1: \def\nfn{\nu F_{\nu}}
2: \def\nfnsy{\left( \nfn \right)_{\rm sy}}
3: \def\nfnssc{\left( \nfn \right)_{\rm SSC}}
4: \def\nfnerc{\left( \nfn \right)_{\rm ERC}}
5: \def\nfnsyn{F_{{\rm sy}, -10}}
6: \def\nfnsscn{F_{{\rm SSC}, -10}}
7: \def\nfnercn{F_{{\rm ERC}, -10}}
8: \def\esy{\epsilon_{\rm sy}}
9: \def\essc{\epsilon_{\rm SSC}}
10: \def\eerc{\epsilon_{\rm ERC}}
11: \def\estar{\epsilon_{\ast}}
12: \def\esyn{\epsilon_{{\rm sy}, -7}}
13: \def\esscn{\epsilon_{{\rm SSC}, -1}}
14: \def\eercn{\epsilon_{{\rm ERC}, 2}}
15: \def\estarn{\epsilon_{\ast, -5}}
16: \def\eb{\epsilon_B}
17: \def\gcr{\gamma_{\rm cr}}
18: \def\fsp{f_{\rm sp}}
19: \def\fsy{f_{\rm sy}}
20: \def\ferc{f_{\rm ERC}}
21: \def\fssc{f_{\rm SSC}}
22: \def\uext{u_{\rm ext}}
23: \def\Bcr{B_{\rm cr}}
24: \def\taur{\tau_{\rm repr}}
25: \def\ls{\lower4pt\hbox{${\buildrel < \over \sim}$}}
26: \def\gs{\lower4pt\hbox{${\buildrel > \over \sim}$}}
27:
28: \documentclass[12pt, preprint]{aastex}
29: %\documentclass[manuscript]{aastex}
30: %\documentclass[preprint2]{aastex}
31:
32: \newcommand{\vdag}{(v)^\dagger}
33: \newcommand{\myemail}{mboett@helios.phy.ohiou.edu}
34:
35: \slugcomment{Submitted to {\it The Astrophysical Journal}}
36:
37: \shorttitle{The WEBT campaign on 3C~279 in 2006}
38: \shortauthors{B\"ottcher et al.}
39:
40: \begin{document}
41:
42: \title{The WEBT Campaign on the Blazar 3C~279 in 2006\footnote{For
43: questions regarding the availability of the data from the WEBT campaign
44: presented in this paper, please contact the WEBT President Massimo
45: Villata at {\tt villata@oato.inaf.it}}}
46:
47: \author{M. B\"ottcher\altaffilmark{2}, S. Basu\altaffilmark{2},
48: M. Joshi\altaffilmark{2}, M. Villata\altaffilmark{3},
49: A. Arai\altaffilmark{4},
50: N. Aryan\altaffilmark{5},
51: I. M. Asfandiyarov\altaffilmark{6},
52: U. Bach\altaffilmark{3},
53: R. Bachev\altaffilmark{7},
54: A. Berduygin\altaffilmark{8},
55: M. Blaek\altaffilmark{9},
56: C. Buemi\altaffilmark{12},
57: A. J. Castro-Tirado\altaffilmark{11},
58: A. De Ugarte Postigo\altaffilmark{11},
59: A. Frasca\altaffilmark{12},
60: L. Fuhrmann\altaffilmark{30,13,3},
61: V. A. Hagen-Thorn\altaffilmark{17},
62: G. Henson\altaffilmark{14},
63: T. Hovatta\altaffilmark{16},
64: R. Hudec\altaffilmark{9},
65: M. Ibrahimov\altaffilmark{6},
66: Y. Ishii\altaffilmark{4},
67: R. Ivanidze\altaffilmark{15},
68: M. Jel\'inek\altaffilmark{11},
69: M. Kamada\altaffilmark{4},
70: B. Kapanadze\altaffilmark{15}
71: M. Katsuura\altaffilmark{4},
72: D. Kotaka\altaffilmark{4},
73: Y. Y. Kovalev\altaffilmark{30,31},
74: Yu. A. Kovalev\altaffilmark{31},
75: P. Kub\'anek\altaffilmark{9},
76: M. Kurosaki\altaffilmark{4},
77: O. Kurtanidze\altaffilmark{15},
78: A. L\"ahteenm\"aki\altaffilmark{16},
79: L. Lanteri\altaffilmark{3},
80: V. M. Larionov\altaffilmark{17},
81: L. Larionova\altaffilmark{17}
82: C.-U. Lee\altaffilmark{18},
83: P. Leto\altaffilmark{10},
84: E. Lindfors\altaffilmark{8},
85: E. Marilli\altaffilmark{12},
86: K. Marshall\altaffilmark{19},
87: H. R. Miller\altaffilmark{19},
88: M. G. Mingaliev\altaffilmark{32},
89: N. Mirabal\altaffilmark{20},
90: S. Mizoguchi\altaffilmark{4},
91: K. Nakamura\altaffilmark{4},
92: E. Nieppola\altaffilmark{16},
93: M. Nikolashvili\altaffilmark{15},
94: K. Nilsson\altaffilmark{8},
95: S. Nishiyama\altaffilmark{4},
96: J. Ohlert\altaffilmark{21},
97: %A. Oksanen\altaffilmark{22},
98: M. A. Osterman\altaffilmark{19},
99: S. Pak\altaffilmark{23},
100: M. Pasanen\altaffilmark{8},
101: C. S. Peters\altaffilmark{24},
102: T. Pursimo\altaffilmark{25},
103: C. M. Raiteri\altaffilmark{3},
104: J. Robertson\altaffilmark{26},
105: T. Robertson\altaffilmark{27}
106: W. T. Ryle\altaffilmark{19},
107: K. Sadakane\altaffilmark{4},
108: A. Sadun\altaffilmark{5},
109: L. Sigua\altaffilmark{15}
110: B.-W. Sohn\altaffilmark{18},
111: A. Strigachev\altaffilmark{7},
112: N. Sumitomo\altaffilmark{4},
113: L. O. Takalo\altaffilmark{8},
114: Y. Tamesue\altaffilmark{4},
115: K. Tanaka\altaffilmark{4},
116: J. R. Thorstensen\altaffilmark{24},
117: G. Tosti\altaffilmark{13},
118: C. Trigilio\altaffilmark{12},
119: G. Umana\altaffilmark{12},
120: S. Vennes\altaffilmark{26},
121: S. Vitek\altaffilmark{11},
122: A. Volvach\altaffilmark{28},
123: J. Webb\altaffilmark{29}
124: M. Yamanaka\altaffilmark{4},
125: H.-S. Yim\altaffilmark{17},
126: }
127:
128: \altaffiltext{2}{Astrophysical Institute, Department of Physics and Astronomy, \\
129: Clippinger 339, Ohio University, Athens, OH 45701, USA}
130: \altaffiltext{3}{Istituto Nazionale di Astrofisica (INAF), Osservatorio Astronomico di Torino,\\
131: Via Osservatorio 20, I-10025 Pino Torinese, Italy}
132: \altaffiltext{4}{Astronomical Institute, Osaka Kyoiku University, Kashiwara-shi, \\
133: Osaka, 582-8582 Japan}
134: \altaffiltext{5}{Department of Physics, University of Colorado at Denver, \\
135: Campus Box 157, P. O. Box 173364, Denver, CO 80217-3364, USA}
136: \altaffiltext{6}{Ulugh Beg Astronomical Institute, Academy of Sciences of Uzbekistan, \\
137: 33 Astronomical Str., Tashkent 700052, Uzbekistan}
138: \altaffiltext{7}{Institute of Astronomy, Bulgarian Academy of Sciences,\\
139: 72 Tsarigradsko Shosse Blvd., 1784 Sofia, Bulgaria}
140: \altaffiltext{8}{Tuorla Observatory, University of Turku, 21500 Piikki\"o, Finland}
141: \altaffiltext{9}{Astronomical Institute, Academy of Sciences of the Czech Republic,\\
142: CZ-251 65 Ondrejov, Czech Republic}
143: \altaffiltext{10}{Istituto di Radioastronomia, Sezione di Noto, C. da Renna Bassa -- \\
144: Loc. Casa di Mezzo C. P. 141, I-96017 Noto, Italy}
145: \altaffiltext{11}{Instituto de Astrofisica de Andalucia, Apartado de Correos, 3004, \
146: E-18080 Granada, Spain}
147: \altaffiltext{12}{Osservatorio Astrofisico di Catania, Viale A.\ Doria 6, \\
148: I-95125 Catania, Italy}
149: \altaffiltext{13}{Osservatorio Astronomico, Universit\`a di Perugia, Via B.\ Bonfigli, \\
150: I-06126 Perugia, Italy}
151: \altaffiltext{14}{East Tennessee State University and SARA Observatory, \\
152: Department of Physics, Astronomy, and Geology, Box 70652, Johnson City, TN 37614}
153: \altaffiltext{15}{Abastumani Observatory, 383762 Abastumani, Georgia}
154: \altaffiltext{16}{Mets\"ahovi Radio Observatory, Helsinki University of Technology, \\
155: Mets\"ahovintie 114, 02540 Kylm\"al\"a, Finland}
156: \altaffiltext{17}{Astronomical Institute, St. Petersburg State University, \\
157: Universitetsky pr.\ 28, Petrodvoretz, 198504 St. Petersburg, Russia}
158: \altaffiltext{18}{Korea Astronomy \& Space Science Institute, 61-1 Whaam-Dong, Yuseong-Gu,\\
159: Daejeon 305-348, Korea}
160: \altaffiltext{19}{Department of Physics and Astronomy, Georgia State University,\\
161: Atlanta, GA 30303, USA}
162: \altaffiltext{20}{Department of Astronomy, University of Michigan, \\
163: 830 Dennison Building, Ann Arbor, MI 48109-1090, USA}
164: \altaffiltext{21}{Michael Adrian Observatory, Astronomie-Stiftung Trebur, \\
165: Fichtenstraße 7, D-65468 Trebur, Germany}
166: %\altaffiltext{22}{Nyr\"ol\"a Observatory, Jyv\"askyl\"an Sirius ry, Kyllikinkatu 1, \\
167: %40950 Jyv\"askyl\"a, Finland}
168: \altaffiltext{23}{Department of Astronomy and Space Science, Kyung Hee University, \\
169: Seocheon, Gilheung, Yongin, Gyeonggi, 446-701, South Korea}
170: \altaffiltext{24}{Department of Physics and Astronomy, Dartmouth College, MS 6127, \\
171: Hannover, NH 03755, USA}
172: \altaffiltext{25}{Nordic Optical Telescope, Apartado 474, E-38700 Santa Cruz de La Palma,\\
173: Santa Cruz de Tenerife, Spain}
174: \altaffiltext{26}{Florida Institute of Technology and SARA Observatory, 150 West University Boulevard, \\
175: Melbourne, FL 32901-6975, USA}
176: \altaffiltext{27}{Ball State University and SARA Observatory, Department of Physics and Astronomy, \\
177: Munice, IN 47306, USA}
178: \altaffiltext{28}{Crimean Astrophysical Observatory, Nauchny, Crimea 98409, Ukraine}
179: \altaffiltext{29}{Florida International University and SARA Observatory,\\
180: University Park Campus, Miami, FL 33199, USA}
181: \altaffiltext{30}{Max-Planck-Institut f\"ur Radioastronomie, Auf dem H\"ugel 69,\\
182: D-53121 Bonn, Germany}
183: \altaffiltext{31}{Astro Space Center of Lebedev Physical Institute, Profsoyuznaya 84/32,\\
184: Moscow 117997, Russia}
185: \altaffiltext{32}{Special Astrophysical Observatory, Nizhnij Arkhyz, Karachai-Cherkessia 369167, Russia}
186:
187:
188: \begin{abstract}
189: The quasar 3C~279 was the target of an extensive multiwavelength
190: monitoring campaign from January through April 2006. An optical-IR-radio
191: monitoring campaign by the Whole Earth Blazar Telescope (WEBT)
192: collaboration was organized around Target of Opportunity X-ray
193: and soft $\gamma$-ray observations with {\it Chandra} and {\it INTEGRAL}
194: in mid-January 2006, with additional X-ray coverage by {\it RXTE} and
195: {\it Swift} XRT. In this paper we focus on the results of the WEBT
196: campaign.
197:
198: The source exhibited substantial variability of optical flux and spectral
199: shape, with a characteristic time scale of a few days. The variability
200: patterns throughout the optical BVRI bands were very closely correlated
201: with each other, while there was no obvious correlation between the
202: optical and radio variability. After the ToO trigger, the optical flux
203: underwent a remarkably clean quasi-exponential decay by about one magnitude,
204: with a decay time scale of $\tau_d \sim 12.8$~d.
205:
206: In intriguing contrast to other (in particular, BL~Lac type) blazars, we
207: find a lag of shorter-wavelength behind longer-wavelength variability throughout
208: the RVB wavelength ranges, with a time delay increasing with increasing frequency.
209: Spectral hardening during flares appears delayed with respect to a rising
210: optical flux. This, in combination with the very steep IR-optical continuum
211: spectral index of $\alpha_o \sim 1.5$ -- 2.0, may indicate a highly oblique
212: magnetic field configuration near the base of the jet, leading to inefficient
213: particle acceleration and a very steep electron injection spectrum.
214:
215: An alternative explanation through a slow (time scale of several days)
216: acceleration mechanism would require an unusually low magnetic field of
217: $B \lesssim 0.2$~G, about an order of magnitude lower than inferred from
218: previous analyses of simultaneous SEDs of 3C~279 and other FSRQs with
219: similar properties.
220:
221: \end{abstract}
222:
223: \keywords{galaxies: active --- Quasars: individual (3C~279)
224: --- gamma-rays: theory --- radiation mechanisms: non-thermal}
225:
226: \section{Introduction}
227:
228: Flat-spectrum radio quasars (FSRQs) and BL~Lac objects are
229: active galactic nuclei (AGNs) commonly unified in the class
230: of blazars. They exhibit some of the most violent high-energy
231: phenomena observed in AGNs to date. Their spectral energy
232: distributions (SEDs) are characterized by non-thermal continuum
233: spectra with a broad low-frequency component in the radio -- UV
234: or X-ray frequency range and a high-frequency component from
235: X-rays to $\gamma$-rays. Their electromagnetic radiation exhibits
236: a high degree of linear polarization in the optical and radio
237: bands and rapid variability at all wavelengths. Radio interferometric
238: observations often reveal radio jets with individual components
239: exhibiting apparent superluminal motion. At least episodically,
240: a significant portion of the bolometric flux is emitted in
241: $> 100$~MeV $\gamma$-rays. 46 blazars have been detected
242: and identified with high confidence in high energy ($> 100$~MeV)
243: $\gamma$-rays by the {\it Energetic Gamma-Ray Experiment Telescope
244: (EGRET)} instrument on board the {\it Compton Gamma-Ray Observatory}
245: \citep[CGRO,][]{hartman99,mhr01}.
246:
247: In the framework of relativistic jet models, the low-frequency (radio
248: -- optical/UV) emission from blazars is interpreted as synchrotron
249: emission from nonthermal electrons in a relativistic jet. The
250: high-frequency (X-ray -- $\gamma$-ray) emission could either be
251: produced via Compton upscattering of low frequency radiation by the
252: same electrons responsible for the synchrotron emission \citep[leptonic
253: jet models; for a recent review see, e.g.,][]{boettcher07a}, or
254: due to hadronic processes initiated by relativistic protons
255: co-accelerated with the electrons \citep[hadronic models, for
256: a recent discussion see, e.g.,][]{muecke01,muecke03}.
257:
258: The quasar 3C279 ($z = 0.538$) is one of the best-observed flat
259: spectrum radio quasars, not at last because of its prominent
260: $\gamma$-ray flare shortly after the launch of {\it CGRO}
261: in 1991. It has been persistently detected by {\it EGRET}
262: each time it was observed, even in its very low quiescent
263: states, e.g., in the winter of 1992 -- 1993, and is
264: known to vary in $\gamma$-ray flux by roughly two orders of
265: magnitude \citep{maraschi94,wehrle98}. It has been monitored
266: intensively at radio, optical, and more recently also X-ray
267: frequencies, and has been the subject of intensive multiwavelength
268: campaigns \citep[e.g.,][]{maraschi94,hartman96,wehrle98}.
269:
270: Also at optical wavelengths, 3C~279 has exhibited substantial
271: variability over up to two orders of magnitude ($R \sim
272: 12.5$ -- 17.5). Variability has been observed on a
273: variety of different time scales, from years, down to
274: intra-day time scales. The most extreme variability patterns
275: include intraday variability with flux decays of $\lesssim
276: 0.1^{\rm mag}$/hr \citep{kb07}. Observations with the {\it
277: International Ultraviolet Explorer} in the very low activity
278: state of the source in December 1992 -- January 1993 revealed
279: the existence of a thermal emission component, possibly related
280: to an accretion disk, with a luminosity of $L_{\rm UV} \sim
281: 2 \times 10^{46}$~erg~s$^{-1}$ if this component is assumed
282: to be emitting isotropically \citep{pian99}. \cite{pian99} have
283: also identified an X-ray spectral variability trend in archival
284: {\it ROSAT} data, indicating a lag of $\sim 2$ -- 3~days of the
285: soft X-ray spectral hardening behind a flux increase. Weak evidence
286: for spectral variability was also found within the {\it EGRET}
287: (MeV -- GeV) energy range \citep{nandi07}. At low $\gamma$-ray flux
288: levels, an increasing flux seems to be accompanied by a spectral
289: softening, while at high flux levels, no consistent trend was
290: apparent.
291:
292: The quasar 3C~279 was the first object in which superluminal motion was
293: discovered \citep{whitney71,cotton79,unwin89}. Characteristic apparent
294: speeds of individual radio components range up to $\beta_{\rm app}
295: \sim 17$ \citep{cotton79,homan03,jorstad04}, indicating pattern flow
296: speeds with bulk Lorentz factors of up to $\Gamma \sim 17$. Radio jet
297: components have occasionally been observed not to follow straight,
298: ballistic trajectories, but to undergo slight changes in direction
299: between parsec- and kiloparsec-scales \citep{homan03,jorstad04}.
300: VLBA polarimetry indicates that the electric field vector is
301: generally well aligned with the jet direction on pc to kpc scales
302: \citep{jorstad04,ojha04,lh05,helmboldt07}, indicating that the magnetic field
303: might be predominantly perpendicular to the jet on those length scales.
304:
305: A complete compilation and modeling of all available SEDs simultaneous
306: with the 11 {\it EGRET} observing epochs has been presented in \cite{hartman01a}.
307: The modeling was done using the time-dependent leptonic (SSC + EIC)
308: model of \cite{bms97,bb00} and yielded quite satisfactory fits for
309: all epochs. The results were consistent with other model fitting works
310: \citep[e.g.,][]{bednarek98,sikora01,moderski03} concluding that the
311: X-ray -- soft $\gamma$-ray portion of the SED might be dominated by
312: SSC emission, while the {\it EGRET} emission might require an additional,
313: most likely external-Compton, component. The resulting best-fit
314: parameters were consistent with an increasing bulk Lorentz factor,
315: but decreasing Lorentz factors of the ultrarelativistic electron
316: distribution in the co-moving frame of the emission region during
317: $\gamma$-ray high states, as compared to lower $\gamma$-ray states
318: \citep{hartman01a}. However, such an interpretation also required
319: changes of the overall density of electrons, and the spectral index
320: of the injected electron power-law distribution, which did not show
321: any consistent trend with $\gamma$-ray luminosity.
322:
323: \cite{hartman01b} have investigated cross correlations between
324: different wavelength ranges, in particular, between optical,
325: X-ray, and $\gamma$-ray variability. In that work, a general
326: picture of a positive correlation between optical, X-ray and
327: $\gamma$-ray activity emerged, but no consistent trends of time
328: lags between the different wavelength ranges were found.
329:
330: The discussion above illustrates that, in spite of the intensive
331: past observational efforts, the physics driving the broadband
332: spectral variability properties of 3C~279 are still rather poorly
333: understood. For this reason, \cite{collmar07b} proposed an
334: intensive multiwavelength campaign in an optical high state of
335: 3C~279, in order to investigate its correlated radio -- IR
336: -- optical -- X-ray -- soft $\gamma$-ray variability. The campaign
337: was triggered on Jan. 5, 2006, when the source exceeded an R-band
338: flux corresponding to R = 14.5. It involved intensive radio, near-IR
339: (JHK), and optical monitoring by the Whole Earth Blazar Telescope
340: \cite[WEBT\footnote{\tt http://www.to.astro.it/blazars/webt}, see,
341: e.g.][and references therein]{raiteri06,villata07} collaboration
342: through April of 2006, focusing on a core period of Jan. and Feb.
343: 2006. In order to illustrate the source's behaviour leading up to
344: the trigger in January 2006, previously unpublished radio and optical
345: data from late 2005 are also included in the analysis presented in this
346: paper. X-ray and soft $\gamma$ observations were carried out by all
347: instruments on board the {\it International Gamma-Ray Astrophysics
348: Laboratory (INTEGRAL)} during the period of Jan. 13 -- 20, 2006.
349: Additional, simultaneous X-ray coverage was obtained by {\it
350: Chandra} and {\it Swift} XRT. These observations were supplemented
351: by extended X-ray monitoring with the {\it Rossi X-Ray Timing
352: Explorer (RXTE)}. In this paper, we present details of the
353: data collection, analysis, and results of the WEBT (radio -- IR
354: -- optical) campaign. Preliminary results of the multiwavelength
355: campaign have been presented in \cite{collmar07a} and
356: \cite{boettcher07a,boettcher07b}, and a final, comprehensive
357: report on the result of the entire multiwavelength campaign
358: will appear in \cite{collmar07b}.
359:
360: Throughout this paper, we refer to $\alpha$ as the energy
361: spectral index, $F_{\nu}$~[Jy]~$\propto \nu^{-\alpha}$. A
362: cosmology with $\Omega_m = 0.3$, $\Omega_{\Lambda} = 0.7$,
363: and $H_0 = 70$~km~s$^{-1}$~Mpc$^{-1}$ is used. In this cosmology,
364: and using the redshift of $z = 0.538$, the luminosity distance
365: of 3C~279 is $d_L = 3.1$~Gpc.
366:
367: \begin{figure}
368: \plotone{f1.eps}
369: \caption{Timeline of the broadband campaign on 3C~279 in 2006,
370: including the optical and radio light curves during the entire
371: campaign period. The gray shaded area indicates the period of the
372: {\it INTEGRAL, Chandra}, and {\it Swift} observations.}
373: \label{timeline}
374: \end{figure}
375:
376: \section{\label{observations}Observations, data reduction,
377: and light curves}
378:
379: 3C~279 was observed in a coordinated multiwavelength campaign
380: at radio, near-IR, optical (by the WEBT collaboration),
381: X-ray ({\it Chandra}, {\it Swift}, {\it RXTE PCA}, {\it INTEGRAL
382: JEM-X}), and soft $\gamma$-ray ({\it INTEGRAL}) energies. The overall
383: timeline of the campaign, along with the measured long-term light
384: curves at radio and optical frequencies is illustrated in Fig.
385: \ref{timeline}. Simultaneous X-ray coverage with all X-ray / soft
386: $\gamma$-ray telescopes mentioned above was obtained in the time
387: frame Jan. 13 -- 20, as indicated by the gray shaded area in Fig.
388: \ref{timeline}. Detailed results of those high-energy observations
389: will be presented in \cite{collmar07b}. Table \ref{observatories}
390: lists all participating observatories which contributed data to
391: the WEBT campaign. In total, 25 ground-based radio, infrared,
392: and optical telescopes in 12 countries on 4 continents contributed
393: 2173 data points.
394:
395: \begin{deluxetable}{lccc}
396: \tabletypesize{\scriptsize}
397: \tablecaption{List of observatories that contributed data to the
398: WEBT campaign}
399: \tablewidth{0pt}
400: \tablehead{
401: \colhead{Observatory} & \colhead{Specifications} & \colhead{frequency / filters} &
402: \colhead{$N_{\rm obs}$}
403: }
404: \startdata
405: \multispan 4 \hss \bf Radio \hss \\
406: \noalign{\smallskip\hrule\smallskip}
407: Mets\"ahovi, Finland & 14 m & 37 GHz & 70 \\
408: Medicina, Italy & 32 m & 5, 8, 22 GHz & 32 \\
409: Noto, Italy & 32 m & 8, 22 GHz & 6 \\
410: RATAN-600, Russia & 576 m (ring) & 1, 2.3, 5, 8, 11, 22 GHz & 138 \\
411: Crimean Astr. Obs., Ukraine (RT-22) & 22 m & 36 GHz & 7 \\
412: %UMRAO, Michigan, USA & 26 m & 4.8, 8, 14.5 GHz & 46 \\
413: \noalign{\smallskip\hrule\smallskip}
414: \multispan 4 \hss \bf Infrared \hss \\
415: \noalign{\smallskip\hrule\smallskip}
416: Roque (NOT), Canary Islands & 2.56 m & J, H, K & 3 \\
417: \noalign{\smallskip\hrule\smallskip}
418: \multispan 4 \hss \bf Optical \hss \\
419: \noalign{\smallskip\hrule\smallskip}
420: Abastumani, Georgia (FSU) & 70 cm & R & 127 \\
421: ARIES, Naintal, India & & R & 63 \\
422: %Armenzano, Italy & 40 cm & B, V, R, I & 315 \\
423: %Bell Obs., Kentucky, USA & 60 cm & R & 12 \\
424: Belogradchik, Bulgaria & 60 cm & V, R, I & 75 \\
425: BOOTES-1, Spain & 30 cm & R & 151 \\
426: Catania, Italy & 91 cm & U, B, V & 33 \\
427: Crimean Astr. Obs., Ukraine & 70 cm & B, V, R, I & 47 \\
428: SMARTS, CTIO, Chile & 90 cm & B, V, R & 33 \\
429: %Heidelberg, Germany & 70 cm & B, R, I & 8 \\
430: Kitt Peak (MDM), Arizona, USA & 130 cm & U, B, V, R, I & 190 \\
431: Kitt Peak (MDM), Arizona, USA & 240 cm & R & 77 \\
432: Michael Adrian Obs., Germany & 120 cm & R & 9 \\
433: Mt. Lemmon, Arizona, USA & 100 cm & B, V, R, I & 214 \\
434: Mt. Maidanak (AZT-22), Uzbekistan & 150 cm & B, V, R, I & 44 \\
435: %Nyr\"ol\"a, Finland & 40 cm SCT & R & 159 \\
436: Osaka Kyoiku, Japan & 51 cm & V, R, I & 494 \\
437: %Perugia, Italy & 40 cm & V, R, I & 140 \\
438: Roque (KVA), Canary Islands & 35 cm & R & 75 \\
439: Roque (NOT), Canary Islands & 256 cm & U, B, V, R, I & 7 \\
440: %Sabadell, Spain & 50 cm & B, R & 4 \\
441: %San Pedro Mart\'\i r, Mexico & 150 cm & B, V, R, I & 185 \\
442: SARA, Arizona, USA & 90 cm & B, V, R, I & 242 \\
443: %Shanghai, China & 156 cm & V, R & 36 \\
444: %Skinakas, Crete & 130 cm & B, V, R, I & 156 \\
445: %Sobaeksan, Korea & 61 cm & B, V, R, I & 133 \\
446: %St. Louis, Missouri, USA & 35 cm & B, R & 16 \\
447: Tenagra, Arizona, USA & 81 cm & B, V, R, I & 19 \\
448: Torino, Italy & 105 cm & B, V, R & 3 \\
449: Tuorla, Finland & 103 cm & R & 84 \\
450: \noalign{\smallskip\hrule}
451: \enddata
452: \label{observatories}
453: \end{deluxetable}
454:
455:
456: \subsection{\label{optical}Optical and near-infrared observations}
457:
458: The observing strategy and data analysis followed to a large extent
459: the standard procedure for the optical data reduction for WEBT campaigns
460: which is briefly outlined below. For more information on standard data
461: reduction procedures for WEBT campaigns see also:
462: \cite{villata00,raiteri01,villata02,boettcher03,villata04a,villata04b,raiteri05,boettcher05}
463:
464: It had been suggested that, optimally, observers perform photometric
465: observations alternately in the B and R bands, and include complete
466: (U)BVRI sequences at the beginning and the end of each observing
467: run. Exposure times should be chosen to obtain an optimal compromise
468: between high precision (instrumental errors less than $\sim 0.03$~mag
469: for small telescopes and $\sim 0.01$~mag for larger ones) and high time
470: resolution. If this precision requirement leads to gaps of 15 -- 20
471: minutes in each light curve, we suggested to carry out observations
472: in the R band only. Observers were asked to perform bias and dark
473: corrections as well as flat-fielding on their frames, and obtain
474: instrumental magnitudes, applying either aperture photometry (using
475: IRAF or CCDPHOT) or Gaussian fitting for the source 3C~279 and four
476: recommended comparison stars. This
477: calibration has then been used to convert instrumental to standard
478: photometric magnitudes for each data set. In the next step, unreliable
479: data points (with large error bars at times when higher-quality data
480: points were available) were discarded. Our data did not provide evidence
481: for significant variability on sub-hour time scales. Consequently, error
482: bars on individual data sets could be further reduced by re-binning
483: on time scales of typically 15 -- 20~min.
484: %In previous WEBT campaigns, systematic offsets between different
485: %instruments and telescopes had sometimes been found, which were
486: %then corrected on a night-by-night basis, wherever the available
487: %data sets contained sufficient independent measurements to clearly
488: %identify such offsets \citep[e.g.,][]{villata02}. However, in the
489: %case of our campaign data, the available data did not allow the
490: %unambiguous identification of significant offsets between individual
491: %instruments, so no offset correction was applied to our data sets.
492: The data resulting at this stage of the analysis are displayed in
493: the top panel of Fig. \ref{timeline}.
494:
495: In order to provide information on the intrinsic broadband
496: spectral shape (and, in particular, a reliable extraction
497: of B - R color indices), the data were then de-reddened using
498: the Galactic Extinction coefficients of \cite{schlegel98},
499: based on $A_B = 0.123$~mag and $E(B-V) =
500: 0.029$~mag\footnote{\tt http://nedwww.ipac.caltech.edu/}.
501:
502: Possible contaminations of the optical color information could
503: generally also arise from contributions from the host galaxy
504: and the optical -- UV emission from an accretion disk around
505: the central supermassive black hole in 3C~279. However, these
506: contributions are not expected to be significant in the case
507: of our campaign data: Assuming absolute magnitudes of $M_V
508: \sim -23$ and $M_B \sim -21$ for typical quasar host galaxies
509: at $z \sim 0.5$ \citep[e.g.][]{floyd04,zakamska06}, their
510: contribution in the V and B band, respectively, at the distance
511: of 3C~279 would be $V_{\rm gal} \sim 19.5$ and $B_{\rm gal}
512: \sim 21.6$, respectively. These are at least about four magnitudes
513: fainter than the actually measured total B and V magnitudes
514: during our campaign, and thus negligible. The possible
515: contribution of an accretion disk can be estimated on the
516: basis of the thermal component for which \cite{pian99} found
517: evidence in IUE observations during the 1992/1993 low state
518: of 3C~279. Their best fit to this component suggests $U \sim
519: 18.6$ and $B \sim 21$ (and much fainter contributions at lower
520: frequencies), which corresponds to a contribution of $\lesssim
521: 2.5$~\% to the total B and U magnitudes measured during our
522: campaign. Therefore, both the host galaxy and the accretion
523: disk contribution are neglected in our further analysis.
524:
525: The only infrared observations obtained for this campaign were
526: one sequence of JHK exposures taken on January 15, 2006, with
527: the 2.56~m NOT on Roque de los Muchachos on the Canary Island
528: of La Palma. The resulting fluxes are included in the SED
529: displayed in Fig. \ref{SED}.
530:
531:
532: \subsubsection{\label{opt_lightcurves}Optical light curves}
533:
534: The optical (and radio) light curves from December 2005 to April
535: 2006 are displayed in Fig. \ref{timeline}. The densest coverage
536: was obtained in the R band, and the figure clearly indicates
537: that the variability in B, V, and I bands closely tracks the
538: R-band behaviour. The coverage in the U band was extremely sparse
539: and does not allow any assessment of the U-band light curve during
540: our campaign. Therefore, the U-band will be ignored in the following
541: discussion, and we will describe the main features of the variability
542: behavior based on the R-band light curve.
543:
544: The optical light curves show variability with magnitude changes of
545: typically $\lesssim 0.5^{\rm mag}$ on time scales of a few days. The
546: most notable exception to this relatively moderate variability is the
547: major dip of the brightnesses in all optical bands right around our
548: coordinated X-ray / soft $\gamma$-ray observations in January 2006
549: ($\approx$~JD~2453742 -- 2453770). In the R-band, the light curve followed
550: an unusually clean exponential decay over 1.1 mags. in 13 days, i.e.,
551: a slope of $dR/dt = 0.085$~mag/day or a flux decay as $F(t) = F(t_0)
552: \, e^{-(t - t_0) / \tau_d}$ with a decay constant of $\tau_d = 12.8$~d.
553: Only moderate intraday deviations on a characteristic scale of
554: $\lesssim 0.1$~mag/d are superposed on this smooth exponential
555: decay.
556:
557: In contrast to the smooth decline of the optical brightness during
558: January 6 -- 20, 2006, the subsequent re-brightening to levels
559: comparable to those before the dip, appears much more erratic
560: and involves a remarkably fast rise by
561: $\sim 0.5^{\rm mag}$ within $\sim 1$~d on Jan. 27 (JD 2453763).
562: Unfortunately, the detailed shape of this fast rise was not well
563: sampled in our data set. If this was a quasi-exponential rise with
564: a slope of $dR/dt \sim 0.5$~mag/d, it would correspond to a rise
565: time scale of $\tau_r \sim 2.2$~d.
566:
567:
568: \subsection{\label{radio}Radio observations}
569:
570: At radio frequencies, 3C~279 was monitored using the 14~m Mets\"ahovi
571: Radio Telescope of the Helsinki University of Technology, at 37~GHz,
572: the 32-m radio telescope of the Medicina Radio Observatory near Bologna,
573: Italy, at 5, 8, and 22~GHz, the 32-m antenna of the Noto Radio Observatory
574: on Sicily, Italy, at 8 and 22~GHz, the 576-m ring telescope (RATAN-600)
575: of the Russian Academy of Sciences, at 1, 2.3, 5, 8, 11, and 22~GHz,
576: and the 22-m RT-22 dish at the Crimean Astrophysical Observatory,
577: Ukraine, at 36~GHz.
578:
579: The Mets\"ahovi data have been reduced with the standard procedure
580: described in \cite{ter98}. The resulting 37~GHz light curve is reasonably
581: well sampled during the period mid-January -- mid-March 2006. Inspection by
582: eye in comparison to the optical light curves displayed in Fig. \ref{timeline}
583: appears to indicate that the optical and 37~GHz light curves follow similar
584: variability patterns with a radio lead before the optical variability
585: by $\sim 5$~days. However, a discrete cross-correlation analysis between
586: the R-band and 37~GHz radio light curves did not reveal a significant
587: signal to confirm this suggestion.
588:
589: Also included in Fig. \ref{timeline} are the radio light curves at 5, 8,
590: and 22~GHz. For details of the analysis of data from the Medicina and
591: Noto radio observatories at those frequencies, see \cite{bach07}. As
592: already apparent in Fig. \ref{timeline}, most of the data at frequencies
593: below 37~GHz were not well sampled on the $\lesssim 3$-months time scale
594: of the 2006 campaign, and any evidence for variability did not show a
595: discernable correlation with the variability at higher (radio and optical)
596: frequencies.
597:
598: All radio data contributed to this campaign are stored in the
599: WEBT archive (see {\tt http://www.to.astro.it/blazars/webt/}
600: for information regarding availability of the data).
601: Radio data at all observed frequencies have been included in the
602: quasi-simultaneous SED for Jan 15, 2006, shown in Fig. \ref{SED}.
603: Given the generally very moderate radio variability at frequencies
604: below 37~GHz, a linear interpolation between the two available data
605: points nearest in time to Jan. 15, 2006, was used to construct an
606: estimate of the actual radio fluxes at that time.
607:
608:
609: \begin{figure}
610: \plotone{f2.eps}
611: \caption{Light curve of the B and R magnitudes and the B -- R color index
612: of 3C~279 over the duration of the entire campaign. }
613: \label{B_R_lc}
614: \end{figure}
615:
616: \section{\label{variability}Optical spectral variability}
617:
618: In this section, we will describe spectral variability
619: phenomena, i.e. the variability of spectral (and color)
620: indices and their correlations with monochromatic source fluxes.
621: We will concentrate here on the optical spectral variability as
622: indicated by a change of the optical color. In particular, our
623: observing strategy was optimized to obtain a good sampling of
624: the B -- R color index as a function of time. Since our data
625: did generally not indicate substantial flux changes on sub-hour
626: time scales, we extracted B -- R color indices wherever both
627: magnitudes were available within 20 minutes of each other.
628: Fig. \ref{B_R_lc} shows the B -- R color history over the entire
629: campaign, compared to the B and R band light curves. Overall,
630: there is no obvious correlation between the light curves and
631: the color behavior of the source on long time scales. However,
632: two short-term sequences attracted our attention: There is a
633: sequence of brightness decline accompanied by a spectral hardening
634: (declining B -- R index) around JD 2453750 (Jan. 14, 2006), and
635: another sequence of a brightness increase accompanied by a
636: spectral softening around JD 2453790 -- 2453806 (Feb. 23 --
637: Mar. 11). However, we caution that incomplete sampling, in
638: particular in the B-band may introduce spurious effects.
639:
640: \begin{figure}
641: \plotone{f3.eps}
642: \caption{Snap-shot optical (BVRI) continuum spectra for 5 epochs
643: during our campaign. All measurements for each individual spectrum
644: were taken within $\le 20$~min. of each other. }
645: \label{opt_sp}
646: \end{figure}
647:
648: A similar trend was recently observed in a multiwavelength WEBT campaign
649: on the quasar 3C~454.3 \citep{villata06}, where it could be interpreted as
650: a ``little blue bump'' due to the unresolved contribution from optical emission
651: lines in the $\sim 2000$ -- 4000~\AA\ wavelength range in the rest frame
652: of the quasar, in particular from Fe~II and Mg~II \citep{raiteri07}.
653: In order to test whether such an interpretation would also be viable in
654: the case of 3C~279, we have compiled several simultaneous snap-shot optical
655: continuum (BVRI) spectra at various brightness levels (Fig. \ref{opt_sp}).
656: All measurements for each individual spectrum displayed in Fig. \ref{opt_sp}
657: have been taken within $\le 20$~min. of each other. Only the spectrum of
658: Jan. 15 shows a significant deviation from a pure power-law in the B
659: band; and in that case, there was simultaneous U band coverage, which
660: matched a straight power-law extrapolation of the VRIJHK spectrum.
661: If the spectral upturn towards the blue end of the spectrum were
662: due to an unresolved Mg~II / Fe~II line contribution, it should
663: emerge even more clearly in the Jan. 24 spectrum, which is characterized
664: by a lower optical continuum flux level than the Jan. 15 spectrum.
665: Therefore, the compilation of spectra in Fig. \ref{opt_sp} does not
666: provide any support for the existence of an essentially non-variable
667: continuum component at the blue end of the spectrum. For this reason,
668: we believe that the color changes that we found in our data are in fact
669: intrinsic to the blazar jet emission.
670:
671: Fig. \ref{color_mag} illustrates our impression from Fig. \ref{B_R_lc},
672: that there is no clear overall trend of source (R-band) brightness with
673: optical spectral hardness. However, the figure clearly illustrates that
674: the scatter of the B -- R color index is significantly larger at larger
675: source brightness, indicating that spectral variability is more likely
676: to occur when the source is bright. Specifically, for $R > 14.5$, the
677: data is consistent with a roughly constant value of B -- R = 1.09,
678: corresponding to a spectral index of $\alpha_o = 1.9$ for a power-law
679: continuum spectrum with $F_{\nu, o} [{\rm Jy}] \propto \nu^{-\alpha_o}$.
680: At brightness levels $R < 14.5$, spectral variability by $\Delta (B - R)
681: \lesssim 0.35$, corresponding to $\Delta\alpha_o \lesssim 0.85$, is
682: observed.
683:
684: \begin{figure}
685: \plotone{f4.eps}
686: \caption{Color-magnitude diagram for the entire campaign period.}
687: \label{color_mag}
688: \end{figure}
689:
690: In Fig. \ref{hysteresis}, we focus on two $\sim 1$-month optical flares,
691: around mid-Jan. -- mid-Feb. 2006, and late Feb. -- late March 2006,
692: as indicated by the gray shaded segments 1 and 2, respectively, in
693: Fig. \ref{B_R_lc}. The number labels in the color-magnitude diagrams
694: in Fig. \ref{hysteresis} indicate the time ordering of the points.
695: While there is no obvious trend discernible in Segment 1, Segment 2
696: suggests the presence of a spectral hysteresis pattern: The spectral
697: softening around JD 2453790 -- 2453806 (Feb. 23 -- Mar. 11), already
698: mentioned above, precedes the main brightness increase. Subsequently,
699: the optical continuum hardens while the source is still in a bright
700: state. We need to caution that due to the poor sampling of the B band
701: light curve during segment 2, the significance of the tentative hysteresis
702: found here may be questionable. However, such hysteresis would naturally
703: lead to a B-band time lag behind the R-band, for which we do find a
704: $3.9 \, \sigma$ evidence from a discrete correlation function analysis
705: as described in the following section.
706:
707: \begin{figure}
708: \plotone{f5.eps}
709: \caption{Color-magnitude diagrams for the two time segments marked in
710: Fig. \ref{B_R_lc}, with time ordering indicated by the numbers in the
711: two panels. }
712: \label{hysteresis}
713: \end{figure}
714:
715: To our knowledge, such a spectral hysteresis has never been observed
716: at optical wavelengths for any flat-spectrum radio quasar. It is
717: reminiscent of the spectral hysteresis occasionally seen at X-ray
718: energies in high-frequency peaked BL~Lac objects \citep[HBLs,
719: e.g.][]{takahashi96,fossati00,kataoka00}. However, the spectral
720: hysteresis observed in the X-rays of HBLs is generally clockwise
721: (i.e., spectral hardening precedes flux rise; softening precedes
722: flux decline), and can be interpreted as the synchrotron signature
723: of fast acceleration of ultrarelativistic electrons, followed by
724: a gradual decline on the radiative cooling time scale
725: \citep[e.g.,][]{kataoka00,kusunose00,lk00,bc02}. In our
726: case, the direction of the spectral hysteresis is counterclockwise
727: (i.e., spectral softening precedes the flux rise; spectral hardening
728: precedes flux decline). Possible physical implications of such
729: hysteresis phenomena will be discussed in \S \ref{discussion}.
730:
731: \begin{figure}
732: \plotone{f6.eps}
733: \caption{Discrete correlation functions between V and R (top panel),
734: and B and R (bottom panel), respectively. The solid curves indicate
735: the best fit with a symmetric Gaussian. This leads to a best-fit maximum
736: correlation at $\tau_0 = (-1.14 \pm 0.48)$~d for the V band and $\tau_0
737: = (-3.75 \pm 0.96)$~d for the B band, indicating a lag of the V and B
738: band light curves behind the R-band. }
739: \label{DCF}
740: \end{figure}
741:
742: \section{\label{crosscorrelations}Inter-band cross-correlations and time lags}
743:
744: The result of an occasional counterclockwise hysteresis in 3C~279,
745: as found in the previous section, immediately suggests the existence
746: of a characteristic time lag of higher-frequency behind lower-frequency
747: variability. In order to investigate this, we evaluated the discrete
748: correlation function \citep[DCF,][]{ek88} between the R band and the
749: other optical light curves. In our notation, a positive value of the
750: time lag $\Delta \tau$ would indicate a lag of the R-band light curve
751: behind the comparison light curve. As mentioned earlier, the radio (and
752: near-IR) light curves are too sparsely sampled and yielded no significant
753: features in the DCF. Fig. \ref{DCF} shows the DCF between the R band and
754: the V band (top panel) and the B band (bottom panel), using a sampling time
755: scale of $\Delta\tau = 1$~d. We have done the same analysis using various
756: other values of $\Delta\tau$, which yielded results consistent with the
757: ones described below. We chose to show the results for $\Delta\tau = 1$~d
758: because they provided the best compromise between dense time scale sampling
759: and reduction of error bars.
760:
761: The DCFs reveal clear correlations between the different optical wavebands,
762: with peak values around 1. This confirms our previous impression from
763: inspection by eye, that the variability patterns in all optical wavebands
764: track each other very closely.
765:
766: \begin{figure}
767: \plotone{f7.eps}
768: \caption{Best-fit time lag of the R band vs. B, V, R, and I bands, as a
769: function of frequency.}
770: \label{tau_frequency}
771: \end{figure}
772:
773: The resulting DCFs have then been fitted with a symmetric Gaussian,
774: $F_0 \, e^{-(\tau - \tau_0)^2/(2 \sigma^2)}$. This analysis yields
775: non-zero offsets of the best-fit maxima, $\tau_0$ at the $\sim 3$ --
776: $4 \sigma$ level. Specifically, we find $\tau_0 = (-1.14 \pm 0.48)$~d
777: for the V band and $\tau_0 = (-3.75 \pm 0.96)$~d for the B band. This
778: indicates a hard time lag of higher-frequency variability behind the
779: variability at lower frequencies in the B-V-R frequency range. However,
780: this trend does not continue into the I band. This is illustrated in
781: Fig. \ref{tau_frequency}, where we plot the best-fit time lags as a
782: function of photon frequency. However, we need to add a note of caution:
783: The rather sparse sampling of the B- and V-band light curves leads to
784: large error bars on the DCFs. Clearly, alternative, more complex
785: representations, e.g., multiple Gaussians and/or asymmetric functions,
786: will also provide acceptable fits to the observed DCFs and may lead
787: to different quantitative results concerning the involved time lags.
788: Future observations with denser B- and V-band sampling are needed in
789: order to test the robustness of the trend found here.
790:
791: The hard lag found in this analysis may be physically related to
792: the $\sim 2$ -- 3~d time lag of the soft X-ray spectral index behind
793: the flux in {\it ROSAT} observations of 3C~279 in December 1992 --
794: January 1993 \citep{pian99}. Possible physical interpretations will
795: be discussed in \S \ref{discussion}.
796:
797:
798: \begin{figure}
799: \plotone{f8.eps}
800: \caption{Simultaneous snap-shot spectral energy distributions of 3C~279 at
801: various epochs. Data pertaining to the multiwavelength campaign in January
802: 2006 are plotted with red triangles. All IR and optical data were taken within
803: $\pm 1/2$~hr of UT 05:00 on Jan. 15, 2006; the X-ray and soft $\gamma$-ray
804: data represent the time-averaged spectra throughout the respective observing
805: windows in Jan. 2006, which included Jan. 15 for all instruments. Model fits
806: to the {\it EGRET} P1 and P2 SEDs are calculated using a time-dependent leptonic
807: (SSC + EC) jet model, and taken from \cite{hartman01a}. The June 2003 SED
808: is from \cite{collmar04}. }
809: \label{SED}
810: \end{figure}
811:
812: \section{\label{spectra}Broad-band spectral energy distribution}
813:
814: The most complete broadband coverage of 3C~279 during our campaign was
815: obtained on January 15, 2006. On that day, the only near-infrared (JHK)
816: exposures were taken with the NOT, and simultaneous radio observations
817: at 5, 8, 22, and 37~GHz were available as well. This is also within the
818: time window of the X-ray and soft $\gamma$-ray observations, although
819: in most cases, a meaningful extraction of spectral information required
820: the integration over most of the high-energy observing period, January
821: 13 -- 20, 2006 \citep{collmar07a}. The total snap-shot SED composed of
822: all available radio, near-IR, optical, X-ray, and soft $\gamma$-ray
823: observations around January 15, 2006 is displayed in Fig. \ref{SED},
824: where all near-IR and optical data are taken within $\pm 1/2$~hr of
825: UT 05:00 on January 15. The figure compares the January 2006 SED to
826: previous SEDs from the bright flare during the first {\it EGRET}
827: observing epoch in June 1991, the low state of December 1992 / January 1993
828: \citep[both SEDs adapted from][]{hartman01a}, and a previous multiwavelength
829: campaign around {\it INTEGRAL} AO-1 observations of 3C~279 in a low state
830: in June 2003 \citep{collmar04}.
831:
832: While the optical (presumably synchrotron) emission component clearly
833: indicates that the source was in an elevated state compared to previous
834: low states, the simultaneous X-ray -- soft-$\gamma$-ray spectrum
835: is perfectly consistent with the low-state spectra of 1992/1993 and
836: 2003. This is a very remarkable result and will be discussed in detail
837: in a companion paper about the results of the entire multiwavelength
838: campaign \citep{collmar07b}.
839:
840: The NIR -- optical continuum can be very well represented by a single
841: power-law with a spectral index of $\alpha_o = 1.64 \pm 0.04$, corresponding
842: to an underlying non-thermal electron distribution with a spectral index
843: of $p = 4.28 \pm 0.08$, if the optical continuum is synchrotron emission. As
844: already pointed out in \cite{hartman01a} for a comparison between various
845: {\it EGRET} observing epochs over $\sim 10$~years and confirmed by our
846: campaign data on time scales of weeks -- months (see \S \ref{variability}),
847: the optical spectral index in 3C~279 does not show any systematic
848: correlation with the brightness state of the source, and Fig. \ref{SED}
849: indicates that in our high-state observations of 2006, the continuum
850: spectral slope is not significantly different from the slopes observed
851: in the low states of 1992/1993 and 2003.
852:
853:
854: \section{\label{discussion}Discussion}
855:
856: In this section, we discuss some general physical implications and
857: constraints that our results can place on source parameters. In the
858: following discussion, we will parameterize the magnetic field in units
859: of Gauss, i.e., $B = 1 \, B_G$~Gauss, and the Doppler boosting factor
860: $D = \left(\Gamma [ 1 - \beta_{\Gamma} \cos\theta_{\rm obs}] \right)^{-1}$
861: in units of 10, i.e., $D = 10 \, D_1$ where $\Gamma$ is the bulk Lorentz
862: factor of the emitting region, $\beta_{\Gamma}$~c is the corresponding
863: speed, and $\theta_{\rm obs}$ is the observing angle. The characteristic
864: variability time scale is of the order of 1 -- a few days, so we write
865: $t_{\rm var}^{\rm obs} \equiv 1 \, t_{\rm var, d}^{\rm obs}$. In the
866: same sense, we parameterize the time lag of the B band behind the R band
867: as $\tau_{\rm BR}^{\rm obs} \equiv 1 \, \tau_{\rm BR, d}^{\rm obs}$~d with
868: $\tau_{\rm BR, d}^{\rm obs} \sim 3$. The observed variability time scale
869: yields an estimate of the size of the emitting region, $R \equiv 10^{15}
870: R_{15}$~cm through $R \lesssim c \, D/(1 + z) \, t_{\rm var}^{\rm obs}$.
871: We find $R_{15} \lesssim 17 \, D_1 \, t_{\rm var, d}^{\rm obs}$.
872:
873: The (co-moving) energies of electrons emitting synchrotron radiation at
874: their characteristic peak frequencies in the R and B bands are
875:
876: \begin{eqnarray}
877: \gamma_R &= 3.7 \times 10^3 \, \left( B_G \, D_1 \right)^{-1/2} \cr
878: \gamma_B &= 4.5 \times 10^3 \, \left( B_G \, D_1 \right)^{-1/2}
879: \label{gamma_BR}
880: \end{eqnarray}
881:
882: The steep underlying electron spectrum with $p = 4.5$, inferred from the
883: steep optical continuum, might indicate that the entire optical spectrum
884: is produced by electrons in the fast-cooling regime. This implies that
885: the radiative cooling time scale of electrons emitting synchrotron radiation
886: in the optical regime is shorter than the characteristic escape time scale
887: of those electrons. The respective time scales in the co-moving frame,
888: $t'_{\rm esc}$ and $t'_{\rm cool}$ can be written as
889:
890: \begin{eqnarray}
891: t'_{\rm esc} &\equiv \eta \, {R \over c} & \lesssim 5.7 \times 10^5 \, \eta \,
892: D_1 \, t_{\rm var, d}^{\rm obs} \; {\rm s} \cr
893: t'_{\rm cool} &= {\gamma \over {\dot\gamma}_{\rm rad}} &\sim 7.7 \times 10^7
894: \, B_G^{-1} \, \gamma^{-1} \, (1 + k)^{-1} \; {\rm s}
895: \label{timescales}
896: \end{eqnarray}
897: where $\eta \ge 1$ is the escape time scale parameter (as defined in
898: the first line of eq. \ref{timescales}), and $k$ is a correction
899: factor accounting for radiative cooling via Compton losses in the Thomson
900: regime in a radiation field with an energy density ${u'}_{\rm rad} \equiv
901: k \, {u'}_{\rm sy}$. Requiring that the cooling time scale is shorter than
902: the escape time scale, at least for electrons emitting in the R band,
903: leads to a lower limit on the magnetic field:
904:
905: \begin{equation}
906: B \gtrsim B_{\rm c} \equiv 1.3 \times 10^{-3} \, (1 + k)^{-2} \, \eta^{-1}
907: \, (t_{\rm var, d}^{\rm obs})^{-2} \; {\rm G}
908: \label{B_cooling}
909: \end{equation}
910: which does not seem to pose a severe constraint, given the values of
911: $B \sim$~a few G typically found for other FSRQs and also for 3C~279
912: from previous SED modeling analyses \citep[e.g.,][]{hartman01a}.
913:
914: Another estimate of the co-moving magnetic field can be found by
915: assuming that the dominant portion of the time-averaged synchrotron
916: spectrum is emitted by a power-law spectrum of electrons with
917: $N_e (\gamma) = n_0 \, V_B \, \gamma^{-p}$ for $\gamma_1 \le \gamma
918: \le \gamma_2$; here, $V_B$ is the co-moving blob volume, and we use
919: $p = 4.5$ as a representative value inferred from the optical continuum
920: slope. The normalization constant $n_0 = n_e \, (1 - p) /
921: \left(\gamma_2^{1 - p} - \gamma_1^{1 - p}\right)$ is related to the
922: magnetic field through an equipartition parameter $e_B \equiv {u'}_B /
923: {u'}_e$ (in the co-moving frame). Note that this equipartition parameter
924: only refers to the energy density of the electrons, not accounting for
925: a (possibly dominant) energy content of a hadronic matter component in
926: the jet. Under these assumptions, the magnetic field can be estimated
927: as described, e.g., in \cite{boettcher03}. Taking the $\nu F_{\nu}$ peak
928: synchrotron flux $f_{\epsilon}^{\rm sy}$ at the dimensionless synchrotron
929: peak photon energy $\epsilon_{\rm sy} \equiv E_{\rm pk, sy}/(m_e c^2)
930: \approx 3 \times 10^{-7}$ as $\sim 10^{-10}$~ergs~cm$^{-2}$~s$^{-1}$,
931: we find
932:
933: \begin{equation}
934: B \gtrsim B_{\rm e_B} \equiv 1.86 \, D_1^{-13/7} \, e_B^{2/7} \,
935: (t_{\rm var, d}^{\rm obs})^{-1} \; {\rm G}.
936: \label{B_eB}
937: \end{equation}
938:
939: This constitutes a more useful and realistic magnetic-field estimate than
940: eq. \ref{B_cooling}. If, indeed, the optical emission is synchrotron emission
941: from a fast-cooling electron distribution, then electrons have been primarily
942: accelerated to a power-law distribution with an injection index of $q = p - 1
943: = 3.5$. This is much steeper than the canonical spectral index of $q \sim 2.2$
944: -- 2.3 found for acceleration on relativistic, parallel shocks
945: \citep[e.g.,][]{gallant99,achterberg01}, and could indicate an oblique
946: magnetic-field orientation \citep[e.g.,][]{ob02,no04}, which would yield
947: a consistent picture with the predominantly perpendicular magnetic-field
948: orientation observed on parsec-scales \citep{jorstad04,ojha04,lh05}. The
949: observed hard lag (B vs. R) may then be the consequence of a gradual spectral
950: hardening of the electron acceleration (injection) spectrum throughout the
951: propagation of a relativistic shock front along the jet. Such a gradual
952: hardening of the electron acceleration spectrum could be the consequence
953: of the gradual build-up of hydromagnetic turbulence through the relativistic
954: two-stream instability \citep[see, e.g.,][]{schlickeiser02}. This
955: turbulence would harden the relativistic electron distribution via
956: 2nd-order Fermi acceleration processes \citep{vv05}. Such a scenario
957: would imply a length scale for the build-up of turbulence of
958: $\Delta r \sim c \, \tau_{\rm BR}^{\rm obs} \, D \, \Gamma / (1 + z)
959: \sim 5.6 \times 10^{-2} \, \tau_{\rm BR, d}^{\rm obs} \, D_1 \, \Gamma_1
960: \sim 0.2$~pc for the characteristic values of 3C~279.
961:
962: Alternatively, the acceleration process could become more efficient along
963: the jet if the magnetic-field configuration gradually evolves into a
964: more quasi-parallel one, on the same length scale of $\sim 0.2$~pc as
965: estimated above. However, this scenario might be in conflict with the
966: predominantly perpendicular magnetic-field orientation seen in the jets
967: of 3C~279 on pc scales.
968:
969: The hard lag in the optical regime may also be indicative of a slow
970: acceleration mechanism, with an acceleration time scale of the order
971: of the observed B vs. R lag. This would imply an acceleration rate of
972:
973: \begin{equation}
974: \dot\gamma_A \sim {\gamma_B - \gamma_R \over \tau_{\rm BR}}
975: \sim 6.8 \times 10^{-2} (\tau_{\rm BR, d}^{\rm obs})^{-1} \,
976: \left( {D_1 \over B_G} \right)^{1/2} \; {\rm s}^{-1}.
977: \label{acceleration}
978: \end{equation}
979: In this scenario, electrons could only be accelerated to at least $\gamma_B$,
980: if the acceleration rate of eq. \ref{acceleration} is larger than the absolute
981: value of the radiative (synchrotron + Compton) cooling rate corresponding to eq.
982: \ref{timescales}. This imposes an upper limit on the magnetic field:
983:
984: \begin{equation}
985: B \lesssim B_{\rm acc} \sim 0.42 \, D_1^{1/3} \, (\tau_{\rm BR, d}^{\rm obs})^{-2/3}
986: \, (1 + k)^{-2/3} \; {\rm G}.
987: \label{B_acc}
988: \end{equation}
989: This can be combined with the estimate in eq. \ref{B_eB} to infer a limit
990: on the magnetic-field equipartition parameter:
991:
992: \begin{equation}
993: e_B \lesssim e_{B, acc} \sim 5.8 \times 10^{-3} \, D_1^{23/3} \,
994: (\tau_{\rm BR, d}^{\rm obs})^{-7/3} \, (t_{\rm var, d}^{\rm obs})^{7/2} \,
995: (1 + k)^{-7/3}.
996: \label{eB_acc}
997: \end{equation}
998: Based on this equipartition parameter, one can use the magnetic-field
999: estimate of eq. \ref{B_eB} to estimate the total amount of co-moving
1000: energy contained in the emission region at any given time:
1001:
1002: \begin{equation}
1003: E'_e \sim {4 \over 3} \, \pi \, R^3 \, {{u'}_B \over e_B}
1004: \sim 2.5 \times 10^{49} \, D_1^{-4} \, \tau_{\rm BR, d}^{\rm obs}
1005: \, (t_{\rm var, d}^{\rm obs})^{-1/2} \, (1 + k) \; {\rm erg}
1006: \label{energy}
1007: \end{equation}
1008: Assuming that the bulk of this energy is dissipated within the
1009: characteristic variability time scale, one can estimate the power
1010: in relativistic electrons in the jet:
1011:
1012: \begin{equation}
1013: L_{\rm jet} \sim {E'_e \over {t'}_{\rm var}} \sim 4.5 \times 10^{43}
1014: \, D_1^{-5} \, \tau_{\rm BR, d}^{\rm obs} \, (t_{\rm var, d}^{\rm obs})^{-3/2}
1015: \, (1 + k) \; {\rm ergs \; s}^{-1}.
1016: \label{Ljet}
1017: \end{equation}
1018:
1019: Previous modeling works of the SEDs of FSRQs in general and 3C~279 in
1020: particular indicated characteristic magnetic field values of a few Gauss,
1021: in approximate equipartition with the ultrarelativistic electron population.
1022: The unusually low equipartition parameter in eq. \ref{eB_acc} could therefore
1023: pose a problem for the slow-acceleration scenario. Note, however, the very
1024: strong dependence of $e_B$ on the Doppler factor ($\propto D^{23/3}$). A
1025: Doppler factor $D \sim 20$ could account for equipartition parameters of
1026: the order of one. Also, the energy requirements of eqs. \ref{energy} and
1027: \ref{Ljet} seem reasonable, and there does not appear to be a strict argument
1028: that would rule this scenario out.
1029:
1030: Another scenario one could think of would be based on a decreasing magnetic
1031: field along the blazar jet, leading to a gradually increasing cooling
1032: break in the underlying electron distribution. This would require that the
1033: cooling time scale for electrons emitting synchrotron radiation in the
1034: optical regime would be equal to or longer than the escape time scale.
1035: Thus, the inequality in eq. \ref{B_cooling} would be reversed. This would
1036: require unreasonably low magnetic fields. Furthermore, this scenario would
1037: be in conflict with the typically observed unbroken snap-shot power-law
1038: continuum spectra throughout the optical-IR range. Therefore, this idea
1039: may be ruled out.
1040:
1041:
1042: \section{\label{summary}Summary}
1043:
1044: We have presented the results of an optical-IR-radio monitoring campaign
1045: on the prominent blazar-type flat-spectrum radio quasar 3C~279 by the
1046: WEBT collaboration in January -- April 2006, around Target of Opportunity
1047: X-ray and soft $\gamma$-ray observations with {\it Chandra} and {\it INTEGRAL}
1048: in mid-January 2006. Previously unpublished radio and optical data from
1049: several weeks leading up to the ToO trigger are also included.
1050:
1051: The source exhibited substantial variability of flux and spectral shape,
1052: in particular in the optical regime, with a characteristic time scale
1053: of a few days. The variability patterns throughout the optical BVRI
1054: bands were very closely correlated with each other, while there was
1055: no significant evidence for a correlation between the optical and
1056: radio variability. After the trigger flux level for the {\it Chandra}
1057: and {\it INTEGRAL} ToOs was reached on Jan. 5, 2006, the optical
1058: flux decayed smoothly by 1.1 mags. within 13 days, until the end
1059: of the time frame of the X-ray and $\gamma$-ray observations. The
1060: decay could be well described by an exponential decay with a
1061: decay time scale of $\tau_d = 12.8$~d. The flux then recovered
1062: to approximately the pre-dip values in a much more erratic way,
1063: including a $\sim 0.5^{\rm mag}$ rise within $\sim 1$~d.
1064:
1065: A discrete correlation function analysis between different optical
1066: (BVRI) bands indicates a hard lag with a time delay increasing with
1067: increasing frequency, reaching $\sim 3$~d for the lag of B behind
1068: R. This appears to be accompanied by a single indication of
1069: counterclockwise spectral hysteresis in a color-intensity diagram
1070: (B-R vs. R). Thus, spectral hardening during flares appears delayed
1071: with respect to a rising optical flux. There is no consistent
1072: overall trend of optical spectral hardness with source brightness.
1073: However, our data indicate that the source displays a rather uniform
1074: spectral slope of $\alpha_o \sim 1.9$ at moderate flux levels ($R >
1075: 14.5$), while spectral variability seems common at high flux levels
1076: ($R < 14.5$).
1077:
1078: The occasional optical spectral hysteresis, in combination with the
1079: very steep IR-optical continuum spectral index of $\alpha_o \sim 1.5$
1080: -- 2.0, may indicate a highly oblique magnetic field configuration near
1081: the base of the jet, leading to inefficient particle acceleration and a
1082: very steep electron injection spectrum. As the emission region propagates
1083: along the jet, a gradual hardening of the primarily injected ultrarelativistic
1084: electron distribution may be caused by the gradual build-up of hydromagnetic
1085: turbulence, which could lead to a gradually increasing contribution of
1086: second-order Fermi acceleration. This would imply a length scale of the
1087: build-up of hydromagnetic turbulence of $\Delta r \sim 0.2$~pc.
1088:
1089: An alternative explanation of the hard lag may be a slow acceleration
1090: mechanism by which relativistic electrons are accelerated on a time scale
1091: of several days. However, even though this model can plausibly explain the
1092: observed variability trends and overall luminosity of the source, it requires
1093: an unusually low magnetic field in the emitting region of $B \lesssim 0.2$~G,
1094: unless rather high Doppler factors of $D ~ 20$ are assumed. Such a small
1095: magnetic field would be about an order of magnitude lower than inferred
1096: from previous analyses of simultaneous SEDs of 3C~279 and other flat-spectrum
1097: radio quasars with similar properties.
1098:
1099:
1100: \acknowledgments
1101: The work of M. B\"ottcher and S. Basu was partially supported by
1102: NASA through INTEGRAL GO grant award NNG~06GD57G and the Chandra
1103: GO program (administered by the Smithsonian Astrophysical Observatory)
1104: through award no. GO6-7101A.
1105: The Mets\"ahovi team acknowledges the support from the Academy of
1106: Finland.
1107: YYK is a research fellow of the Alexamder von Humboldt Foundation.
1108: RATAN-600 observations were partly supported by the Russian Foundation
1109: for Basic Research (project 05-02-17377).
1110: The St. Petersburg team was supported by the Russian Foundation for
1111: Basic Research through grant 05-02-17562.
1112:
1113: \begin{thebibliography}{}
1114:
1115: \bibitem[Achterberg et al.(2001)]{achterberg01}Achterberg, A., Gallant, Y. A.,
1116: Kirk, J. G., \& Guthmann, A. W., 2001, MNRAS, 328, 393
1117:
1118: \bibitem[Bach et al.(2007)]{bach07}Bach, U., et al., 2007, A\&A, 464, 175
1119:
1120: \bibitem[Bednarek(1998)]{bednarek98}Bednarek, W., 1998, A\&A, 336, 123
1121:
1122: \bibitem[B\"ottcher(2007a)]{boettcher07a}B\"ottcher, M., 2007a, in proc.
1123: ``The Multimessenger Approach to Gamma-Ray Sources'', ApSS, in press
1124: (astro-ph/0608713)
1125:
1126: \bibitem[B\"ottcher(2007b)]{boettcher07b}B\"ottcher, M., 2007b, in proc.
1127: ``The Central Engine of Active Galactic Nuclei'', ASPCS, in press
1128:
1129: \bibitem[B\"ottcher \& Bloom(2000)]{bb00}B\"ottcher, M., \& Bloom, S. D., 2000,
1130: AJ, 119, 469
1131:
1132: \bibitem[B\"ottcher \& Chiang(2002)]{bc02}B\"ottcher, M., \& Chiang, J.,
1133: 2002, ApJ, 581, 127
1134:
1135: \bibitem[B\"ottcher, Mause, \& Schlickeiser(1997)]{bms97}B\"ottcher, M.,
1136: Mause, H., \& Schlickeiser, R., 1997, A\&A, 324, 395
1137:
1138: \bibitem[B\"ottcher et al.(2003)]{boettcher03}B\"ottcher, M., et al., 2003, ApJ,
1139: 596, 847
1140:
1141: \bibitem[B\"ottcher et al.(2005)]{boettcher05}B\"ottcher, M., et al., 2005, ApJ,
1142: 631, 169
1143:
1144: \bibitem[Collmar et al.(2004)]{collmar04}Collmar, W., et al., 2004, in proc.
1145: of 5th INTEGRAL Workshop, ESA-SP 552, Ed.: B. Battrick, p. 555
1146:
1147: \bibitem[Collmar et al.(2007a)]{collmar07a}Collmar, W., et al., 2007a, in proc.
1148: of 6th INTEGRAL Workshop, ESA-SP, in press
1149:
1150: \bibitem[Collmar et al.(2007b)]{collmar07b}Collmar, W., et al., 2007b, in preparation
1151:
1152: \bibitem[Cotton et al.(1979)]{cotton79}Cotton, W. D., et al., 1979, ApJ, 229, L115
1153:
1154: \bibitem[Edelson \& Krolik(1988)]{ek88}Edelson, R. A., \& Krolik, J. H.,
1155: 1988, ApJ, 333, 646
1156:
1157: \bibitem[Floyd et al.(2004)]{floyd04}Floyd, D. J. E., Kukula, M. J., Dunlop, J. S.,
1158: McLure, R. J., Miller, L., Percival, W. J., Baum, S. A., \& O'Dea, C. P., 2004,
1159: MNRAS, 355, 196
1160:
1161: \bibitem[Fossati et al.(2000)]{fossati00}Fossati, G., et al., 2000,
1162: ApJ, 541, 166
1163:
1164: \bibitem[Gallant, Achterberg \& Kirk(1999)]{gallant99}Gallant, Y. A.,
1165: Achterberg, A., \& Kirk, J. G., 1999, A\&AS, 138, 549
1166:
1167: \bibitem[Gonz\'alez-P\'erez et al.(2001)]{gp01}Gonz\'alez-P\'erez, J. N.,
1168: Kidger, M. R., \& Mart\'\i n-Luis, F., 2001, AJ, 122, 2055
1169:
1170: \bibitem[Hartman et al.(1996)]{hartman96}Hartman, R. C., et al.,
1171: 1996, ApJ, 461, 698
1172:
1173: \bibitem[Hartman et al.(1999)]{hartman99}Hartman, R. C., et al.,
1174: 1999, ApJS, 123, 79
1175:
1176: \bibitem[Hartman et al.(2001a)]{hartman01a}Hartman, R. C., et al.,
1177: 2001a, ApJ, 553, 683
1178:
1179: \bibitem[Hartman et al.(2001b)]{hartman01b}Hartman, R. C., et al.,
1180: 2001b, ApJ, 558, 583
1181:
1182: \bibitem[Helmboldt et al.(2007)]{helmboldt07}Helmboldt, J. F., et al.,
1183: 2007, ApJ, 658, 203
1184:
1185: \bibitem[Homan et al.(2003)]{homan03}Homan, D. C., Lister, M. L., Kellermann, K. I.,
1186: Cohen, M. H., Ros, E., Zensus, J. A., Kadler, M., \& Vermeulen, R. C., 2003,
1187: ApJ, 589, L9
1188:
1189: \bibitem[Jorstad et al.(2004)]{jorstad04}Jorstad, S. G., Marscher, A. P.,
1190: Lister, M. L., Stirling, A. M., Cawthorne, T. V., G\'omez, J.-L., \&
1191: Gear, W. K., 2004, AJ, 127, 3115
1192:
1193: \bibitem[Kartaltepe \& Balonek(2007)]{kb07}Kartaltepe, J. S., \& Balonek, T. J.,
1194: 2007, ApJ, 133, 2866
1195:
1196: \bibitem[Kataoka et al.(2000)]{kataoka00}Kataoka, J., Takahashi, T.,
1197: Makino, F., Inoue, S., Madejski, G. M., Tashiro, M., Urry, C. M.,
1198: \& Kubo, H., 2000, ApJ, 528, 243
1199:
1200: \bibitem[Kirk, Rieger, \& Mastichiadis(1998)]{kirk98}Kirk, J. G.,
1201: Rieger, F. M., \& Mastichiadis, A., 1998, A\&A, 333, 452
1202:
1203: \bibitem[Kusunose, Takahara, \& Li(2000)]{kusunose00}Kusunose, M., Takahara, F.,
1204: \& Li, H., 2000, ApJ, 536, 299
1205:
1206: \bibitem[Li \& Kusunose(2000)]{lk00}Li, H., \& Kusunose, M., 2000,
1207: ApJ, 536, 729
1208:
1209: \bibitem[Lister \& Homan(2005)]{lh05}Lister, M. L., \& Homan, D. C., 2005,
1210: AJ, 130, 1389
1211:
1212: \bibitem[Maraschi et al.(1994)]{maraschi94}Maraschi, L., et al., 1994,
1213: ApJ, 435, L91
1214:
1215: \bibitem[Mattox, Hartman, \& Reimer(2001)]{mhr01}Mattox, J. R.,
1216: Hartman, R. C., \& Reimer, O., 2001, ApJS, 135, 155
1217:
1218: \bibitem[Moderski et al.(2003)]{moderski03}Moderski, R., Sikora, M.,
1219: Blazejowski, M., 2003, A\&A, 406, 855
1220:
1221: \bibitem[M\"ucke \& Protheroe(2001)]{muecke01}M\"ucke, A., \& Protheroe, R. J.,
1222: 2001, Astropart. Phys., 15, 121
1223:
1224: \bibitem[M\"ucke et al.(2003)]{muecke03}M\"ucke, A., Protheroe, R. J.,
1225: Engel, R., Rachen, J. P., \& Stanev, T., 2003, Astropart. Phys.,
1226: 18, 593
1227:
1228: \bibitem[Nandikotkur et al.(2007)]{nandi07}Nandikotkur, G., Jahoda, K. M.,
1229: Hartman, R. C., Mukherjee, R., Sreekumar, P., B\"ottcher, M., Sambruna, R. M.,
1230: \& Swank, J. H., 2007, ApJ, 657, 705
1231:
1232: \bibitem[Niemiec \& Ostrowski(2004)]{no04}Niemiec, J., \& Ostrowski, M.,
1233: 2004, ApJ, 610, 851
1234:
1235: \bibitem[Ojha et al.(2004)]{ojha04}Ojha, R., Homan, D. C., Roberts, D. H.,
1236: Wardle, J. F. C., Aller, M. F., Aller, H. D., \& Hughes, P. A., 2004, ApJS,
1237: 150, 187
1238:
1239: \bibitem[Ostrowski \& Bednarz(2002)]{ob02}Ostrowski, M., \& Bednarz, J.,
1240: 2002, A\&A, 394, 1141
1241:
1242: \bibitem[Pian et al.(1999)]{pian99}Pian, E., et al., 1999, ApJ, 521, 112
1243:
1244: \bibitem[Raiteri et al.(1998)]{raiteri98}Raiteri, C. M., Villata, M.,
1245: Lanteri, L., Cavallone, M., \& Sobrito, G., 1998, A\&AS, 130, 495
1246:
1247: \bibitem[Raiteri et al.(2001)]{raiteri01}Raiteri, C. M., et al., 2001,
1248: A\&A, 377, 396
1249:
1250: \bibitem[Raiteri et al.(2005)]{raiteri05}Raiteri, C. M., et al., 2005,
1251: A\&A, 438, 39
1252:
1253: \bibitem[Raiteri et al.(2006)]{raiteri06}Raiteri, C. M., et al., 2006,
1254: A\&A, 459, 731
1255:
1256: \bibitem[Raiteri et al.(2007)]{raiteri07}Raiteri, C. M., et al., 2007, A\&,
1257: submitted
1258:
1259: \bibitem[Schlegel et al.(1998)]{schlegel98}Schlegel, D. J., Finkbeiner, D. P.,
1260: \& Davis, M., 1998, ApJ, 500, 525
1261:
1262: \bibitem[Schlickeiser et al.(2002)]{schlickeiser02}Schlickeiser, R., Vainio, R.,
1263: B\"ottcher, M., Schuster, C., Lerche, I., \& Pohl, M., 2002, A\&A, 393, 69
1264:
1265: \bibitem[Sikora et al.(2001)]{sikora01}Sikora, M., Blazejowski, M.,
1266: Begelman, M. C., \& Moderski, R., 2001, ApJ, 554, 1;
1267: Erratum: ApJ, 561, 1154 (2001)
1268:
1269: \bibitem[Smith \& Balonek(1998)]{sb98}Smith, P. S., \& Balonek, T. J.,
1270: 1998, PASP, 110, 1164
1271:
1272: \bibitem[Takahashi et al.(1996)]{takahashi96}Takahashi, T., et al., 1996,
1273: ApJ, 470, L89
1274:
1275: \bibitem[Ter\"asranta et al.(1998)]{ter98}Ter\"asranta, H., et al., 1998,
1276: A\&AS, 132, 305
1277:
1278: \bibitem[Unwin et al.(1989)]{unwin89}Unwin, S. C., Biretta, J. A.,
1279: Hodges, M. W., \& Zensus, J. A., 1989, ApJ, 340, 117
1280:
1281: \bibitem[Villata et al.(2000)]{villata00}Villata, M., et al., 2000,
1282: A\&A, 363, 108
1283:
1284: \bibitem[Villata et al.(2002)]{villata02}Villata, M., et al., 2002,
1285: A\&A, 390, 407
1286:
1287: \bibitem[Villata et al.(2004a)]{villata04a}Villata, M., et al., 2004a,
1288: A\&A, 421, 103
1289:
1290: \bibitem[Villata et al.(2004b)]{villata04b}Villata, M., et al., 2004b,
1291: A\&A, 424, 497
1292:
1293: \bibitem[Villata et al.(2006)]{villata06}Villata, M., et al., 2006, A\&A,
1294: 453, 817
1295:
1296: \bibitem[Villata et al.(2007)]{villata07}Villata, M., et al., 2007,
1297: A\&A, 464, L5
1298:
1299: \bibitem[Virtanen \& Vainio(2005)]{vv05}Virtanen, J. J. P., \& Vainio, R.,
1300: 2005, ApJ, 621, 313
1301:
1302: \bibitem[Wehrle et al.(1998)]{wehrle98}Wehrle, A. E., et al., 1998, ApJ,
1303: 497, 178
1304:
1305: \bibitem[Whitney et al.(1971)]{whitney71}Whitney, A. R., et al., 1971, Science,
1306: 173, 225
1307:
1308: \bibitem[Zakamska et al.(2006)]{zakamska06}Zakamska, N. L., et al., 2006, ApJ,
1309: 132, 1496
1310:
1311: \end{thebibliography}
1312:
1313: \end{document}
1314:
1315: