: L_\gamma$ and optical luminosity $\nu_\mathrm{O} L_\mathrm{O}$.
743: {(b)} Correlation of VHE luminosity $\nu_\gamma L_\gamma$ and
744: radio luminosity $\nu_\mathrm{R} L_\mathrm{R}$. The two data
745: points for PG\,1553+113 (open crosses) are for assumed redshifts
746: of $z=0.1$ ($\nu_\gamma
747: L_{\gamma}=6.4\times10^{43}\,\mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$)
748: and $z=0.3$ ($\nu_\gamma
749: L_{\gamma}=3.1\times10^{45}\,\mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$),
750: respectively. \label{fig:comp:VHE_vs_O}}}
751: \end{figure*}
752:
753: VHE blazars might populate only restricted ranges in X-ray,
754: optical, or radio luminosity distributions. To test this, I use
755: X-ray, optical and radio data from the full set of 246 sources
756: considered by \citet{costamante}. I rejected the sources for which
757: no redshift is known and converted the remaining 183 fluxes into
758: luminosities. These blazars are compared to the VHE blazars in
759: Fig. \ref{fig:comp:CG_lumiHistograms}. In none of the three
760: distributions can substantial deviations of the VHE blazars from
761: the overall set of blazars be found.
762:
763: \begin{figure*}
764: \center{\includegraphics[width=\linewidth]{fig07}
765: \caption{Histograms of {(a)} 1 keV X-ray, {(b)} 5500
766: \AA\, optical and {(c)} 5 GHz radio luminosity of the VHE blazars and all
767: those blazars considered by \citet{costamante} with determined redshifts (183
768: out of the total set of 246 blazars).
769: \label{fig:comp:CG_lumiHistograms}}}
770: \end{figure*}
771:
772: \subsection{Correlations of intrinsic photon index and VHE luminosity with the synchrotron peak frequency}
773:
774: In SSC models the VHE peak, identified with the inverse Compton
775: (IC) peak with a maximum at $\nu^\mathrm{IC}_\mathrm{peak}$,
776: resembles the form \citep[e.g. ][]{fossati} of the synchrotron
777: peak at $\nu^\mathrm{Sy}_\mathrm{peak}$, displaced by the squared
778: Lorentz factor $\nu^\mathrm{IC}_\mathrm{peak} /
779: \nu^\mathrm{Sy}_\mathrm{peak} \sim \gamma^2$
780: \citep{Tavecchio1998}. \citet*{Nieppola} collected
781: (non-simultaneous) multiwavelength data for a large ($>300$) set
782: of blazars, which includes eleven of the VHE blazars under study
783: here. The data, covering frequencies from the 5 GHz radio to the
784: 2.4 keV X-ray domain, were used by \citeauthor{Nieppola} to
785: reconstruct the SEDs of these blazars and to infer the synchrotron
786: peak frequency $\nu_\mathrm{peak}$. I test for correlations
787: between $\nu^\mathrm{Sy}_\mathrm{peak}$ and the VHE luminosity and
788: spectral slope (Fig.~\ref{fig:comp:VHE_vs_Npeak}). According to
789: expectations from SSC models, I find a correlation of the photon
790: index $\Gamma$ with $\nu^\mathrm{Sy}_\mathrm{peak}$. The
791: correlation coefficient is $r=-0.63^{+0.28}_{-0.18}$ and a linear
792: fit to the data as shown in Fig.~\ref{fig:comp:VHE_vs_Npeak}a
793: yields a $\chi^2_\mathrm{red}=12.1/8$.
794: Fig.~\ref{fig:comp:VHE_vs_Npeak}b shows the corresponding data
795: points in the $\nu_\gamma L_\gamma -
796: \nu^\mathrm{Sy}_\mathrm{peak}$ plane, in which no correlation is
797: apparent. Finally, in Fig.~\ref{fig:comp:MBH_vs_Npeak}, I check
798: whether the LBL--HBL transition of the VHE blazars is connected
799: with the BH masses of their host galaxies. Ten of the VHE blazars,
800: for which both $\nu^\mathrm{Sy}_\mathrm{peak}$ and $M_\bullet$
801: measurements are available, show no trend towards a correlation.
802:
803: \begin{figure*}
804: \center{\includegraphics[width=\linewidth]{fig08}
805: \caption{{(a)} Photon index $\Gamma$ vs. synchrotron peak
806: frequency $\nu^\mathrm{Sy}_\mathrm{peak}$ and {(b)} VHE
807: luminosity vs. synchrotron peak frequency
808: $\nu^\mathrm{Sy}_\mathrm{peak}$ for eleven VHE blazars. The
809: synchrotron peak frequencies and the LBL-IBL-HBL classification
810: are taken from \citet{Nieppola}. The two data points for
811: PG\,1553+113 (open crosses) are for assumed redshifts of $z=0.1$
812: ($\Gamma=3.68\pm0.68$; $\nu_\gamma
813: L_{\gamma}=6.4\times10^{43}\,\mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$)
814: and $z=0.3$ ($\Gamma=2.34\pm0.46$; $\nu_\gamma
815: L_{\gamma}=3.1\times10^{45}\,\mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$),
816: respectively, and are not used in the fit and for determining the
817: correlation coefficient in the left Figure {(a)}.\label{fig:comp:VHE_vs_Npeak}}}
818: \end{figure*}
819:
820: \begin{figure*}
821: \center{\includegraphics[width=0.66\linewidth]{fig09}
822: \caption{Estimated BH mass vs. synchrotron peak frequency
823: $\nu^\mathrm{Sy}_\mathrm{peak}$ for those VHE blazars with both
824: quantities known. \label{fig:comp:MBH_vs_Npeak}}}
825: \end{figure*}
826:
827: \subsection{Correlation between the intrinsic photon index and the VHE $\gamma$-ray
828: luminosity}
829: %
830: Fig.~\ref{fig:comp:compasllum}a relates the intrinsic photon
831: indices $\Gamma$ to the VHE $\gamma$-ray luminosities $\nu_\gamma
832: L_\gamma$. For all objects under study (excluding flare states and
833: the PG\,1553+113 data), the $\Gamma-\nu_\gamma L_\gamma$
834: correlation reads as $\Gamma = \Gamma_0 + m \cdot \log_{10}
835: (\nu_\gamma L_\gamma)$ with $\Gamma_0 = 38.5 \pm6.8$ and
836: $m=-0.82\pm0.15$. The fit which led to this correlation has a
837: $\chi^2_\mathrm{red} = 40.18/14$. The line at $\Gamma=2$ denotes
838: the spectral slope at which a maximum in the SED occurs. Looking
839: at the spread of the observations in $\Gamma$ as a function of
840: $\nu_\gamma L_\gamma$, I notice that the distribution sharpens
841: towards $\Gamma=2$. This might reflect that the highest luminosity
842: occurs at $\Gamma=2$. Thus the spread of the data reflects the
843: spread of the shapes of the high energy peaks. The general
844: behaviour above $\Gamma\approx2$ -- the higher the VHE
845: $\gamma$-ray luminosity, the harder the spectrum -- can within SSC
846: models be described with a moving IC peak towards higher energies
847: with increasing luminosity. The five blazars with observed spectra
848: at quiescent and flare states partially behave in a similar manner
849: (Fig.~\ref{fig:comp:compasllum}b): Mkn~421
850: \citep{krawczinski,bla421}, Mkn~501 \citep{MAGIC-501} and
851: 1ES\,2344+514 \citep{MAGIC2344published} also show a spectral
852: hardening during high-flux states, flares and outbursts.
853: %
854: \begin{figure}
855: \center{\includegraphics[width=\linewidth]{fig10a}
856: \includegraphics[width=\linewidth]{fig10b} \caption{{(a)} Intrinsic photon
857: index vs. luminosity. Additional flare states of sources are
858: marked by grey circles. The results of a linear fit of the form
859: $\Gamma=\Gamma_0+m \log_{10} (\nu_\gamma L_\gamma)$ are given in
860: the figure. The two data points for PG\,1553+113 (open crosses),
861: not included in this fit, are for assumed redshifts of $z=0.1$
862: (low luminosity) and $z=0.3$ (high luminosity), respectively. {\it
863: (b)} As before, but only for the five blazars for which low and
864: high VHE $\gamma$ flux states have been observed.
865: \label{fig:comp:compasllum}}}
866: \end{figure}
867: %
868: Fig.~\ref{fig:comp:compasllum2}a shows the corresponding
869: luminosity differences $\Delta(\nu_\gamma L_\gamma)$ and slope
870: differences $\Delta \Gamma$. Mkn\,501 and 1ES\,2344+514 show a
871: similar change in spectral slope and a dynamical range of $\Delta
872: (\nu_\gamma L_\gamma) \approx 20$. The luminosity increase of
873: Mkn~421 observed up to now is much lower with $\Delta (\nu_\gamma
874: L_\gamma) \approx 10$. While the spectral slope of Mkn\,421 also
875: hardens with increasing luminosity, 1ES\,1959+650 and
876: PKS\,2155-304 show within errors no variation in their spectral
877: slopes during flares, while their luminosities increase rather
878: drastically by a factor of 40 and 50, respectively. When plotting
879: the luminosity difference versus the BH masses
880: (Fig.~\ref{fig:comp:compasllum2}b), a trend, although broken by
881: 1ES~1959+650, towards blazars with more massive BHs showing higher
882: dynamical ranges of their emission, is found.
883:
884: \begin{figure}
885: \center{\includegraphics[width=\linewidth]{fig11a}\\
886: \includegraphics[width=\linewidth]{fig11b}
887: \caption{{(a)} Evolution of intrinsic spectral index $\Gamma$
888: and source luminosity from low to high VHE $\gamma$ flux states:
889: luminosity ratio $\Delta (\nu_\gamma L_\gamma) = (\nu_\gamma
890: L_\gamma)_\mathrm{flare}/(\nu_\gamma L_\gamma)_\mathrm{low}$
891: versus the difference of intrinsic photon indices. {(b)}
892: luminosity ratio $\Delta (\nu_\gamma L_\gamma) = (\nu_\gamma
893: L_\gamma)_\mathrm{flare}/(\nu_\gamma L_\gamma)_\mathrm{low}$
894: versus $M_\bullet$. \label{fig:comp:compasllum2}}}
895: \end{figure}
896:
897: \subsection{A test for selection effects: redshift dependencies?} \label{sect:intrinsicspectraz}
898:
899: Possible correlations of $\Gamma$ and $\nu_\gamma L_\gamma$ with
900: the redshift $z$ are not expected, but may identify selection
901: effects in the data set and/or an inaccurate EBL model.
902: Conspicuously, very hard ($\Gamma \ll 2.0$) spectra have up to now
903: been only reconstructed for rather distant ($z\ga 0.1$) blazars
904: (Fig.~\ref{fig:comp:compaspecsz}a), a trend also observed by
905: \citet*{sbs07}, who used a smaller set of VHE blazars. At the same
906: time, none of the measured spectra of nearby sources shows
907: $\Gamma$ much smaller than 2.0, although for these blazars no
908: strong EBL modifications apply and the measured spectra should not
909: differ substantially from the intrinsic ones in the energy region
910: studied here. Why are only blazars with rather hard intrinsic
911: spectra visible at large distances ($z>0.1$)? Soft spectra
912: certainly fall more easily below the current instrumental
913: sensitivity limits. Another explanation for the prevalent hard
914: spectra at large $z$ is an overcorrection of the EBL attenuation
915: effects.
916:
917: Fig.~\ref{fig:comp:compaspecsz}b shows the corresponding
918: distribution of VHE luminosities $\nu_\gamma L_\gamma$ as a
919: function of $z$. Two curves indicate the sensitivity limit of
920: current IACTs \citep[e.g. ][]{HESS1553}, a significant detection
921: of 1 per cent of the flux of a Crab nebula-like source in 25
922: hours, and the sensitivity of previous (before 2002) IACTs of
923: $\approx 10$ per cent of this flux. Interestingly, most of the
924: sources found at $z>0.05$ seem to be rather low-luminosity sources
925: in the sense that their luminosity is not much higher than the
926: current instrumental sensitivity allows for. This means that not
927: only the substantially lower energy thresholds of the current
928: IACTs ($\la 100$~GeV as compared to $\approx 300$~GeV until 2004),
929: but also their increased sensitivity enabled some of the new
930: blazar discoveries. Exceptions to these detections close to the
931: current sensitivity limit are (trivially) the six VHE blazars
932: discovered before 2002, 1ES\,1011+496 (but discovered during an
933: optical flare, while its low-flux state seems to lie below the
934: current instrumental reach, \citealt{1011,MAGIC-UL-Paper}), and
935: 1ES\,1218+304, which in fact seems to have been detected in a
936: non-flare state. Note further that H\,1426+428 has not yet been
937: detected after 2002 (e.g. \citealt{veul,MAGIC-UL-Paper}), which
938: also currently places it below sensitivity limits.
939:
940: \begin{figure*}
941: \center{\includegraphics[width=\linewidth]{fig12}
942: \caption{Correlations of redshift with {(a)} the intrinsic
943: photon index and {(b)} VHE $\gamma$-ray luminosity. For
944: Mkn~421, Mkn~501, 1ES\,2344+514, PKS~2155-304 and 1ES\,1959+650
945: flare flux levels are also included in the plots; the
946: corresponding data points are marked by additional grey circles.
947: The two data points for PG\,1553+113 (open crosses) are for
948: assumed redshifts of $z=0.1$ and $z=0.3$, respectively. The dotted
949: curve in the right Figure {(b)} marks the sensitivity
950: \citep[e.g. ][]{HESS1553} limit of current IACTs (selection
951: effect);
952: the dashed curve indicates the sensitivity of previous (before
953: 2002) IACTs. \label{fig:comp:compaspecsz}}}
954: \end{figure*}
955:
956: \subsection{Correlations of intrinsic VHE $\gamma$-ray emission parameters with the black
957: hole mass} \label{sect:intrinsicspectra}
958:
959: The $\gamma$-ray production is thought to take place at shock
960: fronts inside the AGN jets at very close (sub-parsec) distances
961: from the central BH \citep{jester,uchiyama}. While the jet
962: production and collimation mechanism is still elusive, accepted
963: models are generally based on magnetohydrodynamic
964: \citep{1982MNRAS.199..883B,Kudoh} or electromagnetic jet models.
965: In the latter, a Poynting flux dominated flow is launched from a
966: Kerr BH \citep{1977MNRAS.179..433B} or from the accretion disc
967: \citep{1976MNRAS.176..465B}. The conversion from Poynting
968: dominance into particle dominance is not yet understood. The
969: properties of the blazar $\gamma$-ray emission are expected to be
970: connected to the properties of the central BH, like $M_\bullet$
971: and the BH spin, since scaling laws govern BH physics
972: \citep{2006Natur.444..730M}, in particular length and time-scales
973: of flows \citep{1999ARA&A..37..409M,2004inun.conf..175M}, e.g. the
974: orbital period of the last stable BH orbit. Currently, only
975: $M_\bullet$ can be reliably estimated; the BH spin remains
976: inaccessible by large. Moreover, the environment in which the BH
977: is embedded might be equally important; one of its properties, the
978: accretion rate, is indirectly accessible through the (radio) jet
979: power
980: \citep{2006ApJ...637..669L}, or from multiwavelength modelling
981: \citep{mt03}. A previous study of the connection of spectral
982: properties and $M_\bullet$ for five VHE blazars
983: \citep{Krawczynski1es1959-2004} did not find any correlations with
984: the BH mass.
985:
986: For 15 VHE blazars with known BH masses, I neither find a
987: correlation between $M_\bullet$ and the spectral slope $\Gamma$
988: (Fig.~\ref{fig:comp:compaspecs}a), nor between $M_\bullet$ and the
989: VHE $\gamma$-ray luminosity (Fig.~\ref{fig:comp:compaspecs}b).
990: While the VHE blazar set disfavours a dependency of the VHE
991: $\gamma$-ray emission properties studied here on $M_\bullet$, the
992: uncertainties of the $M_\bullet$ determination are still rather
993: large and might conceal otherwise interesting physics. In
994: particular, indirectly inferred $\sigma$ values from the
995: fundamental plane or bulge luminosity measurements have rather
996: large uncertainties. To improve on the $M_\bullet$ uncertainties,
997: it would be desirable to obtain direct measurements of $\sigma$
998: for all VHE blazars. Moreover, VHE emission properties may also
999: depend sensitively on the BH spin, the accretion rate, or on
1000: properties of the acceleration region in the jet. Also results on
1001: timing properties (see Sect.~\ref{sect:dcsect}) support such
1002: claims.
1003:
1004: \begin{figure*}
1005: \center{\includegraphics[width=\linewidth]{fig13}
1006: \caption{Correlations of black hole mass with {(a)} the
1007: intrinsic photon index and {(b)} VHE $\gamma$-ray
1008: luminosity. For Mkn~421, Mkn~501, 1ES\,2344+514 and 1ES\,1959+650
1009: flare flux levels are also included in the plots; the
1010: corresponding data points are marked by additional grey circles.
1011: \label{fig:comp:compaspecs}}}
1012: \end{figure*}
1013:
1014: \subsection{An upper distance limit for PG 1553+113}
1015: %
1016: \label{sect:comp:501lumilim}
1017: %
1018: The redshift determination for blazars is challenging, as these
1019: AGN generally exhibit only weak spectral lines. In several
1020: attempts, no emission or absorption lines could be found in the
1021: optical/IR spectrum of PG\,1553+113; see e.g. \citet{1553icrc}
1022: for results of a recent Very Large Telescope (VLT) campaign.
1023: The frequently cited initial determination of $z=0.36$
1024: \citep{MillerGreen} was found to be based on a misidentified
1025: emission line and could not be reproduced
1026: \citep*{Falomo,1994ApJS...93..125F}. VLT optical spectroscopy
1027: \citep{Sbar} yields a lower limit of $z>0.09$, while the analysis
1028: of Hubble Space Telescope images leads to the prediction of a
1029: redshift in the range of $z=0.3-0.4$ \citep*{treves-07}. Indirect
1030: methods employing the maximum slope
1031: \citep{HESS1553,MAGIC1553published} or the shape
1032: \citep{2007ApJ...655L..13M} of the VHE $\gamma$-ray spectrum find
1033: upper limits of $z<0.74$ and $z<0.42$, respectively.
1034:
1035: \begin{figure*}
1036: \center{\includegraphics[width=.66\linewidth]{fig14}
1037: \caption{Luminosity evolution for PG\,1553+113 assumed at
1038: different distances. Together with the luminosities of the other
1039: known VHE $\gamma$-ray emitting blazars, the luminosity of
1040: PG\,1553+113 as a function of its assumed distance is shown. The
1041: solid curve includes both the distance and the EBL attenuation
1042: effect, the latter calculated using the `low-IR' model given in
1043: \citet{kneiske4}. The dashed curve illustrates how weak the effect
1044: only by increasing the distance is. The dotted curve indicates the
1045: current IACT sensitivity limit neglecting EBL attenuation effects.
1046: Blazars in flaring state, marked with additional gray circles,
1047: were ignored for determining the redshift limit (see text).
1048: \label{fig:comp:1553lumilim}}}
1049: \end{figure*}
1050: %
1051: With increasing distance, the luminosity of PG\,1553+113 has to
1052: increase stronger than quadratic due to EBL $\gamma\gamma$
1053: absorption as to sustain the measured VHE flux.
1054: Fig.~\ref{fig:comp:1553lumilim} shows the source luminosity (1)
1055: when only considering the distance effect and (2) when also taking
1056: into account the EBL absorption. Due to the exponential behaviour
1057: of the EBL attenuation, the latter effect is by far dominant. I
1058: assume here that PG\,1553+113 is an `off the shelf' blazar, i.e.
1059: with no extraordinarily high luminosity $\nu_\gamma L_\gamma$.
1060: This assumption is difficult to quantify; given the overall
1061: dynamical range of the (non-flare) blazar luminosities in this
1062: study of $\approx 75$, I consider the case in which the luminosity
1063: of PG\,1553+113 is not more than 30 (1000) times higher than the
1064: highest luminosity found in the sample. Further, PG\,1553+113 has
1065: in two years of observations not shown any apparent flaring
1066: behaviour; within a factor of three, the measured flux was
1067: constant \citep{HESS1553,MAGIC1553published,icrc1553hess}.
1068: Therefore I consider only sources in non-flare states, of which
1069: 1ES\,1218+304 with $\nu_\gamma L_\gamma=1.3\times10^{45}\,
1070: \mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{sr}^{-1}$ is the most
1071: luminous one. A 30-times higher luminosity then implies a limit of
1072: $z<0.45$, while an extreme luminosity of $\nu_\gamma
1073: L_{\gamma}=1.3\times10^{48}\,
1074: \mathrm{erg}\,\mathrm{sr}^{-1}\,\mathrm{s}^{-1}$ yields a limit of
1075: $z<0.64$. These limits do not only depend on a good knowledge of
1076: the EBL over a wide range in redshift, but also on the assumed
1077: maximum VHE blazar luminosity that strongly depends on the Doppler
1078: factor $\delta$. In any case, either a strikingly high luminosity
1079: or a very high $\delta$ is needed to explain the observations
1080: should PG\,1553+113 be more distant than $z\ga0.35$. Note that
1081: $\delta \la 20$ suffices for most of the blazars modelled up to
1082: now during non-outburst times, as also for flare observations
1083: \citep[e.g. Mkn\,421, ][]{Maraschi60}. SSC modelling for
1084: PG\,1553+113 resulted in $\delta=21$
1085: \citep{costamante,MAGIC1553published}.
1086:
1087: The difficulties in finding emission and absorption lines might
1088: indicate a very close alignment of the jet axis of PG~1553+113 to
1089: our line of sight. Very-long baseline interferometry (VLBI)
1090: imaging is available for some VHE HBLs detected before 2002
1091: \citep{1999ApJ...525..176P,2002ApJ...579L..67E,2004ApJ...600..115P}
1092: and for BL~Lacertae \citep*{2000ApJS..129...61D}, while additional
1093: measurements of the recently found VHE blazars are underway
1094: \citep*{PinerAK}. VLBI essentially confirms the close alignment of
1095: the jet to our line of sight on sub-parsec scale and finds opening
1096: angles of typically few degrees. These results cannot yet be used
1097: for quantitative correlation studies though.
1098:
1099: We have to eventually consider the possibilities that PG\,1553+113
1100: is a rather distant source and the blazar populations at large
1101: distances show significantly different properties than the
1102: close-by objects at $z<0.2$, and that such very extreme objects
1103: are so rare that a sufficiently large volume had to be probed to
1104: find one of them.
1105:
1106: \subsection{Correlations with the X-ray duty cycle and the VHE variability time-scale}
1107: \label{sect:dcsect}
1108: %
1109: Following a method described in \citet{Krawczynski1es1959-2004}, I
1110: determine the time fraction (`duty cycle') for which the
1111: $(2-10)\,\mathrm{keV}$ X-ray flux exceeds the average flux by 50
1112: per cent. In this paper, for a blazar to be regarded `on duty'
1113: this deviation is additionally required to be significant on the
1114: $3\sigma$ level. 2-10 keV X-ray light curves are obtained from the
1115: All-Sky Monitor detector\footnote{available at
1116: http://xte.mit.edu/} on board the Rossi X-ray Timing Explorer
1117: ({\it RXTE}) and are available from 1996 January 5 on.
1118: Fig.~\ref{fig:comp:compadutycycle} shows the corresponding light
1119: curves along with the resulting duty cycles. Objects which are
1120: classified as extreme BL~Lacs (Mkn~501, 1ES\,2344+514,
1121: H\,1426+428, H\,2356-309 and PKS\,0548-322,
1122: \citealt{2001A&A...371..512C}) show substantially higher activity
1123: than the other blazars. Note also the outstanding role of Mkn 421
1124: with its substantially higher duty cycle as compared to all other
1125: objects. Fig.~\ref{fig:comp:compaspecstime}a shows the duty cycle
1126: as a function of the BH mass for the VHE blazars. No trends are
1127: visible, except for the observation that the three blazars with
1128: the most massive BHs ($\log(M_\bullet/\mathrm{M}_{\sun})>8.8$) do
1129: not have duty cycles in excess of 17 per cent. This goes in line
1130: with a speculation of an anticorrelation between the X-ray flare
1131: duty cycle and $M_\bullet$ seen in the five VHE blazars studied in
1132: \citet{Krawczynski1es1959-2004}.
1133: %
1134: \begin{figure*}
1135: \includegraphics[width=.64\linewidth]{fig15}
1136: \caption{X-ray (2-10~keV) light curves of VHE $\gamma$-ray
1137: emitting blazars. The flare duty cycle, i.e. the fraction of time
1138: in which the respective object significantly exceeds its average
1139: X-ray flux by 50 per cent is shown. For convenience the ranges of
1140: the vertical axes are fixed for all plots. FSRQ: Flat spectrum
1141: radio quasar, LBL: low-frequency peaked BL~Lac object; HBL:
1142: high-frequency peaked BL~Lac object, synchrotron peak in the
1143: UV/X-ray range; XBL: extreme BL~Lac object.
1144: \label{fig:comp:compadutycycle}}
1145: \end{figure*}
1146: The distribution of the X-ray duty cycle as a function of
1147: luminosity is found to be rather flat
1148: (Fig.~\ref{fig:comp:compaspecstime}b), supporting the claim that
1149: variability on all scales is a defining property of blazars. Note,
1150: however, that Mkn\,501 and PKS\,2155-304, the objects which show
1151: the fastest variability in VHE $\gamma$-rays, are among the
1152: objects with a rather low duty cycle.
1153:
1154: \begin{figure*}
1155: \center{\includegraphics[width=\linewidth]{fig16}
1156: \caption{Correlations of the X-ray duty cycle with {(a)} the
1157: BH mass and {(b)} the VHE
1158: luminosity.\label{fig:comp:compaspecstime}}}
1159: \end{figure*}
1160:
1161: Blazars are characterised by a highly variable emission. In
1162: particular the VHE $\gamma$-ray emission is often found to be more
1163: variable than the emission at other wavelenghts. Still,
1164: comprehensive data on the temporal behaviour of VHE blazar
1165: emission has often been collected only for blazars in outburst,
1166: and thus, over short time spans. Although VHE light curves ranging
1167: over ten years' worth of measurements were collected occasionally
1168: \citep[e.g.][]{MAGIC2344published,MAGIC-501,tlucmad}, the sampling
1169: is only sparse, and continuous, unbiased long-term monitoring
1170: campaigns have not been started until 2005
1171: \citep{magicmonitoring,veritasmonitoring,hessmonitoring}.
1172:
1173: Variability time-scales $\tau$ are intimately linked to the
1174: extension of the region $R$ from which the observed emission
1175: originates by the causality condition $R \la \delta c \tau /
1176: (1+z)$. \citet{Cui2004} has suggested that the flare hierarchy
1177: seen in Mkn~501 long-term data implies a scale-invariant nature of
1178: the flare process and that there might not be any fundamental
1179: difference among long, intermediate and rapid flares. Although not
1180: thoroughly understood, the flares in blazars might be related to
1181: internal shocks in the jet \citep{ReesM87,Spada} or to major
1182: ejection events of new components of relativistic plasma into the
1183: jet \citep*{BoettcherAA,MasKirk}. Different flare time-scales thus
1184: may be caused by a hierarchy of inhomogeneities in the jet,
1185: energised so as to produce flares. As a timing property of the VHE
1186: $\gamma$-ray emission, the minimum flux doubling times for the VHE
1187: blazars were collected from literature and plotted versus
1188: $M_\bullet$ (Fig.~\ref{fig:comp:compaspecstime2}a) and the VHE
1189: luminosity (Fig.~\ref{fig:comp:compaspecstime2}b).
1190: %
1191: \begin{figure*}
1192: \center{\includegraphics[width=\linewidth]{fig17}
1193: \caption{Correlations of the VHE $\gamma$ flux doubling time
1194: $\tau$ with {(a)} the BH mass and {(b)} the VHE
1195: luminosity. Upper limits on $\tau$ are taken from
1196: \citet{icrc2005hess} and from the references given in
1197: Tab.~\ref{tab:parvhe}. Additional data points represent the
1198: recently found fast doubling times for Mkn~501 \citep{MAGIC-501} and PKS\,2155-304
1199: \citep{2155flare}. Further for Mkn~421
1200: \citep{Gaidos1996}, 1ES\,2344+514 and 1ES\,1959+650 doubling
1201: times during flare states are also included, which are marked by
1202: additional grey circles. The two data points for PG\,1553+113
1203: represent assumed redshifts of $z=0.1$ (lower luminosity) and
1204: $z=0.3$ (higher luminosity), respectively.
1205: \label{fig:comp:compaspecstime2}}}
1206: \end{figure*}
1207: %
1208: Because of the very limited data base of VHE variability
1209: measurements, which also is biased towards high-flux states and
1210: outbursts, time-scales are mostly given as upper limits, which
1211: disables strong conclusions.
1212: %
1213:
1214: The observed VHE flux doubling times do not scale with the BH mass
1215: (Fig.~\ref{fig:comp:compaspecstime2}a), which may simply mean that
1216: (1) the flaring mechanism is working in a much smaller region than
1217: the BH radius/least stable orbit and more importantly (2) the BH
1218: and its properties as such do not influence the flaring process
1219: substantially, and the embedding environment of the BH and the jet
1220: environment play more dominant roles (e.g., the accretion power of
1221: the system). The extremely short doubling times of $\tau<5$~min
1222: recently found for Mkn~501 \citep{MAGIC-501} and PKS~2155-304
1223: \citep{2155flare} clearly do not support such a scaling either;
1224: even more, they are vastly incompatible with the rotation period
1225: of particles on the last stable orbit, which e.g. for the BH in
1226: Mkn~501 is $T=8.4$~d. Additionally, the size scale implied by
1227: $\tau<5$~min requires large Doppler factors $\delta$ in the
1228: $\gamma$-ray production regions as to avoid self-absorption. In
1229: contrast to the expected scaling behaviour of the flow properties
1230: around BH with their masses, the three AGN that host rather
1231: massive BHs, PKS\,2155-304, Mkn~501 and Mkn~421 were found to
1232: exhibit the shortest variability time-scales. Until minute-scale
1233: flaring was found in PKS\,2155-304, one could have argued that
1234: short time variability could only be measured for high fluxes due
1235: to the proximity of the respective sources: Mkn 421 and Mkn 501
1236: are the closest blazars at $z<0.034$, while PKS~2155-304 is
1237: located at $z=0.116$. The present observations could, however,
1238: still be the cause of a selection effect: Minute-scale flares were
1239: found only during exceptional high-flux states of the respective
1240: sources so far.
1241:
1242: In Fig.~\ref{fig:comp:geometrical} I translate the BH masses into
1243: the corresponding gravitational radii $r_\mathrm{g}=GM/c^2$
1244: (=$0.5$~Schwarzschild radius) and the observed minimal variability
1245: time-scales into the sizes of the emission regions $R \la \delta c
1246: \tau / (1+z)$, assuming $\delta=10$. For blazars in which fast
1247: variability has been observed, the extreme compactness of the
1248: emission region is apparent, being clearly comparable to or
1249: smaller than the Schwarzschild radius of the central BH. Even if
1250: the flares were driven by extremely large Doppler factors, say
1251: $\delta\approx100$, this would still result in emission regions
1252: smaller than the Schwarzschild radius for the Mkn\,501 and
1253: PKS\,2155-304 flares.
1254:
1255: \begin{figure*}
1256: \center{\includegraphics[width=.78\linewidth]{fig18}
1257: \caption{Gravitational radius $r_\mathrm{g}$ vs. upper limits on
1258: the size of the VHE $\gamma$-ray emission region $R$ determined by
1259: the causality condition. $\delta_1 = 10 \cdot \delta$ denotes the
1260: Doppler factor. The lines correspond to emission region sizes of
1261: the last stable orbit radius (dotted line), of the Schwarzschild
1262: radius (dashed line), and of 10 per cent and 1 per cent of the
1263: Schwarzschild radius, respectively (dot-dashed lines).
1264: \label{fig:comp:geometrical}}}
1265: \end{figure*}
1266:
1267: \section{Concluding remarks}
1268: \label{sect:conc}
1269: %
1270: Before the new generation of IACT became operational, only six
1271: firmly detected extragalactic VHE $\gamma$-ray sources were known
1272: \citep[e.g.,][]{Moriicrc}; not for all of them differential energy
1273: spectra had been inferred. To date, there are 17 BL Lac objects
1274: and the FSRQ 3C~279 known to emit VHE $\gamma$-rays. This
1275: substantially enlarged set of blazars called for a synoptic
1276: study. I collected and derived intrinsic properties of the VHE
1277: $\gamma$-ray emission (luminosity, spectral hardness, temporal
1278: properties) and further included X-ray, optical and radio
1279: emission properties. As an accessible property of the BHs in the
1280: centres of the blazar host galaxies, the BH mass was also used.
1281: The studies yield the following results.
1282: %
1283: \begin{itemize}
1284: \item[(i)] So far, only blazars with $M_\bullet \ga 10^8
1285: \mathrm{M}_{\sun}$ show VHE $\gamma$ emission. Whether this
1286: experimental finding constitutes a selection effect or will reveal
1287: interesting physics, remains to be seen.
1288: %
1289: \item[(ii)] The VHE luminosity of the blazars under study and the
1290: corresponding X-ray luminosity show a hint of a correlation, as
1291: expected from leptonic acceleration/SSC models.
1292: %
1293: \item[(iii)] A correlation between the spectral slope in the VHE region
1294: and the peak location of the synchrotron peak is found. Such a
1295: correlation is also expected from SSC models.
1296: %
1297: \item[(iv)] There were no correlations found between the $\gamma$
1298: emission properties and the $M_\bullet$ of the galaxies that host
1299: the blazars. Also, no correlation of $M_\bullet$ could be observed
1300: with the flare duty cycles and the flare time-scales. Thus, VHE
1301: $\gamma$-ray emission properties may not dominantly depend on
1302: $M_\bullet$. Other possibly interesting BH parameters are not yet
1303: within instrumental reach.
1304: %
1305: \item[(v)] There is an indication that the VHE $\gamma$ luminosity is
1306: correlated with the spectral hardness. This correlation can be
1307: formulated as a decrease of $\Delta \Gamma \approx 0.82$ per
1308: decade of luminosity.
1309: %
1310: \item[(vi)] This behaviour is also found for some individual blazars
1311: that were observed in different emission states: There are
1312: indications that in some variable sources the observed spectra
1313: become harder with increasing luminosity, while in others no
1314: hardening is found.
1315: %
1316: \end{itemize}
1317:
1318: Investigations of the temporal properties of the X-ray emission
1319: show that the blazars with the most massive BHs have rather low
1320: duty cycles. The temporal behaviour of the VHE $\gamma$-ray
1321: emission is not well studied for most of the VHE blazars, but the
1322: present data do not seem to support a scaling of the flux doubling
1323: time-scales with the BH masses.
1324:
1325: Obviously we are still dealing with a low number of sources and
1326: certainly with an incomplete source sample, which leaves regions
1327: in the parameter space empty. Nevertheless, the rather impressive
1328: number of 17 VHE blazars has permitted first comparative studies.
1329: It also allows
1330: conclusions concerning EBL models: A hint at a marginal
1331: correlation between the intrinsic spectral hardness and the source
1332: distance is likely due to an EBL over-prediction. This result was
1333: also recently quantified by \citet{HESSAGNNature} using the two
1334: distant blazars 1ES\,1101-232 and H~2356-309. In clarifying the
1335: situation, more distant sources particularly with intrinsically
1336: hard spectra (as e.g. 1ES\,0229+200) will play a decisive role;
1337: PG\,1553+113 with its yet undetermined distance might also turn
1338: out to be a good candidate for such analyses once a measurement of
1339: its redshift succeeds. In the meantime, the observed luminosity
1340: distribution of the studied VHE $\gamma$ blazars was used to
1341: constrain the unknown distance of PG\,1553+113 by assuming that the
1342: properties of this blazar are not too different from the most
1343: extreme objects in the blazar sample. Conversely, a large distance
1344: of PG\,1553+113 implies an unusually high luminosity or an
1345: unusually high Doppler factor.
1346:
1347: It can be said that the era of blazar astronomy has been entered
1348: -- astronomy being understood as the study of generic properties
1349: of a given class of objects. With the currently known VHE
1350: blazars and the given instrumental sensitivity, VHE blazar
1351: astronomy starts to become less biased: Not only blazars with hard
1352: spectra or during outbursts have been detected, but also low
1353: (quiescent?) emission states in the VHE range are now being
1354: observed and studied.
1355:
1356: With continuing discoveries of new sources, it will become easier
1357: to disentangle EBL absorption effects and intrinsic absorption
1358: effects from the measured spectra. Already currently, the VHE
1359: blazar sample contains groups of objects at very similar
1360: distances, e.g. 1ES\,1218+304 and 1ES\,1101-232 ($\Delta
1361: z=0.004$), or the triplet 1ES\,2344+514, Mkn~180 and
1362: 1ES\,1959+650 ($\Delta z=0.003$). Such groups are ideal for
1363: direct comparisons between the respective individual objects, as
1364: they are subject to very similar EBL attenuation.
1365:
1366: Since 2003, on average three new blazars have been detected per
1367: year. When comparing this rate to the ten years it took to
1368: discover the first six VHE blazars, one can with good reason hope
1369: to accumulate many more sources in the near future, which will
1370: refine synoptic blazar studies as the pioneering one conducted
1371: here. In addition, the first observation of new object classes
1372: like LBLs and FSRQs will certainly help to sharpen the VHE view of
1373: a possible blazar sequence and the underlying physics in the near
1374: future.
1375:
1376: \section*{Acknowledgments}
1377: Luigi Costamante kindly provided the full object list from
1378: \citet{costamante}. I would like to thank Eckart Lorenz, Nina
1379: Nowak, W\l odek Bednarek and Hinrich Meyer for fruitful
1380: discussions on this study and the {\it RXTE} team for providing
1381: the all-sky monitor X-ray data. The financial support by Max
1382: Planck Society is gratefully acknowledged. This research has made
1383: use of NASA's Astrophysics Data System.
1384:
1385: %\bibliographystyle{mn2e}
1386: %\bibliography{papers}
1387:
1388: \begin{thebibliography}{}
1389:
1390: \bibitem[\protect\citeauthoryear{Aharonian}{Aharonian}{2000}]{AharonianHad} Aharonian F.~A., 2000, NewA, 5, 377
1391: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{1999a}]{AharonianMkn501-1999a} Aharonian F., et~al. (HEGRA collaboration), 1999a, \aap, 342, 69
1392: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{1999b}]{AharonianMkn4211999} Aharonian F.~A., et~al. (HEGRA collaboration), 1999b, \aap, 350, 757
1393: \bibitem[\protect\citeauthoryear{{Aharonian} et~al.,}{{Aharonian} et~al.}{2001a}]{2001A&A...366...62A} {Aharonian} F.~A., et~al. (HEGRA collaboration), 2001a, \aap, 366, 62
1394: \bibitem[\protect\citeauthoryear{{Aharonian} et~al.,}{{Aharonian} et~al.}{2001b}]{Aharonian2001} {Aharonian} F.~A., et~al. (HEGRA collaboration), 2001b, \apj, 546, 898
1395: \bibitem[\protect\citeauthoryear{{Aharonian} et~al.,}{{Aharonian} et~al.}{2002}]{2002AA...384L..23A} {Aharonian} F., et~al. (HEGRA collaboration), 2002, \aap, 384, L23
1396: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2004}]{Tluczykont} Aharonian F.~A., et~al. (HEGRA collaboration), 2004, \aap, 421, 529
1397: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2005a}]{Aharonian2005} Aharonian F.~A., et~al. (H.E.S.S. collaboration), 2005a, \aap, 430, 865
1398: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2005b}]{HESS2005} Aharonian F.~A., et~al. (H.E.S.S. collaboration), 2005b, \aap, 436, L17
1399: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2005c}]{2005A&A...441..465A} Aharonian F., et~al. (H.E.S.S. collaboration), 2005c, \aap, 441, 465
1400: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2006a}]{HESSAGNNature} Aharonian F.~A., et~al. (H.E.S.S. collaboration), 2006a, \nat, 440, 1018
1401: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2006b}]{HESS1553} Aharonian F.~A., et~al. (H.E.S.S. collaboration), 2006b, \aap, 448, L19
1402: \bibitem[\protect\citeauthoryear{{Aharonian} et~al.,}{{Aharonian} et~al.}{2006c}]{hess87n} {Aharonian} F., et~al. (H.E.S.S. collaboration), 2006c, \sci, 314, 1425
1403: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2007a}]{1101} Aharonian F., et~al. (H.E.S.S. collaboration), 2007a, \aap, 470, 475
1404: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2007b}]{2155flare} Aharonian F., et~al. (H.E.S.S. collaboration), 2007b, \apj, 664, L71
1405: \bibitem[\protect\citeauthoryear{Aharonian et~al.,}{Aharonian et~al.}{2007c}]{0347} Aharonian F., et~al. (H.E.S.S. collaboration), 2007c, \aap, 473, L25
1406: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2006a}]{NadiaPaper} Albert J., et~al. (MAGIC collaboration), 2006a, \apj, 639, 761
1407: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2006b}]{MAGIC1218} Albert J., et~al. (MAGIC collaboration), 2006b, \apj, 642, L119
1408: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2006c}]{MAGIC180} Albert J., et~al. (MAGIC collaboration), 2006c, \apj, 648, L105
1409: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2007a}]{MAGIC1553published} Albert J., et~al. (MAGIC collaboration), 2007a, \apj, 654, L119
1410: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2007b}]{MAGIC2344published} Albert J., et~al. (MAGIC collaboration), 2007b, \apj, 662, 892
1411: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2007c}]{MAGIC421published} Albert J., et~al. (MAGIC collaboration), 2007c, \apj, 663, 125
1412: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2007d}]{MAGICBLLac} Albert J., et~al. (MAGIC collaboration), 2007d, \apj, 666, L17
1413: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2007e}]{1011} Albert J., et~al. (MAGIC collaboration), 2007e, \apj, 667, L21
1414: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2007f}]{MAGIC-501} Albert J., et~al. (MAGIC collaboration), 2007f, \apj, 669, 862
1415: \bibitem[\protect\citeauthoryear{Albert et~al.,}{Albert et~al.}{2008}]{MAGIC-UL-Paper} Albert J., et~al. (MAGIC collaboration), 2008, \apj, preprint (arXiv:0706.4453)
1416: \bibitem[\protect\citeauthoryear{Baixeras et~al.,}{Baixeras et~al.}{2004}]{Baixeras} Baixeras C., et~al. (MAGIC collaboration), 2004, Nucl. Instr. Meth. A, 518, 188
1417: \bibitem[\protect\citeauthoryear{Barth, Ho \& Sargent}{Barth et~al.}{2003}]{Barth2003} Barth A.~J., Ho L.~C., Sargent W. L.~W., 2003, \apj, 583, 134
1418: \bibitem[\protect\citeauthoryear{{Bednarek}}{{Bednarek}}{1993}]{1993ApJ...402L..29B} {Bednarek} W., 1993, \apj, 402, L29
1419: \bibitem[\protect\citeauthoryear{Benbow \& B\"uhler}{Benbow \& B\"uhler}{2007}]{HESS-UL-ICRC} Benbow W., B\"uhler R. (H.E.S.S. collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0709.4598)
1420: \bibitem[\protect\citeauthoryear{Benbow, Boisson, B\"uhler \& Sol}{Benbow et~al.}{2007}]{icrc1553hess} Benbow W., Boisson C., B\"uhler R., Sol H. (H.E.S.S. collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0709.4602)
1421: \bibitem[\protect\citeauthoryear{Bender \& Kormendy}{Bender \& Kormendy}{2003}]{Bender-03} Bender R., Kormendy J., 2003, in Shaver P., Dilella L., Gim\'{e}nez A., eds, Astronomy, Cosmology and Fundamental Physics, p.~262 (Berlin and Heidelberg: Springer)
1422: \bibitem[\protect\citeauthoryear{{Bettoni}, {Falomo}, {Fasano}, {Govoni}, {Salvo} \& {Scarpa}}{{Bettoni} et~al.}{2001}]{Bettoni} {Bettoni} D., {Falomo} R., {Fasano} G., {Govoni} F., {Salvo} M., {Scarpa} R., 2001, \aap, 380, 471
1423: \bibitem[\protect\citeauthoryear{{Blandford}}{{Blandford}}{1976}]{1976MNRAS.176..465B} {Blandford} R.~D., 1976, \mnras, 176, 465
1424: \bibitem[\protect\citeauthoryear{{Blandford} \& {K\"{o}nigl}}{{Blandford} \& {K\"{o}nigl}}{1979}]{1979ApJ...232...34B}{Blandford} R.~D., {K\"{o}nigl} A., 1979, \apj, 232, 34
1425: \bibitem[\protect\citeauthoryear{{Blandford} \& {Payne}}{{Blandford} \& {Payne}}{1982}]{1982MNRAS.199..883B}{Blandford} R.~D., {Payne} D.~G., 1982, \mnras, 199, 883
1426: \bibitem[\protect\citeauthoryear{{Blandford} \& {Znajek}}{{Blandford} \& {Znajek}}{1977}]{1977MNRAS.179..433B}{Blandford} R.~D., {Znajek} R.~L., 1977, \mnras, 179, 433
1427: \bibitem[\protect\citeauthoryear{{B{\l}a{\.z}ejowski}, {Blaylock}, {Bond} et~al.,}{{B{\l}a{\.z}ejowski} et~al.}{2005}]{bla421}{B{\l}a{\.z}ejowski} M., {Blaylock} G., {Bond} I.~H., et~al., 2005, \apj, 630, 130
1428: \bibitem[\protect\citeauthoryear{{B{\"o}ttcher}, {Mause} \& {Schlickeiser}}{{B{\"o}ttcher} et~al.}{1997}]{BoettcherAA} {B{\"o}ttcher} M., {Mause} H., {Schlickeiser} R., 1997, \aap, 324, 395
1429: \bibitem[\protect\citeauthoryear{{Buckley} et~al.,}{{Buckley} et~al.}{1996}]{1996ApJ...472L...9B} {Buckley} J.~H., et~al. (Whipple collaboration), 1996, \apj, 472, L9
1430: \bibitem[\protect\citeauthoryear{{Catanese} et~al.,}{{Catanese} et~al.}{1998}]{Catanese1998} {Catanese} M., et~al. (Whipple collaboration), 1998, \apj, 501, 616
1431: \bibitem[\protect\citeauthoryear{Chadwick et~al.,}{Chadwick et~al.}{1999}]{Chadwick} Chadwick P.~M., et~al., 1999, \apj, 513, 161
1432: \bibitem[\protect\citeauthoryear{Cogan}{Cogan}{2007}]{VERITAS-UL-ICRC} Cogan P. (VERITAS collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0709.3695)
1433: \bibitem[\protect\citeauthoryear{Coppi}{Coppi}{1992}]{coppi2} Coppi P.~S., 1992, MNRAS, 258, 657
1434: \bibitem[\protect\citeauthoryear{Cortina et~al.,}{Cortina et~al.}{2005}]{Cortina} Cortina J., et~al. (MAGIC collaboration), 2005, in {Proceedings of the 29th International Cosmic Ray Conference} Vol.~5, Pune, pp 359--362
1435: \bibitem[\protect\citeauthoryear{Costamante \& Ghisellini}{Costamante \& Ghisellini}{2002}]{costamante} Costamante L., Ghisellini G., 2002, \aap, 384, 56
1436: \bibitem[\protect\citeauthoryear{{Costamante}, {Ghisellini}, {Giommi}, {Tagliaferri}, {Celotti}, {Chiaberge}, {Fossati}, {Maraschi}, {Tavecchio}, {Treves} \& {Wolter}}{{Costamante} et~al.}{2001}]{2001A&A...371..512C} {Costamante} L., {Ghisellini} G., {Giommi} P., {Tagliaferri} G., {Celotti} A., {Chiaberge} M., {Fossati} G., {Maraschi} L., {Tavecchio} F., {Treves} A., {Wolter} A., 2001, \aap, 371, 512
1437: \bibitem[\protect\citeauthoryear{Costamante, Benbow, Boisson, Pita \& Sol}{Costamante et~al.}{2007}]{icrc2005hess} Costamante L., Benbow W., Boisson C., Pita S., Sol H. (H.E.S.S. collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0710.4057, p.~122)
1438: \bibitem[\protect\citeauthoryear{Cui}{Cui}{2004}]{Cui2004} Cui W., 2004, \apj, 605, 662
1439: \bibitem[\protect\citeauthoryear{{Daniel}, {Badran}, {Bond}, {Boyle} et~al.,}{{Daniel} et~al.}{2005}]{2005ApJ...621..181D} {Daniel} M.~K., et~al. (Whipple collaboration), 2005, \apj, 621, 181
1440: \bibitem[\protect\citeauthoryear{{Dar} \& {Laor}}{{Dar} \& {Laor}}{1997}]{1997ApJ...478L...5D} {Dar} A., {Laor} A., 1997, \apjl, 478, L5
1441: \bibitem[\protect\citeauthoryear{{Denn}, {Mutel} \& {Marscher}}{{Denn} et~al.}{2000}]{2000ApJS..129...61D} {Denn} G.~R., {Mutel} R.~L., {Marscher} A.~P., 2000, \apjs, 129, 61
1442: \bibitem[\protect\citeauthoryear{{Dermer} \& {Schlickeiser}}{{Dermer} \& {Schlickeiser}}{1994}]{1994ApJS...90..945D} {Dermer} C.~D., {Schlickeiser} R., 1994, \apjs, 90, 945
1443: \bibitem[\protect\citeauthoryear{{Djorgovski} \& {Davis}}{{Djorgovski} \& {Davis}}{1987}]{1987ApJ...313...59D} {Djorgovski} S., {Davis} M., 1987, \apj, 313, 59
1444: \bibitem[\protect\citeauthoryear{{Donato}, {Ghisellini}, {Tagliaferri} \& {Fossati}}{{Donato} et~al.}{2001}]{2001AA...375..739D} {Donato} D., {Ghisellini} G., {Tagliaferri} G., {Fossati} G., 2001, \aap, 375, 739
1445: \bibitem[\protect\citeauthoryear{{Edwards} \& {Piner}}{{Edwards} \& {Piner}}{2002}]{2002ApJ...579L..67E} {Edwards} P.~G., {Piner} B.~G., 2002, \apjl, 579, L67
1446: \bibitem[\protect\citeauthoryear{{Falomo}}{{Falomo}}{1996}]{1996MNRAS.283..241F} {Falomo} R., 1996, \mnras, 283, 241
1447: \bibitem[\protect\citeauthoryear{Falomo \& Treves}{Falomo \& Treves}{1990}]{Falomo} Falomo R., Treves A., 1990, PASP, 102, 1120
1448: \bibitem[\protect\citeauthoryear{{Falomo}, {Kotilainen} \& {Treves}}{{Falomo} et~al.}{2002}]{2002ApJ...569L..35F} {Falomo} R., {Kotilainen} J.~K., {Treves} A., 2002, \apj, 569, L35
1449: \bibitem[\protect\citeauthoryear{{Falomo}, {Scarpa} \& {Bersanelli}}{{Falomo} et~al.}{1994}]{1994ApJS...93..125F} {Falomo} R., {Scarpa} R., {Bersanelli} M., 1994, \apjs, 93, 125
1450: \bibitem[\protect\citeauthoryear{Fazio \& Stecker}{Fazio \& Stecker}{1970}]{fs70} Fazio G.~G., Stecker F.~W., 1970, Nat, 226, 135
1451: \bibitem[\protect\citeauthoryear{{Ferrarese} \& {Merritt}}{{Ferrarese} \& {Merritt}}{2000}]{2000ApJ...539L...9F} {Ferrarese} L., {Merritt} D., 2000, \apj, 539, L9
1452: \bibitem[\protect\citeauthoryear{{Fossati}, {Maraschi}, {Celotti}, {Comastri} \& {Ghisellini}}{{Fossati} et~al.}{1998}]{fossati} {Fossati} G., {Maraschi} L., {Celotti} A., {Comastri} A., {Ghisellini} G., 1998, MNRAS, 299, 433
1453: \bibitem[\protect\citeauthoryear{Gaidos et~al.,}{Gaidos et~al.}{1996}]{Gaidos1996} Gaidos J.~A., et~al. (Whipple collaboration), 1996, Nature, 383, 319
1454: \bibitem[\protect\citeauthoryear{{Gebhardt}, {Bender}, {Bower}, {Dressler}, {Faber} et~al.,}{{Gebhardt} et~al.}{2000}]{Gebhardt} {Gebhardt} K., {Bender} R., {Bower} G., {Dressler} A., {Faber} S.~M., et~al., 2000, \apj, 539, L13
1455: \bibitem[\protect\citeauthoryear{Goebel, Backes, Bretz, Hayashida, Hsu, Mannheim, Mazin \& Wagner}{Goebel et~al.}{2007}]{magicmonitoring} Goebel F., Backes M., Bretz T., Hayashida M., Hsu C.-C., Mannheim K., Mazin D., Wagner R.~M., 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0709.2032)
1456: \bibitem[\protect\citeauthoryear{Gould \& Schr\'{e}der}{Gould \& Schr\'{e}der}{1966}]{gould} Gould R.~J., Schr\'{e}der G., 1966, \prl, 16, 252
1457: \bibitem[\protect\citeauthoryear{{Graham}}{{Graham}}{2007}]{2007MNRAS.379..711G} {Graham} A.~W., 2007, \mnras, 379, 711
1458: \bibitem[\protect\citeauthoryear{{Gu}, {Cao} \& {Jiang}}{{Gu} et~al.}{2001}]{2001MNRAS.327.1111G} {Gu} M., {Cao} X., {Jiang} D.~R., 2001, \mnras, 327, 1111
1459: \bibitem[\protect\citeauthoryear{Hartman, Bertsch, Bloom et~al.,}{Hartman et~al.}{1999}]{hartman} Hartman R.~C., Bertsch D.~L., Bloom S.~D., et~al., 1999, \apjs, 123, 79
1460: \bibitem[\protect\citeauthoryear{Hauser \& Dwek}{Hauser \& Dwek}{2001}]{hauser} Hauser G.~H., Dwek E., 2001, ARA\&A, 39, 249
1461: \bibitem[\protect\citeauthoryear{Hinton}{Hinton}{2004}]{HESStech} Hinton J., 2004, NewAR, 48, 331
1462: \bibitem[\protect\citeauthoryear{{Hogg}, {Baldry}, {Blanton} \& {Eisenstein}}{{Hogg} et~al.}{2002}]{hogg-02} {Hogg} D.~W., {Baldry} I.~K., {Blanton} M.~R., {Eisenstein} D.~J., 2002, arXiv:astro-ph/0210394
1463: \bibitem[\protect\citeauthoryear{{Horan} \& {Finley}}{{Horan} \& {Finley}}{2001}]{2001ICRC....7.2622H} {Horan} D., {Finley} J.~P., 2001, in Proceedings of the 27th International Cosmic Ray Conference, Hamburg, p.~2622
1464: \bibitem[\protect\citeauthoryear{Horan et~al.,}{Horan et~al.}{2002}]{Horan} Horan D., et~al. (Whipple collaboration), 2002, \apj, 571, 753
1465: \bibitem[\protect\citeauthoryear{Horan et~al.,}{Horan et~al.}{2004}]{Horan04} Horan D., et~al. (Whipple collaboration), 2004, \apj, 603, 51
1466: \bibitem[\protect\citeauthoryear{Jester, Harris, Marshall \& Meisenheimer}{Jester et~al.}{2006}]{jester} Jester S., Harris D.~E., Marshall H.~L., Meisenheimer K., 2006, \apj, 648, 900
1467: \bibitem[\protect\citeauthoryear{Kashlinsky}{Kashlinsky}{2005}]{2005PhR...409..361K} Kashlinsky A., 2005, {Phys. Rep.}, 409, 361
1468: \bibitem[\protect\citeauthoryear{Katarzy{\'n}ski, Ghisellini, Tavecchio, Gracia \& Maraschi}{Katarzy{\'n}ski et~al.}{2006a}]{SS3} Katarzy{\'n}ski K., Ghisellini G., Tavecchio F., Gracia J., Maraschi L., 2006a, \mnras, 368, L52
1469: \bibitem[\protect\citeauthoryear{Katarzy{\'n}ski, Ghisellini, Mastichiadis, Tavecchio \& Maraschi}{Katarzy{\'n}ski et~al.}{2006b}]{SS4}Katarzy{\'n}ski K., Ghisellini G., Mastichiadis A., Tavecchio F., Maraschi L., 2006b, \aap, 453, 47
1470: \bibitem[\protect\citeauthoryear{{Kerrick} et~al.,}{{Kerrick} et~al.}{1995}]{1995ApJ...452..588K} {Kerrick} A.~D., et~al. (Whipple collaboration), 1995, \apj, 452, 588
1471: \bibitem[\protect\citeauthoryear{Kneiske, Bretz, Mannheim \& Hartmann}{Kneiske et~al.}{2004}]{kneiske4}Kneiske T.~M., Bretz T., Mannheim K., Hartmann D.~H., 2004, \aap, 413, 807
1472: \bibitem[\protect\citeauthoryear{{Kormendy} \& {Richstone}}{{Kormendy} \& {Richstone}}{1995}]{1995ARAA..33..581K}{Kormendy} J., {Richstone} D., 1995, \araa, 33, 581
1473: \bibitem[\protect\citeauthoryear{Krawczynski}{Krawczynski}{2007}]{veul} Krawczynski H. (VERITAS collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0710.0089)
1474: \bibitem[\protect\citeauthoryear{Krawczynski et~al.,}{Krawczynski et~al.}{2001}]{krawczinski}Krawczynski H., et~al., 2001, \apj, 559, 187
1475: \bibitem[\protect\citeauthoryear{Krawczynski et~al.}{Krawczynski et~al.}{2004}]{Krawczynski1es1959-2004}Krawczynski H., et~al., 2004, \apj, 601, 151
1476: \bibitem[\protect\citeauthoryear{{Krennrich} et~al.,}{{Krennrich} et~al.}{2001}]{2001ApJ...560L..45K} {Krennrich} F., et~al. (Whipple collaboration), 2001, \apjl, 560, L45
1477: \bibitem[\protect\citeauthoryear{Krennrich et~al.,}{Krennrich et~al.}{2002}]{Krennrich2002}Krennrich F., et~al. (Whipple collaboration), 2002, \apj, 575, L9
1478: \bibitem[\protect\citeauthoryear{Kudoh, Aoki, Koide \& Shibata}{Kudoh et~al.}{1999}]{Kudoh} Kudoh T., Aoki S., Koide S., Shibata K., 1999, Astron. Nachr., 320, 311
1479: \bibitem[\protect\citeauthoryear{Laor}{{Laor}}{2000}]{laor2000} {Laor} A., 2000, \apj, 543, L111
1480: \bibitem[\protect\citeauthoryear{{Liu}, {Jiang} \& {Gu}}{{Liu} et~al.}{2006}]{2006ApJ...637..669L} {Liu} Y., {Jiang} D.~R., {Gu} M.~F., 2006, \apj, 637, 669
1481: \bibitem[\protect\citeauthoryear{{Lynden-Bell}}{{Lynden-Bell}}{1969}]{1969Natur.223..690L} {Lynden-Bell} D., 1969, Nature, 223, 690
1482: \bibitem[\protect\citeauthoryear{{Malkov} \& {Drury}}{{Malkov} \& {Drury}}{2001}]{2001RPPh...64..429M}{Malkov} M.~A., {Drury} L.~O., 2001, Rep. Prog. Phys., 64, 429
1483: \bibitem[\protect\citeauthoryear{Mannheim}{Mannheim}{1993}]{MannheimProton}Mannheim K., 1993, \aap, 269, 67
1484: \bibitem[\protect\citeauthoryear{Maraschi, Ghisellini \& Celotti}{Maraschi et~al.}{1992}]{maraschi}Maraschi L., Ghisellini G., Celotti A., 1992, \apjl, 397, L5
1485: \bibitem[\protect\citeauthoryear{Maraschi \& Tavecchio}{Maraschi \& Tavecchio}{2003}]{mt03}Maraschi L., Tavecchio, F., 2003, \apj, 593, 667
1486: \bibitem[\protect\citeauthoryear{Maraschi et~al.,}{Maraschi et~al.}{1999}]{Maraschi60}Maraschi L., et~al., 1999, \apjl, 526, L81
1487: \bibitem[\protect\citeauthoryear{Mastichiadis \& Kirk}{Mastichiadis \& Kirk}{1997}]{MasKirk} Mastichiadis A., Kirk J.~G., 1997, \aap, 320, 19
1488: \bibitem[\protect\citeauthoryear{Mazin}{Mazin}{2003}]{DanielDiplom}Mazin D., 2003, {Diplomarbeit}, Universit\"at Hamburg
1489: \bibitem[\protect\citeauthoryear{{Mazin} \& {Goebel}}{{Mazin} \& {Goebel}}{2007}]{2007ApJ...655L..13M}{Mazin} D., {Goebel} F., 2007, \apj, 655, L13
1490: \bibitem[\protect\citeauthoryear{{McHardy}, {Koerding}, {Knigge}, {Uttley} \& {Fender}}{{McHardy} et~al.}{2006}]{2006Natur.444..730M} {McHardy} I.~M., {Koerding} E., {Knigge} C., {Uttley} P., {Fender} R.~P., 2006, \nat, 444, 730
1491: \bibitem[\protect\citeauthoryear{{McLure} \& {Dunlop}}{{McLure} \& {Dunlop}}{2002}]{McLure}{McLure} R.~J., {Dunlop} J.~S., 2002, \mnras, 331, 795
1492: \bibitem[\protect\citeauthoryear{Melia \& K\"onigl}{Melia \& K\"onigl}{1989}]{mk89}Melia F., K\"onigl A., 1989, \apj, 340, 162
1493: \bibitem[\protect\citeauthoryear{Metcalf \& Magliocchetti}{Metcalf \& Magliocchetti}{2006}]{mm06}Metcalf R.~B., Magliocchetti M., 2006, \mnras, 365, 101
1494: \bibitem[\protect\citeauthoryear{Miller \& Green}{Miller \& Green}{1983}]{MillerGreen}Miller H.~R., Green R.~F., 1983, \baas, 15, 957
1495: \bibitem[\protect\citeauthoryear{{Mirabel}}{{Mirabel}}{2004}]{2004inun.conf..175M} {Mirabel} I.~F., 2004, in {Sch\"onfelder} V., {Lichti} G., {Winkler} C., eds, ESA SP-552: 5th INTEGRAL Workshop on the INTEGRAL Universe, p.~175 (Noordwijk: ESA)
1496: \bibitem[\protect\citeauthoryear{{Mirabel} \& {Rodr{\'{\i}}guez}}{{Mirabel} \& {Rodr{\'{\i}}guez}}{1999}]{1999ARA&A..37..409M}{Mirabel} I.~F., {Rodr{\'{\i}}guez} L.~F., 1999, ARA\&A, 37, 409
1497: \bibitem[\protect\citeauthoryear{Mori}{Mori}{2003}]{Moriicrc}Mori M., 2003, in {Proceedings of the 28th International Cosmic Ray Conference}, Tsukuba, Vol.~8., p.~161
1498: \bibitem[\protect\citeauthoryear{M\"ucke \& Protheroe}{M\"ucke \& Protheroe}{2001}]{MueckeProtheroe}M\"ucke A., Protheroe R.~J., 2001, \aph, 15, 121
1499: \bibitem[\protect\citeauthoryear{Nieppola, Tornikoski \& Valtaoja}{Nieppola et~al.}{2006}]{Nieppola}Nieppola E., Tornikoski M., Valtaoja E., 2006, \aap, 445, 441
1500: \bibitem[\protect\citeauthoryear{Nikishov}{Nikishov}{1962}]{nikishov}Nikishov A.~I., 1962, Sov. Phys. JETP, 14, 393
1501: \bibitem[\protect\citeauthoryear{{Nishijima}}{{Nishijima}}{2002}]{2002PASA...19...26N} {Nishijima} K. (CANGAROO collaboration), 2002, PASA, 19, 26
1502: \bibitem[\protect\citeauthoryear{Nishiyama et~al.,}{Nishiyama et~al.}{1999}]{Nishiyama1999}Nishiyama T., et~al. (Utah Seven Telescope Array collaboration), 1999, in Proceedings of the 26th International Cosmic Ray Conference Vol.~3, Salt Lake City, p.~370
1503: \bibitem[\protect\citeauthoryear{{Piner} \& {Edwards}}{{Piner} \& {Edwards}}{2004}]{2004ApJ...600..115P}{Piner} B.~G., {Edwards} P.~G., 2004, \apj, 600, 115
1504: \bibitem[\protect\citeauthoryear{{Piner}, {Unwin}, {Wehrle}, {Edwards}, {Fey} \& {Kingham}}{{Piner} et~al.}{1999}]{1999ApJ...525..176P}{Piner} B.~G., {Unwin} S.~C., {Wehrle} A.~E., {Edwards} P.~G., {Fey} A.~L., {Kingham} K.~A., 1999, \apj, 525, 176
1505: \bibitem[\protect\citeauthoryear{Piner, Pant \& Edwards}{Piner et~al.}{2007}]{PinerAK}Piner B.~G., Pant N., Edwards P.~G., 2007, \apj, in press, preprint (arXiv:0801.2749)
1506: \bibitem[\protect\citeauthoryear{{Primack}, {Bullock} \& {Somerville}}{{Primack} et~al.}{2005}]{primack421}{Primack} J.~R., {Bullock} J.~S., {Somerville} R.~S., 2005, in {Aharonian} F.~A., {V{\"o}lk} H.~J., {Horns} D., eds, High Energy Gamma-Ray Astronomy, AIP Conf. Ser. 745, pp 23--33
1507: \bibitem[\protect\citeauthoryear{Punch}{Punch}{2007}]{hessmonitoring}Punch M. (H.E.S.S. collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0710.4057, p.~106)
1508: \bibitem[\protect\citeauthoryear{Punch et~al.,}{Punch et~al.}{1992}]{Punch1992}Punch M., et~al. (Whipple collaboration), 1992, \nat, 358, 477
1509: \bibitem[\protect\citeauthoryear{Quinn et~al.,}{Quinn et~al.}{1996}]{Quinn1996}Quinn J., et~al. (Whipple collaboration), 1996, \apj, 456, L83
1510: \bibitem[\protect\citeauthoryear{Raue, Benbow, Costamante \& Horns}{Raue et~al.}{2007}]{0229}Raue M., Benbow W., Costamante L., Horns D. (H.E.S.S. collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0710.4057, p.~134)
1511: \bibitem[\protect\citeauthoryear{{Rebillot}, {Badran}, {Blaylock} et~al.,}{{Rebillot} et~al.}{2006}]{2006ApJ...641..740R}{Rebillot} P.~F., {Badran} H.~M., {Blaylock} G., et~al., 2006, \apj, 641, 740
1512: \bibitem[\protect\citeauthoryear{Rees}{Rees}{1978a}]{ReesM87}Rees M.~J., 1978a, MNRAS, 184, 61\textsc{p}
1513: \bibitem[\protect\citeauthoryear{{Rees}}{{Rees}}{1978b}]{1978Natur.275..516R}{Rees} M.~J., 1978b, \nat, 275, 516
1514: \bibitem[\protect\citeauthoryear{Richstone et~al.,}{Richstone et~al.}{1998}]{Richstone-98}Richstone D., et~al., 1998, \nat, 395, A14
1515: \bibitem[\protect\citeauthoryear{{Sbarufatti}, Treves, Falomo, Heidt, Kotilainen \& Scarpa}{{Sbarufatti} et~al.}{2006}]{Sbar}{Sbarufatti} B., Treves A., Falomo R., Heidt J., Kotilainen J., Scarpa R., 2006, AJ, 132, 1
1516: \bibitem[\protect\citeauthoryear{Schroedter et~al.,}{Schroedter et~al.}{2005}]{Schroedter}Schroedter M., et~al. (Whipple collaboration), 2005, \apj, 634, 947
1517: \bibitem[\protect\citeauthoryear{{Sikora}, {Begelman} \& {Rees}}{{Sikora} et~al.}{1994}]{1994ApJ...421..153S}{Sikora} M., {Begelman} M.~C., {Rees} M.~J., 1994, \apj, 421, 153
1518: \bibitem[\protect\citeauthoryear{{Spada}, {Ghisellini}, {Lazzati} \& {Celotti}}{{Spada} et~al.}{2001}]{Spada} {Spada} M., {Ghisellini} G., {Lazzati} D., {Celotti} A., 2001, \mnras, 325, 1559
1519: \bibitem[\protect\citeauthoryear{Spergel et~al.,}{Spergel et~al.}{2007}]{WMAPpaperpublished} Spergel D.~N., et~al., 2007, \apjs, 170, 377
1520: \bibitem[\protect\citeauthoryear{Stecker, de Jager \& Salamon}{Stecker et~al.}{1992}]{sjs92} Stecker F.~W., de Jager O.~C., Salamon M.~H., 1992, \apj, 390, 49
1521: \bibitem[\protect\citeauthoryear{Stecker, Malkan \& Scully}{Stecker et~al.}{2006}]{steckerNEW} Stecker F.~W., Malkan M.~A., Scully S.~T., 2006, \apj, 648, 774
1522: \bibitem[\protect\citeauthoryear{Stecker, Baring \& Summerlin}{Stecker et~al.}{2007}]{sbs07} Stecker F.~W., Baring, M.~G., Summerlin E.~J., 2007, \apj, 667, L29
1523: \bibitem[\protect\citeauthoryear{Steele et~al.,}{Steele et~al.}{2007}]{veritasmonitoring} Steele D. (VERITAS collaboration), Carini M. T., Charlot P., Kurtanidze O., Lahteenmaki A., Montaruli T., Sadun A. C., Villata M., 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0709.3869)
1524: \bibitem[\protect\citeauthoryear{Superina, Benbow, Boutelier, Dubus \& Giebels}{Superina et~al.}{2007}]{0548icrc} Superina G., Benbow W., Boutelier T., Dubus G., Giebels B. (H.E.S.S. collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0710.4057, p.~138)
1525: \bibitem[\protect\citeauthoryear{{Tavecchio}, {Maraschi} \& {Ghisellini}}{{Tavecchio} et~al.}{1998}]{Tavecchio1998} {Tavecchio} F., {Maraschi} L., {Ghisellini} G., 1998, \apj, 509, 608
1526: \bibitem[\protect\citeauthoryear{Teshima et~al.,}{Teshima et~al.}{2007}]{279} Teshima M., et~al. (MAGIC collaboration), 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0709.1475)
1527: \bibitem[\protect\citeauthoryear{Tluczykont, Shayduk, Kalekin \& Bernardini}{Tluczykont et~al.}{2007}]{tlucmad} Tluczykont M., Shayduk M., Kalekin O., Bernardini E., 2007, J. Phys. Conf. Ser., 60, 318
1528: \bibitem[\protect\citeauthoryear{{Tremaine} et~al.,}{{Tremaine} et~al.}{2002}]{2002ApJ...574..740T} {Tremaine} S., et~al., 2002, \apj, 574, 740
1529: \bibitem[\protect\citeauthoryear{Treves, Falomo \& Uslenghi}{Treves et~al.}{2007}]{treves-07} Treves A., Falomo R., Uslenghi M., 2007, \aap, 473, L17
1530: \bibitem[\protect\citeauthoryear{Uchiyama, Urry, Cheung, Jester, van Duyne, Coppi, Sambruna, Takahashi, Tavecchio \& Maraschi}{Uchiyama et~al.}{2006}]{uchiyama} Uchiyama Y., Urry C.~M., Cheung C.~C., Jester S., van Duyne J., Coppi P., Sambruna R.~M., Takahashi T., Tavecchio F., Maraschi L., 2006, \apj, 648, 910
1531: \bibitem[\protect\citeauthoryear{Urry \& Padovani}{Urry \& Padovani}{1995}]{UrryPadovani} Urry C.~M., Padovani P., 1995, PASP, 107, 803
1532: \bibitem[\protect\citeauthoryear{{Venters} \& {Pavlidou}}{{Venters} \& {Pavlidou}}{2007}]{2007ApJ...666..128V} {Venters} T.~M., {Pavlidou} V., 2007, \apj, 666, 128
1533: \bibitem[\protect\citeauthoryear{{Wagner}, Dorner, Goebel, Hengstebeck, Kranich, Mazin, Nowak \& Tescaro}{{Wagner} et~al.}{2007}]{1553icrc}{Wagner} R.~M., Dorner D., Goebel F., Hengstebeck T., Kranich D., Mazin D., Tescaro D. (MAGIC collaboration), Nowak N., 2007, in {Proceedings of the 30th International Cosmic Ray Conference}, Merida, preprint (arXiv:0711.1586)
1534: \bibitem[\protect\citeauthoryear{{Woo} \& {Urry}}{{Woo} \& {Urry}}{2002a}]{2002ApJ...579..530W}{Woo} J.-H., {Urry} C.~M., 2002a, \apj, 579, 530
1535: \bibitem[\protect\citeauthoryear{{Woo} \& {Urry}}{{Woo} \& {Urry}}{2002b}]{2002ApJ...581L...5W}{Woo} J.-H., {Urry} C.~M., 2002b, \apj, 581, L5
1536: \bibitem[\protect\citeauthoryear{{Woo}, {Urry}, {van der Marel}, {Lira} \& {Maza}}{{Woo} et~al.}{2005}]{2005ApJ...631..762W}{Woo} J.-H., {Urry} C.~M., {van der Marel} R.~P., {Lira} P., {Maza} J., 2005, \apj, 631, 762
1537: \bibitem[\protect\citeauthoryear{{Wu}, {Liu} \& {Zhang}}{{Wu} et~al.}{2002}]{2002AA...389..742W} {Wu} X.-B., {Liu} F.~K., {Zhang} T.~Z., 2002, \aap, 389, 742
1538:
1539: \end{thebibliography}
1540:
1541: \end{document}
1542: