1: \documentclass{emulateapj}
2: \usepackage{txfonts}
3:
4:
5: \slugcomment{ApJL, in press}
6:
7: \shorttitle{Radio--X-ray Flare in Cygnus~X-1}
8: \shortauthors{Wilms et al.}
9:
10: \begin{document}
11:
12: \title{Correlated radio--X-ray variability of Galactic Black Holes: A
13: radio--X-ray flare in Cygnus~X-1}
14:
15: \author{
16: J\"orn Wilms\altaffilmark{1},
17: Katja Pottschmidt\altaffilmark{2},
18: Guy G. Pooley\altaffilmark{3},
19: Sera Markoff\altaffilmark{4},
20: Michael A. Nowak\altaffilmark{5},
21: Ingo Kreykenbohm\altaffilmark{6,7},
22: Richard E. Rothschild\altaffilmark{2}
23: }
24:
25: \altaffiltext{1}{Dr.\ Karl Remeis-Observatory, University of
26: Erlangen-Nuremberg, Sternwartstr.~7, 96049 Bamberg, Germany, joern.wilms@sternwarte.uni-erlangen.de}
27: \altaffiltext{2}{Center for Astrophysics and Space Sciences,
28: University of California, San Diego, La Jolla, CA 92093-0424, USA,
29: kpottschmidt@ucsd.edu, rrothschild@ucsd.edu}
30: \altaffiltext{3}{Mullard Radio Astronomy Observatory, Cavendish
31: Laboratory, Madingley Road, Cambridge CB3 0HE, UK, guy@mrao.cam.ac.uk}
32: \altaffiltext{4}{Astronomical Institute ``Anton Pannekoek'',
33: University of Amsterdam, Kruislaan 403, Amsterdam, 1098 SJ, The
34: Netherlands, sera@astro.uva.nl}
35: \altaffiltext{5}{MIT Kavli Institute for Astrophysics and Space
36: Research and \textsl{Chandra} X-ray Center, NE80-6077, 77
37: Massachusetts Ave., Cambridge, MA 02139, USA, mnowak@alum.mit.edu}
38: \altaffiltext{6}{Institut f\"ur Astronomie und Astrophysik --
39: Astronomie, Sand 1, 72076 T\"ubingen, Germany,
40: kreyken@astro.uni-tuebingen.de}
41: \altaffiltext{7}{\textsl{INTEGRAL} Science Data Centre, Chemin
42: d'\'Ecogia 16, 1290 Versoix, Switzerland}
43:
44: \email{joern.wilms@sternwarte.uni-erlangen.de}
45:
46: \begin{abstract}
47: We report on the first detection of a quasi-simultaneous
48: radio--X-ray flare of Cygnus~X-1. The detection was made on 2005
49: April 16 with pointed observations by the Rossi X-ray Timing
50: Explorer and the Ryle telescope, during a phase where the black hole
51: candidate was close to a transition from the its soft into its hard
52: state. The radio flare lagged the X-rays by $\sim$7\,minutes,
53: peaking at 3:20\,hours barycentric time (TDB 2453476.63864). We
54: discuss this lag in the context of models explaining such flaring
55: events as the ejection of electron bubbles emitting synchrotron
56: radiation.
57: \end{abstract}
58:
59: \keywords{X-rays: stars -- X-rays: binaries -- Black Hole Physics --
60: accretion, accretion disks}
61:
62:
63: \section{Introduction}
64: With the increased availability of simultaneous radio and observations
65: in the last decade, there is now a large amount of evidence available
66: pointing towards a very close interaction between the accretion disk
67: and the jet in black hole X-ray binaries and active galactic nuclei
68: (AGN). Most convincingly, this disk-jet interaction has been shown for
69: microquasars, i.e., black hole binaries with strongly relativistic
70: jets such as \object{GRS~1915+105} or \object{GRO J1655$-$40}. In
71: these systems, the correlated flaring in the X-rays,
72: optical/infra-red, and radio seen at certain times is generally
73: interpreted as the evidence for (ballistic) ejection events of
74: synchrotron radiation-emitting electron bubbles \citep[][ and
75: therein]{rothstein:05a,fender:04a,kleinwolt:02a,eikenberry:98a}. In
76: this model, the X-ray flare represents the ejection of the synchrotron
77: radiation emitting bubble, which then adiabatically expands within the
78: jet flow and cools down, resulting in the peak of the emission
79: shifting downwards in frequency with time
80: \citep{vanderlaan:66a,hjellming:88a}. Simultaneous broadband
81: observations of such events, which show minute-long delays between the
82: different wave bands, are consistent with this picture
83: \citep{mirabel:94a,mirabel:98a,pooley:97a,eikenberry:98a}. The model
84: has also been confirmed by proper motion measurements in the radio,
85: which reveal intrinsic jet speeds of $\gtrsim$$0.57c$ for GRS~1915+105
86: \citep{millerjones:05a}. Comparable behavior was also detected in
87: \object{3C\,120}, suggesting that similar ejections also occur in
88: active galactic nuclei, on correspondingly longer timescales
89: \citep{marscher:02a}.
90:
91: For black hole binaries with weakly relativistic jets, the evidence
92: for jet-disk-interaction is less direct. This evidence includes the
93: correlation between X-ray states and radio emission in black hole
94: transients \citep[e.g., in
95: \object{GX~339$-$4}][]{corbel:03a,belloni:05a} and the success of
96: modeling the radio to X-ray broad band spectrum of black hole
97: candidates with outflow-dominated models \citep[][ and
98: therein]{markoff:05a,markoff:04a}. Furthermore, at least for
99: \object{Cygnus~X-1}, there is also evidence for the presence of an
100: energetically significant, strong outflow
101: \citep{stirling:98a,stirling:01a,gallo:05a,millerjones:06a}. A
102: relativistic jet with $v\gtrsim 0.3c$ has been associated with radio
103: flares in this system \citep{fender:06a}.
104:
105: Apart from GRS~1915+105, however, none of these observations shows
106: direct evidence for a causal connection between the X-rays and jet on
107: time scales of minutes. Prompted by this lack of quasi-simultaneous
108: short-term radio--X-ray correlations, in 1998 we initiated a long term
109: monitoring campaign of Cyg~X-1 with the Rossi X-ray Timing Explorer
110: (\textsl{RXTE}) and the Ryle telescope. Biweekly 3--10\,ksec long
111: simultaneous observations started in 1999. Previous searches for
112: flares in campaign data taken between 1999 and mid-2003 did not reveal
113: evidence for coherent short term activity in both bands, although a
114: significant correlation on time scales of weeks was found, especially
115: above $\sim$10\,keV \citep{gleissner:04a,wilms:05a}. In this
116: \emph{Letter}, we report on the observation made on 2005 April 16, in
117: which the first clear quasi-simultaneous radio--X-ray flare was
118: detected in Cyg~X-1. The remainder of this \emph{Letter} is
119: structured as follows. In \S\ref{sec:obs} we describe the
120: observations, followed by the analysis of the flare in
121: \S\ref{sec:flare}. We discuss the physics of the flare in the context
122: of emission models for the radio and X-ray emission in
123: \S\ref{sec:discuss}.
124:
125: \section{Observations and Data Reduction}\label{sec:obs}
126: We use data from both instruments on-board the \textsl{RXTE}, the low
127: energy Proportional Counter Array \citep[PCA;][]{jahoda:05a} and the
128: High Energy X-ray Timing Experiment \citep[HEXTE;][]{rothschild:98a}.
129: The data analysis was performed using the standard \textsl{RXTE} data
130: analysis software, HEASOFT 6.1.2. Spectral fitting was performed with
131: XSPEC 11.3.2aa \citep{arnaud:96a}.
132:
133: A crucial part of the observation happened during the early phase of
134: the \textsl{RXTE} observation, shortly after the source rose above the
135: Earth's horizon. Due to auroral emission in the far ultra-violet and
136: soft X-rays and due to cosmic ray reprocessing in the hard X-rays, the
137: Earth's atmosphere is not completely X-ray dark. The typical
138: 2.5--20\,keV X-ray flux at the typical magnetic latitude of the
139: \textsl{RXTE} orbit is too low, however, to influence our measurements
140: \citep{petrinec:00a,sazonov:07a}. We therefore use all data taken
141: whilst the source was $\ge$$1^\circ$ above the Earth's horizon and had
142: a source offset of $\le$$0\fdg{}01$. We use PCA data from the
143: standard1 mode, which gives the full 2.5--128\,keV PCA count rate with
144: a time resolution of 0.125\,s and no energy information, and from the
145: standard2f mode, a binned data mode with a 128 channel energy
146: resolution and a time resolution of 16\,s. X-ray light curves were
147: extracted with the intrinsic time resolution of each mode and then
148: barycentered and rebinned.
149:
150: The Ryle telescope data were taken at 15\,GHz with a time resolution
151: of $\sim$8\,s. The typical $1\sigma$ uncertainty of the radio
152: measurements is 9\,mJy. The observations are interrupted every
153: $\sim$1600\,s for phase calibration observations of J2007+4029. The
154: amplitude calibration of the Ryle data corresponds to the flux scale
155: of \citet{baars:77a} and is performed using nearby observations of
156: 3C48 and 3C286. See \citet{pooley:97a} for further information on the
157: Ryle telescope.
158:
159:
160: \begin{figure}
161: \plotone{f1.eps}
162: \caption{\textsl{RXTE}-ASM (lower curve, left $y$-axis) and Ryle telescope
163: 15\,GHz (upper curve, right $y$-axis) lightcurves of Cyg~X-1 from
164: 2004 March until 2005 June. The time of the simultaneous observation
165: is indicated by the dashed vertical line. Gaps are shown if their
166: duration is $\ge$4\,d.}\label{fig:longterm}
167: \end{figure}
168:
169: \section{A Quasi-Simultaneous Radio--X-ray Flare}\label{sec:flare}
170:
171: \begin{figure}
172: \plotone{f2.eps}
173: \caption{Barycentric 15\,GHz flux density (left axis, dotted line) and
174: 2--128\,keV count rate lightcurves (right axis, solid line) of
175: Cyg~X-1 for the flare event of 2005 April 16, both binned to a
176: resolution of 32\,s. The bottom axis gives the barycentric time in
177: hours, with tick-marks separated by 10\,minutes.}\label{fig:lc}
178: \end{figure}
179:
180: As shown in Fig.~\ref{fig:longterm}, 2005 April marks the possible end
181: of a longer X-ray flaring episode of Cyg~X-1 that started in early
182: 2004 \citep{wilms:05a}. While clearly defined radio flares are not
183: uncommon in Cyg~X-1 \citep[e.g.,][]{hjellming:73a}, increased radio
184: emission and radio flaring are generally seen when the source is in
185: the intermediate state between the hard and the soft states, while the
186: radio is weak once the X-ray source approaches the soft state
187: \citep[][ and therein]{wilms:05a}. At the time of our pointed
188: observations, the soft X-ray flux had just come down from a large
189: flare. Shortly after the observation, the 1\,d averaged 15\,GHz flux
190: peaked, reaching a maximum of $\sim$30\,mJy, close to the brightest
191: radio flux of Cyg~X-1 during 2004/2005.
192:
193: Figure~\ref{fig:lc} shows the 15\,GHz radio flux and the \textsl{RXTE}
194: PCA count rate lightcurve measured on 2005 April 16. Close to the
195: start of the observation, a radio flare is readily apparent. The total
196: duration of the flare is $\sim$15\,minutes. During this interval the
197: 15\,GHz flux increased by a factor of $\sim$3 to a peak radio flux of
198: 70\,mJy. This radio flux is among the highest seen during the Ryle
199: monitoring\footnote{The most exceptional radio flare was that of 2004
200: February 20, which reached a peak flux of 140\,mJy at 15\,GHz, the
201: largest flux ever seen for this source with the Ryle telescope
202: \citep{fender:06a}.}. Previous radio flares, however, did not occur
203: during pointed \textsl{RXTE} observations
204: \citep{gleissner:04a,fender:06a}, and the source monitoring provided
205: by the \textsl{RXTE}-ASM is too coarse to pick up such short lived
206: X-ray events.
207:
208: \textsl{RXTE} started observing Cyg~X-1 about 10 minutes before the
209: peak radio flux. The X-ray lightcurve shown in Fig.~\ref{fig:lc} shows
210: a similar shape to the radio one, although with more substructure.
211: The earlier maximum of the X-ray flare did not allow \textsl{RXTE} to
212: catch the start of the X-ray flare, or determine whether the maximum
213: X-ray flux seen is indeed the peak of the X-ray flare. A cross
214: correlation (CCF) analysis using the algorithm of \citet{scargle:89a}
215: reveals a $413\pm 165$\,s time lag of the radio with respect to the
216: X-rays, where the $1\sigma$ uncertainty was determined using a
217: standard bootstrapping method with 1000 realizations. Other
218: approaches to calculate the CCF for non-uniformly sampled data
219: \citep{alexander:97a,edelson:88a} give essentially the same result.
220: With a maximum \citet{scargle:89a}-CCF of 0.38, this analysis formally
221: confirms the general similarity of the X-ray and radio lightcurves
222: (Fig.~\ref{fig:corr}). Since the substructure of the X-ray
223: lightcurve, i.e., the two smaller flares after the main flare, is
224: clearly different from that in the radio flare and since the start of
225: the X-ray flare is not covered by our observations, the peak CCF value
226: is not higher. For the same reasons, the formal uncertainty of the
227: lag measurement is rather large.
228:
229:
230: \begin{figure}
231: \plotone{f3.eps}
232: \caption{Cross-correlation function \citep{scargle:89a} for the radio
233: data with respect to the \textsl{RXTE}-PCA data. A negative time lag
234: indicates the radio lagging the X-rays.}\label{fig:corr}
235: \end{figure}
236:
237: To characterize the shape of the radio flare, we fit the radio data
238: (rebinned to a resolution of 8\,s) with the sum of a linear flux trend
239: and a Gaussian representing the flare,
240: \begin{equation}\label{eq:flarefunc}
241: f(t) = a (t-t')+b + A \exp\left( -\frac{(t-T)^2}{\sigma^2}\right)
242: \end{equation}
243: where $t'$ is a reference time, taken as the center of the time
244: interval analyzed ($t'={\rm JD}\,2453476.64635$). The barycentric time
245: of maximum flux occured at $T={\rm JD}\,2453476.63864$ with an
246: uncertainty of 15\,s (uncertainties are at the $1\sigma$ level for one
247: interesting parameter). The peak flux of the flare component is
248: $A=39.2\pm2.5$\,mJy and the width of the component is $\sigma=210\pm
249: 16$\,s. The flare is superimposed to a continuum with
250: $b=25.3\pm0.7$\,mJy, decreasing linearly with $a=-13\pm2\,{\rm
251: mJy}\,{\rm h}^{-1}$. The high quality of the fit is indicated by
252: its low reduced $\chi^2$ ($\chi^2_{\rm red}=1.11$ for 233 degrees of
253: freedom). The radio flare is therefore symmetric around its maximum.
254: This symmetry is a marked difference compared to the asymmetric flare
255: of 2004 February 20 \citep{fender:06a}.
256:
257: Modeling the \textsl{RXTE}-PCA light curve is complicated by the flare
258: already being in progress when the measurements started. Furthermore,
259: contrary to the radio data, where the scatter in the lightcurve is
260: mainly due to the measurement uncertainty, the X-ray data are
261: dominated by strong low-frequency noise onto which the X-ray flare is
262: superimposed. Consequently, the empirical model of
263: Eq.~(\ref{eq:flarefunc}) does not result in a good description of the
264: X-ray data.
265:
266: To study the spectral evolution of Cyg~X-1 during the flare, we
267: perform a spectral analysis of the 2.5--20\,keV PCA standard2f data at
268: 16\,s time resolution using a simple photoabsorbed powerlaw, which
269: proves sufficient to describe the spectrum at this lower signal to
270: noise level. For spectra taken during the flare, at PCA count rates
271: above $1000\,{\rm counts}\,{\rm s}^{-1}\,{\rm PCU}^{-1}$, the mean
272: power law index $\Gamma=2.10\pm 0.03$. The spectrum hardens outside of
273: the flare to $\Gamma=1.98\pm 0.03$ (errors given are the standard
274: deviation of the individual power law fits to the standard2f spectra),
275: a value typical for the intermediate state of this source \citep{wilms:05a}.
276:
277: That Cyg~X-1 was in the intermediate state on the day of the flare can
278: also be confirmed by modeling the 2.5--150\,keV PCA and HEXTE spectrum
279: of an \textsl{RXTE} observation performed 4\,h after the flare (to
280: avoid possible ``contamination'' by the flaring activity) with the sum
281: of a photoabsorbed ($N_{\rm H}=6\times 10^{21}\,{\rm cm}^{-2}$, held
282: fixed), exponentially cut-off broken power law model. This empirical
283: model has been shown to give a good characterization of the spectral
284: shape of Cyg~X-1 \citep{wilms:05a}. The spectral parameters are a
285: lower photon index $\Gamma_1=2.01\pm 0.01$, breaking at $E_{\rm
286: break}=10.0^{+0.3}_{-0.2}\,{\rm keV}$ into a power law with
287: $\Gamma_2=1.62^{+0.02}_{-0.03}$. At $E_{\rm cut}=26\pm3\,{\rm keV}$
288: the exponential cutoff starts with a folding energy of $E_{\rm
289: fold}=137\pm 11\,{\rm keV}$ (all uncertainties are at the 90\%
290: level). In addition, a Fe K$\alpha$ line from neutral iron is present
291: with an equivalent width of 135\,eV. The 3--10\,keV source flux is
292: $6.17\times 10^{-9}\,{\rm ergs}\,{\rm cm}^{-2}\,{\rm s}^{-1}$. The
293: parameters of the continuum are again consistent with an intermediate
294: state and fits well with the empirical picture that radio flaring in
295: black hole candidates occurs most frequently in this state
296: \citep{fender:04b,wilms:05a}.
297:
298:
299: \begin{figure}
300: \plotone{f4.eps}
301: \caption{1--128\,keV \textsl{RXTE}-PCA count rate versus hardness
302: ratio for Cyg X-1 observations from the 2000--2005 \textsl{RXTE}
303: monitoring (black) and for data taken during the 2005 April 16 flare
304: (red). The dotted line indicates the location at which observations
305: with maximum radio flux are found in the long term monitoring. The
306: hardness ratio was determined from background subtracted PCA top
307: layer data from channels 8--14 and 15--25,
308: respectively. \textit{A color version of this figure is available in
309: the online edition of ApJL.}}\label{fig:hardness_int}
310: \end{figure}
311:
312:
313: To allow the interpretation of the observed softening during the flare
314: with the general behavior of Cyg~X-1, Fig.~\ref{fig:hardness_int}
315: shows the X-ray hardness intensity diagram for the PCA top layer
316: standard2f 16\,s spectra in the context of the pointed \textsl{RXTE}
317: observations of the monitoring campaign. For black hole transients
318: this diagram is seen to have an approximate \textsf{q}-shape \citep[][
319: and therein]{belloni:06a}. As a persistent hard state source, Cyg~X-1
320: is typically found in the top right corner of the diagram. Outside of
321: the flare, the source is situated at a hardness of $\sim$0.55 with a
322: typical 1--128\,keV PCA count rate of
323: $\sim$$900\,{\rm counts}\,{\rm s}^{-1}\,{\rm PCU}^{-1}$. During the
324: flare the source softens and brightens. It leaves the region of the
325: diagram where Cyg~X-1 is usually found during our monitoring campaign,
326: by moving to higher count rates, for this hardness, than usually
327: observed.
328:
329: Flaring behavior in Cyg~X-1 is usually observed whenever the source is
330: close to the ``jet line'' in its hardness intensity diagram, while the
331: radio flux gets quenched once the source moves away from the jet line
332: to the left of the diagram \citep{gallo:03a,fender:04b}. In spectral
333: fits based on the \texttt{eqpair}-model of \citet{coppi:99a}, the
334: radio fluxes of Cyg~X-1 are at their maximum when the compactness
335: ratio $\ell_{\rm h}/\ell_{\rm s}\sim 3$ \citep[Fig.~16 of][; the
336: compactness ratio is a measure for the relative importance of the
337: energy dumped into the Comptonizing plasma and that dissipated in the
338: accretion disk]{wilms:05a}, corresponding to a soft power law index of
339: $\Gamma_1\sim 2.1$. From our database of spectra of Cyg~X-1, we find
340: that these observations have a $(7.1\mbox{--}11.7\,{\rm
341: keV})/(4.6\mbox{--}7.1\,{\rm keV})$ hardness ratio of 0.52,
342: indicated by the dotted line in Fig.~\ref{fig:hardness_int}.
343: During the 2005 April 16 observation, the source was therefore close
344: to this line of maximum radio flux and approached it asymptotically
345: during the flare.
346:
347:
348: \section{Discussion}\label{sec:discuss}
349:
350: In this \emph{Letter} we have presented the first evidence for a
351: direct relationship of the X-ray and radio emission in Cyg~X-1 on
352: timescales of minutes. The data show the radio to lag the X-rays by
353: $413\pm 165$\,s and the X-ray spectral shape to approach the X-ray
354: hardness ratio where the source is typically found at its largest
355: radio flux in our long term monitoring, hinting towards a general
356: similarity of the physics of individual flare events and the overall
357: radio--X-ray connection. Although the X-ray data do not cover the
358: start of the X-ray flare, explaining the rather large uncertainty of
359: the lag determination, the morphological similarities between the
360: X-ray and radio lightcurves also suggest that the same event is
361: observed in both wavebands.
362:
363: Similar events in microquasars show that lags with timescales of
364: several 100\,s are typical for the coherent behavior of these systems,
365: such as a lag of 310$\pm$20\,s between the soft X-rays and the IR
366: \citep{eikenberry:98a} and 800\,s between the X-rays and the radio
367: \citep{pooley:97a} in GRS~1915+105. The timescale observed in Cyg~X-1
368: allows us to place an upper limit to the physical separation of the
369: X-ray and radio emitting regions of the accretion/ejection flow. We
370: assume that the emission coincides with the imaged jet and that the
371: jet is perpendicular to the orbital plane of the HDE~226868/Cyg~X-1
372: system \citep[although this is not a priori
373: certain;][]{maccarone:02d}, which has an inclination of $30^\circ$
374: \citep{gies:82a,dolan:92a}. Taking light travel time effects into
375: account, for jet speeds of $0.3\,c$, the lower limit implied by
376: observations of the transient jet ejection discussed by
377: \citet{fender:06a}, the measured delay implies a separation of $1.1\pm
378: 0.5$\,AU between the location of X-ray and radio emission. If the jet
379: is relativistic instead, with a speed of $0.99c$, the distance
380: increases to $5.8\pm2.3$\,AU. Note that similar values for the length
381: of the jet are obtained by considering that the $\sim$10\,minute
382: duration of the radio flare is roughly equal to the dynamical
383: timescale of the jet. Assuming a distance of 2.5\,kpc, these values
384: imply to a maximum projected angular separation between the X-ray and
385: radio emitting region of $10^{-3}$\,mas.
386:
387: What is the physics of the observed event? In the model of
388: \citet{fender:04b} for transient radio events, the inner edge of a
389: thin accretion disk is posited to move rapidly towards the black hole.
390: The temperature at the inner edge therefore increases, leading to a
391: softening of the source in the \textsl{RXTE}-PCA as more disk photons
392: enter the instrument's band pass. This X-ray flare is then followed by
393: the ejection of an electron bubble, which rapidly expands, producing
394: the observed radio emission. At least qualitatively, this behavior and
395: also the time scales deduced above seem to agree with our
396: observations, although the model was originally invented for the large
397: scale variability of black hole candidates and not for such short
398: events as the one discussed here. Note that while the main flare
399: dominates the measured time lag and therefore the sizes estimated
400: above, it is followed by two short spikes, which are both present in
401: the radio and the X-ray lightcurves, but at different time delays.
402: These spikes could indicate that more than one blob of material was
403: ejected at different speeds but cannot be separated once the blobs
404: have expanded and their radiation peaks in the radio. Such a behavior
405: could be typical for flares in Cyg~X-1, since the large radio flare of
406: Cyg~X-1 from 2005 Feb 20 also shows very little substructure
407: \citep{fender:06a}.
408:
409: The lack of further detections of radio--X-ray flares in over
410: 1.5\,Msec of simultaneous radio--X-ray data precludes a more detailed
411: discussion of the properties of the ejected material from comparing
412: the source behavior during different flares. The observation of the
413: flare itself, however, stresses the importance of long-term
414: multi-wavelength campaigns to detect such rare events which are
415: necessary to further our insight into the physics of the emission from
416: black holes.
417:
418: \acknowledgments
419:
420: We acknowledge the support of NASA contract NAS5-30720, NASA grant
421: NNG05GK55G, and a travel grant from the Deutscher Akademischer
422: Austauschdienst. We thank the \textsl{RXTE} schedulers for their
423: efforts in making the simultaneous radio--X-ray observations possible
424: and the Aspen Center for Physics for its hospitality during the early
425: stages of the preparation of this manuscript.
426:
427:
428: {\it Facilities:} \facility{RXTE (PCA,HEXTE)}, \facility{Ryle ()}
429:
430:
431: \begin{thebibliography}{39}
432: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
433:
434: \bibitem[{{Alexander}(1997)}]{alexander:97a}
435: {Alexander}, T. 1997, in Astronomical Time Series, ed. D.~Maoz, A.~Sternberg,
436: \& E.~M. Leibowitz (Dordrecht: Kluwer), 163
437:
438: \bibitem[{Arnaud(1996)}]{arnaud:96a} Arnaud, K.~A. 1996, in
439: Astronomical Data Analysis Software and Systems {V}, ed. J.~H.
440: Jacoby \& J.~Barnes, Astron.\ Soc.\ Pacific Conf.\ Ser.\ 101 (San
441: Francisco: Astron.\ Soc.\ Pacific), 17
442:
443: \bibitem[{{Baars} {et~al.}(1977){Baars}, {Genzel}, {Pauliny-Toth}, \&
444: {Witzel}}]{baars:77a}
445: {Baars}, J.~W.~M., {Genzel}, R., {Pauliny-Toth}, I.~I.~K., \& {Witzel}, A.
446: 1977, A\&A, 61, 99
447:
448: \bibitem[{{Belloni} {et~al.}(2005){Belloni}, {Homan}, {Casella}, {van der
449: Klis}, {Nespoli}, {Lewin}, {Miller}, \& {M{\'e}ndez}}]{belloni:05a}
450: {Belloni}, T., {Homan}, J., {Casella}, P., {van der Klis}, M., {Nespoli}, E.,
451: {Lewin}, W.~H.~G., {Miller}, J.~M., \& {M{\'e}ndez}, M. 2005, A\&A, 440, 207
452:
453: \bibitem[{{Belloni} {et~al.}(2006){Belloni}, {Parolin}, {Del Santo}, {Homan},
454: {Casella}, {Fender}, {Lewin}, {M{\'e}ndez}, {Miller}, \& {van der
455: Klis}}]{belloni:06a}
456: {Belloni}, T., {Parolin}, I., {Del Santo}, M., {Homan}, J., {Casella}, P.,
457: {Fender}, R.~P., {Lewin}, W.~H.~G., {M{\'e}ndez}, M., {Miller}, J.~M., \&
458: {van der Klis}, M. 2006, MNRAS, 367, 1113
459:
460: \bibitem[{Coppi(1999)}]{coppi:99a}
461: Coppi, P. 1999, in High Energy Processes in Accreting Black Holes, ed.
462: J.~Poutanen \& R.~Svensson, Astron.\ Soc.\ Pacific Conf.\ Ser.\ 161 (San
463: Francisco: Astron.\ Soc.\ Pacific), 375
464:
465: \bibitem[{Corbel {et~al.}(2003)Corbel, Nowak, Fender, Tzioumis, \&
466: Markoff}]{corbel:03a}
467: Corbel, S., Nowak, M.~A., Fender, R.~P., Tzioumis, A.~K., \& Markoff, S. 2003,
468: A\&A, 400, 1007
469:
470: \bibitem[{Dolan(1992)}]{dolan:92a}
471: Dolan, J.~F. 1992, ApJ, 384, 249
472:
473: \bibitem[{Edelson \& Krolik(1988)}]{edelson:88a}
474: Edelson, R.~A., \& Krolik, J.~H. 1988, ApJ, 333, 646
475:
476: \bibitem[{{Eikenberry} {et~al.}(1998){Eikenberry}, {Matthews}, {Morgan},
477: {Remillard}, \& {Nelson}}]{eikenberry:98a}
478: {Eikenberry}, S.~S., {Matthews}, K., {Morgan}, E.~H., {Remillard}, R.~A., \&
479: {Nelson}, R.~W. 1998, ApJ, 494, L61
480:
481: \bibitem[{Fender \& Belloni(2004)}]{fender:04a}
482: Fender, R., \& Belloni, T. 2004, ARA\&A, 42, 317
483:
484: \bibitem[{{Fender} {et~al.}(2004){Fender}, {Belloni}, \& {Gallo}}]{fender:04b}
485: {Fender}, R.~P., {Belloni}, T.~M., \& {Gallo}, E. 2004, MNRAS, 355, 1105
486:
487: \bibitem[{Fender {et~al.}(2006)Fender, Stirling, Spencer, Brown, Pooley,
488: Muxlow, \& {Miller-Jones}}]{fender:06a}
489: Fender, R.~P., Stirling, A.~M., Spencer, R.~E., Brown, I., Pooley, G.~G.,
490: Muxlow, T. W.~B., \& {Miller-Jones}, J. C.~A. 2006, MNRAS, 369, 603
491:
492: \bibitem[{Gallo {et~al.}(2005)Gallo, Fender, Kaiser, Russell, Morganti,
493: Oosterloo, \& Heinz}]{gallo:05a}
494: Gallo, E., Fender, R., Kaiser, C., Russell, D., Morganti, R., Oosterloo, T., \&
495: Heinz, S. 2005, Nature, 436, 819
496:
497: \bibitem[{Gallo {et~al.}(2003)Gallo, Fender, \& Pooley}]{gallo:03a}
498: Gallo, E., Fender, R.~P., \& Pooley, G.~G. 2003, MNRAS, 344, 60
499:
500: \bibitem[{Gies \& Bolton(1982)}]{gies:82a}
501: Gies, D.~R., \& Bolton, C.~T. 1982, ApJ, 260, 240
502:
503: \bibitem[{Gleissner {et~al.}(2004)Gleissner, Wilms, Pooley, Nowak, Pottschmidt,
504: Markoff, Heinz, Klein-Wolt, Fender, \& Staubert}]{gleissner:04a}
505: Gleissner, T., Wilms, J., Pooley, G.~G., Nowak, M.~A., Pottschmidt, K.,
506: Markoff, S., Heinz, S., Klein-Wolt, M., Fender, R.~P., \& Staubert, R. 2004,
507: A\&A, 425, 1061
508:
509: \bibitem[{Hjellming(1973)}]{hjellming:73a}
510: Hjellming, R.~M. 1973, ApJ, 182, L29
511:
512: \bibitem[{{Hjellming} \& {Johnston}(1988)}]{hjellming:88a}
513: {Hjellming}, R.~M., \& {Johnston}, K.~J. 1988, ApJ, 328, 600
514:
515: \bibitem[{Jahoda {et~al.}(2005)Jahoda, Markwardt, Radeva, Rots, Stark, Swank,
516: Strohmayer, \& Zhang}]{jahoda:05a}
517: Jahoda, K., Markwardt, C.~B., Radeva, Y., Rots, A.~H., Stark, M.~J., Swank,
518: J.~H., Strohmayer, T.~E., \& Zhang, W. 2005, ApJS, 163, 401
519:
520: \bibitem[{{Klein-Wolt} {et~al.}(2002){Klein-Wolt}, {Fender}, {Pooley},
521: {Belloni}, {Migliari}, {Morgan}, \& {van der Klis}}]{kleinwolt:02a}
522: {Klein-Wolt}, M., {Fender}, R.~P., {Pooley}, G.~G., {Belloni}, T., {Migliari},
523: S., {Morgan}, E.~H., \& {van der Klis}, M. 2002, MNRAS, 331, 745
524:
525: \bibitem[{{Maccarone}(2002)}]{maccarone:02d}
526: {Maccarone}, T.~J. 2002, MNRAS, 336, 1371
527:
528: \bibitem[{{Markoff} \& {Nowak}(2004)}]{markoff:04a}
529: {Markoff}, S., \& {Nowak}, M.~A. 2004, ApJ, 609, 972
530:
531: \bibitem[{Markoff {et~al.}(2005)Markoff, Nowak, \& Wilms}]{markoff:05a}
532: Markoff, S., Nowak, M.~A., \& Wilms, J. 2005, ApJ, 635, 1203
533:
534: \bibitem[{{Marscher} {et~al.}(2002){Marscher}, {Jorstad}, {G{\'o}mez}, {Aller},
535: {Ter{\"a}sranta}, {Lister}, \& {Stirling}}]{marscher:02a}
536: {Marscher}, A.~P., {Jorstad}, S.~G., {G{\'o}mez}, J.-L., {Aller}, M.~F.,
537: {Ter{\"a}sranta}, H., {Lister}, M.~L., \& {Stirling}, A.~M. 2002, Nature,
538: 417, 625
539:
540: \bibitem[{{Miller-Jones} {et~al.}(2006){Miller-Jones}, Fender, \&
541: Nakar}]{millerjones:06a}
542: {Miller-Jones}, J. C.~A., Fender, R.~P., \& Nakar, E. 2006, MNRAS, 367, 1432
543:
544: \bibitem[{Miller-Jones {et~al.}(2005)Miller-Jones, {McCormick}, Fender,
545: Spencer, Muxlow, \& Pooley}]{millerjones:05a}
546: Miller-Jones, J. C.~A., {McCormick}, D.~G., Fender, R.~P., Spencer, R.~E.,
547: Muxlow, T. W.~B., \& Pooley, G.~G. 2005, MNRAS, 363, 867
548:
549: \bibitem[{Mirabel {et~al.}(1998)Mirabel, Dhawan, Chaty, {Rodr\'\i{}guez},
550: {Mart\'\i}, Robinson, Swank, \& Geballe}]{mirabel:98a}
551: Mirabel, I.~F., Dhawan, V., Chaty, S., {Rodr\'\i{}guez}, L.~F., {Mart\'\i}, J.,
552: Robinson, C.~R., Swank, J., \& Geballe, T.~R. 1998, A\&A, 330, L9
553:
554: \bibitem[{{Mirabel} \& {Rodr\'\i{}guez}(1994)}]{mirabel:94a}
555: {Mirabel}, I.~F., \& {Rodr\'\i{}guez}, L.~F. 1994, Nature, 371, 46
556:
557: \bibitem[{Petrinec {et~al.}(2000)Petrinec, {McKenzie}, Imhof, Mobilia, \&
558: Chenette}]{petrinec:00a}
559: Petrinec, S.~M., {McKenzie}, D.~L., Imhof, W.~L., Mobilia, J., \& Chenette,
560: D.~L. 2000, J.\ Atmos.\ Solar Terr.\ Phys., 62, 875
561:
562: \bibitem[{Pooley \& Fender(1997)}]{pooley:97a}
563: Pooley, G.~G., \& Fender, R.~P. 1997, MNRAS, 292, 925
564:
565: \bibitem[{Rothschild {et~al.}(1998)Rothschild, Blanco, Gruber, Heindl,
566: {MacDonald}, Marsden, Pelling, Wayne, \& Hink}]{rothschild:98a}
567: Rothschild, R.~E., Blanco, P.~R., Gruber, D.~E., Heindl, W.~A., {MacDonald},
568: D.~R., Marsden, D.~C., Pelling, M.~R., Wayne, L.~R., \& Hink, P.~L. 1998,
569: ApJ, 496, 538
570:
571: \bibitem[{{Rothstein} {et~al.}(2005){Rothstein}, {Eikenberry}, \&
572: {Matthews}}]{rothstein:05a}
573: {Rothstein}, D.~M., {Eikenberry}, S.~S., \& {Matthews}, K. 2005, ApJ, 626, 991
574:
575: \bibitem[{Sazonov {et~al.}(2007)Sazonov, Churazov, Sunyaev, \&
576: Revnivtsev}]{sazonov:07a} Sazonov, S., Churazov, E., Sunyaev, R.,
577: \& Revnivtsev, M. 2007, MNRAS, 377, 1726
578:
579: \bibitem[{Scargle(1989)}]{scargle:89a}
580: Scargle, J.~D. 1989, ApJ, 343, 874
581:
582: \bibitem[{{Stirling} {et~al.}(1998){Stirling}, {Spencer}, \&
583: {Garrett}}]{stirling:98a}
584: {Stirling}, A., {Spencer}, R., \& {Garrett}, M. 1998, New Astronomy Review, 42,
585: 657
586:
587: \bibitem[{Stirling {et~al.}(2001)Stirling, Spencer, {de la Force}, {Garrett},
588: {Fender}, \& {Ogley}}]{stirling:01a}
589: Stirling, A.~M., Spencer, R.~E., {de la Force}, C.~J., {Garrett}, M.~A.,
590: {Fender}, R.~P., \& {Ogley}, R.~N. 2001, MNRAS, 327, 1273
591:
592: \bibitem[{van~der Laan(1966)}]{vanderlaan:66a}
593: van~der Laan, H. 1966, Nature, 211, 1131
594:
595: \bibitem[{Wilms {et~al.}(2006)Wilms, Nowak, Pottschmidt, {et~al.}}]{wilms:05a}
596: Wilms, J., Nowak, M.~A., Pottschmidt, K., {et~al.} 2006, A\&A, 447, 245
597:
598: \end{thebibliography}
599:
600: \end{document}
601: