1: \documentclass[numberedappendix,apj]{emulateapj}
2: \usepackage{apjfonts} % Times fonts
3: %% manuscript produces a one-column, double-spaced document:
4: %\documentclass[manuscript]{aastex}
5: %% preprint2 produces a double-column, single-spaced document:
6: %\documentclass[12pt,preprint]{aastex}
7: %% \documentclass[preprint2,longabstract]{aastex}
8:
9: \newcommand{\vdag}{(v)^\dagger}
10: \newcommand{\myemail}{Matthias.Kadler@nasa.gov}
11:
12: \makeindex
13: \citeindextrue
14:
15: \shorttitle{The Trails of Superluminal Jet Components in 3C\,111}
16: \shortauthors{Kadler et al.}
17: \received{2007 September 26}
18: \accepted{2007 December 19}
19: \journalinfo{Astrophysical~journal}
20: \submitted{}
21:
22: \begin{document}
23:
24: \title{The Trails of Superluminal Jet Components in 3C\,111}
25:
26: \author{M. Kadler\altaffilmark{1,2}
27: E. Ros\altaffilmark{2},
28: M. Perucho\altaffilmark{2},
29: Y. Y. Kovalev\altaffilmark{2,3},
30: D. C. Homan\altaffilmark{4},
31: I. Agudo\altaffilmark{5,2},
32: K. I. Kellermann\altaffilmark{6},
33: M. F. Aller\altaffilmark{7},
34: H. D. Aller\altaffilmark{7},
35: M. L. Lister\altaffilmark{8},
36: and
37: J. A. Zensus\altaffilmark{2}
38: }
39: \altaffiltext{1}{Astrophysics Science Division, NASA's Goddard Space Flight Center, Greenbelt Road, Greenbelt, MD 20771, USA; \mbox{Matthias.Kadler@nasa.gov}}
40: \altaffiltext{2}{Max-Planck-Institut f\"ur Radioastronomie, Auf dem H\"ugel\,69, 53121 Bonn, Germany; ros, perucho, ykovalev, \mbox{azensus@mpifr-bonn.mpg.de}}
41: \altaffiltext{3}{Astro Space Center of Lebedev Physical Institute, Profsoyuznaya 84/32, 117997 Moscow, Russia}
42: \altaffiltext{4}{Astronomy Department, Department of Physics and Astronomy, Denison University, Granville, OH 43023, U.S.A.; \mbox{homand@denison.edu}}
43: \altaffiltext{5}{Instituto de Astrof\'{\i}sica de Andaluc\'{\i}a (CSIC), Apartado 3004, E-18080 Granada, Spain; \mbox{iagudo@iaa.es}}
44: \altaffiltext{6}{National Radio Astronomy Observatory, 520 Edgemont Road, Charlottesville, VA 22903, U.S.A; \mbox{kkellerm@nrao.edu}}
45: \altaffiltext{7}{Astronomy Department, University of Michigan, Ann Arbor, MI 48109-1042, U.S.A.; mfa, \mbox{haller@umich.edu}}
46: \altaffiltext{8}{Department of Physics, Purdue University, 525 Northwestern Avenue, West Lafayette, IN 47907, U.S.A.; \mbox{mlister@physics.purdue.edu}}
47:
48: \begin{abstract}
49: In 1996, a major radio flux-density outburst occured in the broad-line radio galaxy 3C\,111.
50: It was followed by a particularly bright plasma ejection associated with a superluminal
51: jet component, which has shaped the parsec-scale structure of 3C\,111 for almost a decade.
52: Here, we present results from 18 epochs of Very Long Baseline Array (VLBA) observations
53: conducted since 1995 as
54: part of the VLBA 2\,cm Survey and MOJAVE monitoring programs.
55: This major event allows us to study a variety of processes associated with outbursts of radio-loud
56: AGN in much greater detail than has been possible in other cases: the primary perturbation gives rise to
57: the formation of a leading and a following component, which are interpreted as
58: a forward and a backward-shock. Both components evolve in characteristically different ways
59: and allow us to draw conclusions about the work flow of jet-production events; the expansion, acceleration
60: and recollimation
61: of the ejected jet plasma in an environment with steep pressure and density gradients are revealed;
62: trailing components are formed in the wake of the primary perturbation possibly as a result of coupling to
63: Kelvin-Helmholtz instability pinching modes from the interaction of the jet with the external medium. The
64: interaction of the jet with its ambient medium
65: is further described by the linear-polarization signature of jet components traveling
66: along the jet and passing a region of steep pressure/density gradients.
67: \end{abstract}
68:
69:
70: \keywords{galaxies: individual: 3C111 --
71: galaxies: active --
72: galaxies: jets --
73: galaxies: nuclei }
74:
75: \section{Introduction}
76: Direct evidence for the existence of bulk relativistic outflows along the jets in
77: blazars and other radio-loud active galactic nuclei (AGN) comes from
78: Very-Long-Baseline Interferometry (VLBI) observations.
79: The first evidence for apparently superluminal
80: structural changes was found from changes in the fringe visibility curves
81: of 3C\,279 and 3C\,273 \citep{Whi71,Coh71}.
82: Subsequent higher-quality VLBI
83: observations \citep[see, e.g., compilation by ][and references therein]{Ver94} have established
84: the ``core-jet'' type milliarcsecond-scale structure of compact extragalactic jets:
85: the core being a bright and unresolved flat-spectrum component at the end of a linear
86: structure, and the jet being composed out of individual steep-spectrum components or ``knots''.
87: The knots frequently move away from the core with apparent velocities exceeding the
88: speed of light.
89: Monitoring observations of large source samples
90: \citep{Ver94,Jor01,Kel04,Pin07}
91: have provided important statistical tools for probing
92: relativistic beaming and the intrinsic properties of extragalactic radio jets \citep{Coh07},
93: their intrinsic brightness temperatures \citep{Hom06}, or their Lorentz factor
94: distribution \citep{Kel04} and luminosity function \citep{Car07}.
95:
96:
97: The relativistic-jet model \citep[e.g.,][]{Bla79}
98: has become the de-facto paradigm in multiwavelength research on blazars and other AGN,
99: but VLBI observations
100: have demonstrated that the basic concept of ballistically-moving isolated jet knots
101: is clearly oversimplified:
102: jet curvature \citep[e.g.,][]{Ver94}, stationary components \citep[e.g.,][]{Jor01}, and
103: non-radial and accelerated motions \citep[e.g.,][]{Kel04}, are found to be common
104: features of relativistic jets.
105: Within individual jets, there are often characteristic velocities
106: suggesting the presence of an underlying continuous jet flow, but the ``components''
107: themselves most likely represent patterns moving at a different speed than the underlying
108: flow, e.g., as hydrodynamically propagating shocks \citep{Mar85,Hug85}.
109:
110: Recent years have brought major improvements in numerical simulations of relativistic jets
111: \citep[see, e.g.,][for a review]{Gom05}.
112: It is now possible to simulate three-dimensional
113: relativistic jets \citep[e.g.,][]{Alo03} and to compute
114: the relativistic processes \citep[e.g.,][]{Gom97} that transfer hydrodynamic
115: results into observed brightness distributions (e.g., relativistic light abberation and
116: light travel time delays).
117: In particular,
118: interactions between strong perturbations or shocks with the underlying jet flow and the
119: jet-ambient medium can be simulated \citep{Agu01}.
120: With these new techniques, it is now possible to compare the generation, propagation and evolution
121: of emission features in simulated and observed relativistic jets.
122:
123: The nearby (z$=$0.049)\footnote{Assuming $H_0=71$\,km\,s$^{-1}$\,Mpc$^{-1}$,
124: $\Omega_{\rm M} = 0.3$, $\Omega_\Lambda = 0.7$ ($1{\rm \,mas} = 1.0{\rm \,pc}$).} broad-line radio galaxy 3C\,111 (PKS B\,0415+379)
125: shows a classical FR\,II morphology on kiloparsec-scales
126: spanning more than 200$^{\prime\prime}$ with a highly collimated jet connecting
127: the central core and the northeastern lobe in position angle 63$^\circ$ while
128: no counterjet is observed towards the southwestern lobe \citep{Lin84}. This asymmetry
129: is usually explained via relativistic boosting of the jet and de-boosting of
130: the counter-jet.
131: 3C\,111 exhibits the brightest compact radio core at cm/mm wavelengths of all
132: FR\,II radio galaxies, a blazar-like spectral energy distribution \citep{Sgu05}, and
133: it was one of the first (and only) radio galaxies in which superluminal motion
134: was detected \citep{Goe87,Pre88}.
135: Moreover, the (sub-) parsec
136: scale jet of 3C 111 is intimately related to its high-energy emission: \citet{Mar06}
137: reports a disk-jet connection, similar to the well-established one in 3C 120 \citep{Mar02}, in the sense that dips in the X-ray
138: light curve indicate accretion events which are followed by VLBI jet component ejections.
139: Recently, R.C.\,Hartman \& M. Kadler (in prep.) showed that
140: the gamma-ray source 3EG\,J0416+3650 can be decomposed into multiple individual
141: sources inside the EGRET full-band point-spread function,
142: revealing a significant signal from the nominal position of 3C\,111
143: in the higher-resolution, high-energy band above 1\,GeV.
144: This association of 3C\,111 with 3EG\,J0416+3650, which had originally been suggested
145: by \citet{Har99} and \citet{Sgu05},
146: makes 3C\,111 one of the very rare radio galaxies detected
147: at gamma-ray energies and supports the view that this
148: source may be considered a lower-luminosity version of powerful radio-loud quasars.
149:
150: Here, we report
151: the results from ten years of Very-Long-Baseline Interferometry (VLBI) observations of
152: 3C\,111 as part of
153: the VLBA 2\,cm Survey\footnote{\tt http://www.cv.nrao.edu/2cmsurvey/} \citep{Kel98,Zen02,Kel04,Kov05} and its follow-up program
154: MOJAVE\footnote{\tt http://www.physics.purdue.edu/MOJAVE} \citep{Lis05,Hom06b}.
155: We investigate the parsec-scale source
156: structure during a major flux-density outburst and during its aftermath.
157: We find that this outburst was associated with the formation of an exceptionally bright feature
158: in the jet of 3C\,111.
159: A variety of processes (beyond the predictions of simple ballistic motion models) are observed and
160: discussed in view of modern relativistic-jet simulations.
161: In Sect.~\ref{sect:obs}, our observations
162: and the data reduction are described. A detailed report of the observational results is given in
163: Sect.~\ref{sect:results}. In Sect.~\ref{sect:discussion}, we discuss the various
164: processes observed in the jet of 3C\,111 as a result of
165: the outburst and during the propagation of the new jet feature along the jet. In Sect.~\ref{sect:conclusions},
166: we put these results into the context of future simulations and observations
167: with the goal of understanding the production mechanisms of AGN jets.
168:
169: \section{Observations and data analysis}
170: \label{sect:obs}
171: 3C\,111 has been monitored as part of the VLBA 2\,cm Survey program since
172: April 1995.
173: The observational details are given by
174: \citet{Kel98}.
175: Following the methods described there,
176: the data from 17 epochs of VLBA observations of 3C\,111
177: between 1995 and 2005 (see Table~\ref{tab:3c111_journal})
178: were phase and amplitude self calibrated
179: and the brightness distribution was determined
180: via hybrid mapping.
181: An additional epoch from June 2000 was made available to us by
182: G. Taylor.
183: The polarization calibration was performed as described in \citet{Lis05}.
184: Two-dimensional Gaussian components were fitted in the $(u,v)$-domain
185: to the fully calibrated visibility data of each epoch
186: using the program {\sc difmap} \citep{She97}.
187: %We added new components to the model until the residual map was smooth and showed no significant
188: %features above the noise level.
189: The parameters of each model fit at the various epochs are
190: given in Table~\ref{tab:3c111-modelfits}.
191: The models were aligned by assuming the
192: westernmost component (namely, the ``core'') to
193: be stationary so that the position of jet components can be measured relative
194: to it.
195: Because of the coupling of the flux densities of nearby model components,
196: the uncertainties in the component flux densities are larger than the formal (statistical) errors
197: unless the given model component is far enough
198: from its closest neighbor. Throughout this paper, errors of
199: 15\,\% are assumed for the flux densities of individual model-fit
200: components.
201: In most cases, this should be considered a conservative estimate that accounts for absolute
202: calibration uncertainties and formal model-fitting uncertainties \citep[see, e.g.,][]{Hom02}.
203: Position uncertainties were determined
204: internally from the deviations of the data from linear motion.
205:
206:
207: \section{Results}
208: \label{sect:results}
209:
210: \begin{figure}[t]
211: \centering
212: \includegraphics[clip,angle=-90,width=\linewidth]{f1}
213: \caption{University of Michigan Radio Astronomy Observatory light curves of 3C\,111 at 4.8\,GHz, 8\,GHz, and
214: 14.5\,GHz. The shaded
215: areas indicate the ejection epochs of the individually labeled jet components
216: as discussed in Sect.~\ref{sect:vlbaresults}.
217: The lightest shading corresponds to minor ejections of the relatively weak
218: components I,J,K,L with flux densities $S$ below $0.2$\,Jy, medium shading corresponds to
219: components B,C,G,H,M,N with $0.2{\rm Jy} < S < 0.6 {\rm Jy}$ and the darkest shading to
220: components E and F with $S > 0.6 {\rm Jy}$.
221: }
222: \label{fig:3c111_umrao}
223: \end{figure}
224:
225: \begin{figure}[t]
226: \centering
227: \includegraphics[clip,angle=-90,width=\linewidth]{f2}
228: \caption{Spectral-index curves of 3C\,111 between 14.5\,GHz and 8\,GHz (top), and
229: 8\,GHz and 4.8\,GHz (bottom) from the UMRAO monitoring program. The shaded
230: areas are the same as in Fig.~\ref{fig:3c111_umrao}.
231: }
232: \label{fig:3c111_spit}
233: \end{figure}
234:
235: \subsection{The 1996 Radio Outburst of 3C\,111}
236:
237: \begin{figure*}[t!]
238: \centering
239: \includegraphics[clip, angle=-90,width=\linewidth]{f3}
240: \caption{
241: Naturally weighted images of the parsec-scale jet of 3C\,111 from the 2\,cm VLBA monitoring.
242: A common restoring beam of $(0.5 \times 1.0)$\,mas at P.A. $0^\circ$ was used.
243: The total recovered flux density in each image, the rms noise, and the lowest contours for each image are
244: given in Table~\ref{tab:3c111_journal}.
245: Contours increase logarithmically by a factor of $2$. Only components E, G, H and their
246: corresponding trailing components are indicated by circles enclosing a cross.
247: \label{fig:3c111_allepochs}
248: }
249: \end{figure*}
250:
251: A strong flux density outburst of 3C\,111 occurred in
252: 1996, which was first visible in the mm band and some months later at
253: lower radio frequencies. This outburst was first detected
254: at 90\,GHz with the IRAM interferometer at Plateau de Bure
255: in January 1996 with flux densities greater than $10$\,Jy \citep{Ale98},
256: at 37\,GHz in March 1996, and at 22\,GHz in August 1996 with the
257: Mets\"ahovi radio observatory \citep{Ter04}.
258: Figure~\ref{fig:3c111_umrao} shows the single-dish radio light curves of
259: 3C\,111 at 4.8\,GHz, 8\,GHz, and
260: 14.5\,GHz obtained from the UMRAO radio-flux-density monitoring program \citep{All03}.
261: These data show that from early 1996 on, the radio-flux density of 3C\,111
262: was rising at 14.5\,GHz, reaching its maximum in late 1996. At the two lower
263: frequencies, the flux-density maximum was reached at subsequent later times,
264: in mid 1997 at 8\,GHz and in late 1997 at 4.8\,GHz.
265: The profile of the outburst in the flux-density vs. time domain shows a narrow, high-amplitude
266: peak between early 1996 and late 1997, which is almost symmetric. After late 1997, a slower-decreasing
267: component dominates the light curves, most clearly visible at 14.5\,GHz.
268:
269: Figure~\ref{fig:3c111_umrao} shows that the flare propagated through
270: the spectrum as qualitatively expected by standard jet theory \citep[e.g.,][]{Mar85}: high-frequency
271: radio emission comes from the most compact regions of the jet, the emission
272: peak shifts to lower frequencies as a newly ejected jet component travels down
273: the jet and becomes optically thin. The peak flux density shifted with frequency
274: at about $10$\,GHz\,yr$^{-1}$.
275:
276: The evolution of the spectral index, $\alpha$ ($S\sim\nu^\alpha$), for (14.5/8.0)\,GHz and (8.0/4.8)\,GHz is shown in
277: Fig.~\ref{fig:3c111_spit}. Before 1996, the sampling was too sparse to derive the change
278: of the spectral index in the (14.5/8.0)\,GHz band. Between, 8.0\,GHz and 4.8\,GHz,
279: the spectral index was approximately $-0.7$ during the pre-1996 period. The radio flux-density outburst
280: in 1996 corresponded to a subsequent flattening of the spectrum with a maximum
281: spectral index, $\alpha \sim 0$,
282: in the (14.5/8.0)\,GHz band reached in mid 1996. In the post-outburst period
283: between 1998 and 2004, $\alpha$ was typically in the range $-0.5$ to $-0.7$ between 14.5\,GHz
284: and 8.0\,GHz and slightly steeper ($-0.7$ to $-0.9$) in the (8.0/4.8)\,GHz band. The overall
285: steeper spectral index at lower frequencies can be understood as the contribution of
286: optically thin
287: large-scale emission from the radio lobes of 3C\,111 to these single-dish light curves.
288:
289: \subsection{VLBA Monitoring Results}
290: \label{sect:vlbaresults}
291:
292: Figure~\ref{fig:3c111_allepochs} shows the variable parsec-scale structure of 3C\,111
293: at 18 different epochs
294: of VLBA observations between 1995.26 and 2005.73.
295: The variable source structure can be described by a classical one-sided
296: core-jet morphology in the first two epochs with typical velocities of the
297: outward moving jet components of about $1.4$ to $1.7$\,mas\,yr$^{-1}$ corresponding
298: to about $5$\,$c$. In 1996.82 a new jet component, even brighter than the core,
299: dominated the source structure. By 1997.19, this new component was even brighter ($\sim 3.4$\,Jy)
300: and in the following epochs
301: it traveled along the jet while it became gradually more stretched out along the jet-ridge line.
302:
303: \paragraph{Model Fitting:}
304: In Fig.~\ref{fig:3c111_rad}, the radial distance of the various model fit components from the core
305: is shown as a function of time.
306: The component identification was based on a comparison of the positions and
307: flux densities, and a linear regression of the distances from the core as a function
308: of time was used to determine the kinematics.
309: The derived component velocities are tabulated
310: in Table~\ref{tab:3c111_kinematics}.
311: The early outer jet components (A, B, C, D) of the 1995.26 epoch can be
312: traced over two to four epochs before their flux densities fall below the detection
313: threshold (compare Fig.~\ref{fig:3c111_fluxes}).
314: In late 1996 and early 1997, the source structure was dominated by the emission of the core
315: and the newly formed jet components E and F, with E being the leading component. The two components
316: traveled outwards with a mean apparent velocity of $(1.00 \pm 0.02)$\,mas\,yr$^{-1}$
317: and $(0.64 \pm 0.07)$\,mas\,yr$^{-1}$, respectively.
318: Before mid 1997, component F was substantially brighter than component E but after that, its flux
319: density dropped steeply. F was not detected at any epoch later than 1998.18, while E was still
320: about $800$\,mJy at that time. The light curves of E and F reproduce qualitatively the
321: two-component shape of the flux-density outburst in Fig.~\ref{fig:3c111_umrao}
322: with component F being responsible for the narrower and higher-amplitude peak between early 1996 and
323: late 1997 and component E dominating the slower-decreasing tail of the outburst after late 1997
324: (compare Fig.~\ref{fig:3c111_fluxes} and discussion below).
325: In the following epochs, E split into four distinct components (E\,1, E\,2, E\,3, E\,4)
326: at distances of 3.5\,mas to 4.5\,mas from the core.
327: E\,2, E\,3, and E\,4 all moved at subsequently slower speeds than E\,1, resulting in an elongated
328: morphological structure of the associated emission complex.
329:
330: \begin{figure}
331: \centering
332: \includegraphics[clip,angle=-90,width=\linewidth]{f4}
333: \caption{Core separation of model-fit components vs. time.
334: Crosses represent components which could not be cross identified over the epochs.
335: Position uncertainties have been estimated from the internal deviations of the data
336: from linear motions for each component. The uncertainties for the position of E\,1 have been determined
337: separately for the pre-2004 and post-2004 epochs because of the partial resolution of
338: this component after 2004. Uncertainties for components with less than
339: three epochs were estimated from other components at similar positions in the jet and with similar flux
340: densities.
341: }
342:
343: \label{fig:3c111_rad}
344: \end{figure}
345:
346: \begin{figure}
347: \centering
348: \includegraphics[clip,angle=-90,width=\linewidth]{f5}
349: \caption{Flux-density evolution of the core and the jet components with time.
350: For clarity, only components ejected before 2001.50 are shown.
351: The flux densities of E\,1, E\,2, E\,3, and E\,4 were added for the post-1999 epochs and
352: a flux-density weighted effective position was calculated
353: to display the flux-density evolution of the blended feature
354: that would be visible at lower resolution.
355: Note that components E\,4 and G are blended
356: in epoch 2004.27 and that the flux density of E\,4 may be overestimated for this epoch.
357: }
358:
359: \label{fig:3c111_fluxes}
360: \end{figure}
361:
362: In later epochs, new components have been ejected from the core
363: into the jet. The two strongest components
364: (G,H) can be traced through the following eight and nine monitoring epochs,
365: respectively. Component H split into three individual components in 2004.27 and a fourth associated
366: component was seen from 2004.80 on. In the following, we refer to the components E\,1, and H\,1 as the
367: ``leading components'' and to E\,2, E\,3, E\,4, H\,2, H\,3, and H\,4 as the ``trailing components''
368: of E and H, respectively.
369:
370: \begin{figure}[t!]
371: \centering
372: \includegraphics[clip,angle=-90,width=\linewidth]{f6}
373: \caption{Flux-density evolution of component E and its trailing components vs. the distance
374: traveled from the core.
375: Note that component E\,4 is blended with component G
376: in epoch 2004.27 and that its flux density may be overestimated for this epoch.
377: }
378: \label{fig:3c111_fluxes2a}
379: \end{figure}
380: \begin{figure}[h!]
381: \centering
382: \includegraphics[clip,angle=-90,width=\linewidth]{f7}
383: \caption{Flux-density evolution of component G and H and its trailing components vs. distance traveled from the core.
384: }
385: \label{fig:3c111_fluxes2b}
386: \end{figure}
387:
388: %A connection between total flux density outbursts and VLBI component ejections
389: %in extragalactic flat-spectrum radio sources is well established \citep[e.g.,][and references therein]{Sav02}.
390: For the pc-scale jet of 3C\,111, the
391: ejection epochs of the individual jet components can be determined
392: from the linear regression by
393: back-extrapolating the component trajectories to the core. In Fig.~\ref{fig:3c111_umrao}, these ejection
394: epochs and the associated uncertainties are indicated as shaded areas. It is apparent that the ejection of
395: the components E and F coincides with the onset of the major flux-density outburst in 1996 described above.
396: The following major component ejections (G, H, and the combined M/N event) all have direct counterparts
397: in local maxima of the radio light curve, especially at 14.5\,GHz.
398: Figure~\ref{fig:3c111_spit} shows that all these ejection epochs coincided with local maxima of the
399: spectral index in the 14.5/8.0\,GHz band.
400: Between 2002 and 2004, a number of minor
401: component ejections took place but the regression-fit quality (due to the neareby components, the low
402: flux densities and the small time baseline) only moderatly constrains the ejection epochs. In addition,
403: the time sampling of UMRAO observations in this time range is relatively poor, in particular from mid 2001 to
404: mid 2003.
405:
406: \begin{figure}[t!]
407: \centering
408: \includegraphics[clip,angle=-90,width=\linewidth]{f8}
409: \caption{Brightness temperatures of model-fit components as a function of their distance to the core. The brightness temperatures
410: of components belonging to the E--, F-- and H--components are indicated by filled black circles. The solid line represents a least-squares
411: fit to all but the E--, F-- and H--components. The slope of the regression curve is $-2.4 \pm 0.2$.
412: }
413: \label{fig:3c111_tb}
414: \end{figure}
415: \begin{figure}[h!]
416: \centering
417: \includegraphics[clip,angle=-90,width=\linewidth]{f9}
418: \caption{Brightness temperatures of component E, its leading and trailing components,
419: and component F
420: as a function of their distance to the core. The two regimes of brightness-temperature gradient
421: discussed in the text are indicated with dashed lines.
422: }
423: \label{fig:3c111_tb_E}
424: \end{figure}
425:
426: \paragraph{Flux Density Evolution:}
427: Figure~\ref{fig:3c111_fluxes} shows the brightness evolution of the core and the jet components
428: that have been ejected prior to 2001.5.
429: Apparently, the trailing components E\,4 and H\,4 appeared first in a rising
430: state, i.e., they first increased in flux density before they became fainter in later epochs. Component
431: F showed an extraordinary steep decrease in brightness in 1997--1998.
432: %As mentioned before, not the leading component E but the following component F dominated the
433: %light curve in the first year after the outburst.
434: In Fig.~\ref{fig:3c111_fluxes2a} and Fig.~\ref{fig:3c111_fluxes2b}, the flux-density evolution of the components
435: E, G, and H and the associated leading and trailing components
436: are shown with distance traveled from the core, respectively.
437: The ejecta first rose in flux density within the inner 1\,mas from the core, then they showed a decline
438: about almost three orders of magnitude in the following decade, exhibiting a plateau or broad local maximum
439: in 1998--2000 at a distance from 2--4\,mas from the core.
440: Component H and its leading and trailing components exhibited a similar behavior although on about an order of magnitude
441: lower flux-density levels and at slightly further downstream, 4--6\,mas from the core.
442: Component G, in spite of the fact that it does not appear to have split into leading and trailing
443: components like E and H, did exhibit a pronounced flux density maximum after 2002, as well,
444: about 6\,mas from the core.
445:
446: \paragraph{The T$_{\rm b}$ gradient along the jet:}
447: Following \citet{Kad04,Kad05}, the power-law index $s$, which describes the brightness
448: temperature gradient via $T_{\rm b} \propto r^{s}$, can be parametrized as
449: \begin{equation}\label{eq:s}
450: s\,=\,l\,+\,n\,+\,b\,(1\,-\,\alpha)
451: \end{equation}
452: where $l$, $n$ and $b$ are the power law indices that describe the
453: gradients of jet diameter $d \propto r^l$, particle density $n_e \propto r ^n$, and
454: magnetic field $B \propto r^b$ with distance $r$ from the core, respectively.
455: Therefore, measuring the brightness temperature gradient
456: provides a method to constrain the critical physical properties
457: along the jet and abrupt changes in the $T_{\rm b}$-gradient
458: can highlight regions in the jet where the density, magnetic field, or jet diameter
459: change rapidly.
460: %Conservation of magnetic energy along the jet
461: %implies a constant energy partition along the jet:
462: %$B^2 \propto n_e$, $\Rightarrow b = n/2$.
463: %With $\alpha = -0.5$, it follows
464: %\begin{equation}\label{eq:s}
465: %s\,=\,l\,+\,7/4\,n \quad .
466: %\end{equation}
467:
468: Figure~\ref{fig:3c111_tb} shows the brightness temperatures of all jet components
469: in the parsec-scale jet of 3C\,111 at 2\,cm wavelength between 1995 and 2005 as
470: a function of their core distance. In general, the brightness temperature of all components
471: decreased as the components traveled outwards but an approximation with a simple power
472: law does not yield a good fit to the full data set ($\chi_{\rm red}^2=1.8$, 115 degrees of freedom [d.o.f.]).
473: Visual inspection of Fig.~\ref{fig:3c111_tb} shows that this is due to the E--, F-- and H-- components
474: and their leading and trailing components, respectively.
475: This behavior is different than
476: expected for a straight and stable jet geometry in which the power-law dependences of the particle
477: density, the magnetic field strength and the jet diameter on the core distance predicts
478: that the brightness temperature along the jet can be described with a
479: well-defined power-law index $s$. Most extragalactic parsec-scale jets
480: which do not show pronounced curvature, show a power-law decrease with
481: increasing distance from the core and power-law indices typically around $-2.5$
482: \citep{Kad05}.
483: In fact, excluding the E--, F-- and H--components from the fit yields a statistically better
484: result ($\chi_{\rm red}^2=1.3$, 52 d.o.f.) and a gradient of $-2.4 \pm 0.2$.
485: In Sect.~\ref{sect:smaller1mas} and Sect.~\ref{sect:3-5mas}, we discuss possible
486: physical reasons for the different behavior and nature of these components.
487: The measured relation between component sizes and distance along the jet is
488: affected by a large degree of
489: scatter and does not provide independent information from the flux
490: density and
491: brightness temperature plots described above. Therefore we do not
492: show plots of component size versus jet distance.
493:
494: The brightness-temperature gradient of component E was first flat or inverted immediately after
495: the creation of this new component within approximately 1\,mas from the core
496: and then reached steep values of $-2.5$ to $-2.8$ (regime I; compare Fig.~\ref{fig:3c111_tb_E})
497: through 1997 when the component
498: traveled from 1\,mas to 2\,mas.
499: Between 2\,mas and 4\,mas, the determination of the brightness-temperature gradient requires
500: an identification of component E with either component E\,1 or E\,3 (see below).
501: Independently of this identification, the brightness-temperature
502: gradient eventually changed to very steep values ($s<-5$) beyond 5\,mas from the core (regime II).
503: Component F began its very rapid decline in brightness temperature at a very small distance
504: from the core ($< 0.7$\,mas) with an extremely steep $T_{\rm b}$-gradient ($s <-8$).
505:
506: \begin{figure}[t!]
507: \centering
508: \includegraphics[trim=1.3cm 0cm 0cm 0cm,width=\linewidth]{f10}
509: \caption{Naturally weighted images of the linear-polarization structure of 3C\,111
510: between 2002 and 2005.
511: The restoring-beam dimensions and orientations
512: for each epoch are indicated by a cross to the left of each
513: Stokes-I image.
514: Stokes I contours start at 1\,mJy/beam and increase by factors of
515: $2$. Fractional polarization is over-plotted on the Stokes-I contours in
516: color.
517: To the right of the Stokes-I images are the polarization
518: intensity contours starting at 1\,mJy/beam and increasing by factors
519: $\sqrt{2}$. The polarization contours are over-plotted with
520: tick-marks representing the electric vector position angle. A single
521: Stokes-I 1\,mJy contour
522: surrounds the polarization image to show registration.
523: The dotted line marks the distance of 3.3\,mas from the core
524: where the most pronounced changes of the polarization properties take place
525: (see text).
526: }
527: \label{fig:3c111_pol}
528: \end{figure}
529:
530: \paragraph{Linear Polarization:}
531: From 1995 to 2002, 2\,cm-Survey observations were done in
532: left circular polarization only, so no linear-polarization
533: information can be derived from these data.
534: MOJAVE observations (from 2002 on) are done in full-polarimetric mode.
535: %and we have supplemented our 3C\,111 VLBA 2\,cm data with another
536: %full-polarimetric data set from October 2002 (PI: Taylor).
537: Figure~\ref{fig:3c111_pol} shows our polarization
538: data through September 2005.
539: %We do not show the May 2006 image
540: %because of its poor polarization sensitivity.
541:
542: In October of 2002, component H was about 2.5 mas from the base of
543: the jet and showed a fractional polarization of 5\% to 10\% increasing
544: towards the downstream side of the component. The electric
545: vector position angle (EVPA) displayed by the component was approximately
546: aligned with the jet.
547: In this epoch, the jet material just downstream of
548: component H
549: at $\sim 3.3$\,mas from the base of the jet
550: was more highly polarized, exceeding 20\% fractional
551: polarization on the jet's southern side, and the EVPA of the
552: polarization turned to be about 45$^\circ$ to the main jet
553: direction.
554:
555: By August of 2003, component H had entered a region approximately $3.3$\,mas from the core
556: and its observed polarization was now similar to
557: the emission in this same region observed in the previous epoch.
558: The observed fractional polarization of H now climbed sharply to values in
559: excess of 20\% toward the jet's southern side
560: while there was no detectable polarization from the northern
561: side of H. The observed EVPA of H had rotated further to be approximately
562: 60$^\circ$ to the main jet direction. However, the EVPA
563: on the southern-most side of H was
564: approximately perpendicular to the jet.
565:
566: After component H passed through this region (epochs April 2004
567: through September 2005), it split into a number of subcomponents as
568: described earlier, and its polarization gradually became more
569: uniform. Consistent fractional polarization of 5\% to 10\% was approached
570: with the electric vectors approximately perpendicular to the local
571: jet direction.
572:
573: The much weaker component, I,
574: developed polarization very similar to H as it passed through
575: the same region, about 3.3 mas from the core, with fractional
576: polarization exceeding 20\% toward the southern side of the
577: jet with an EVPA at approximately 45$^\circ$ to the main jet axis. This
578: is also the same region of the jet in which component E had broken
579: up into a number of sub-components. In future epochs, we
580: will have the opportunity to follow component K as it passes
581: through this same region.
582:
583: \section{Discussion}
584: \label{sect:discussion}
585: In this section, we discuss the
586: aftermath of the major outburst in 3C\,111 in 1996
587: and the following component ejections through 2005.
588: We organize the subsections of our discussion
589: according to the downstream distance from the VLBI core
590: where we observe the effect of interest.
591:
592: \subsection{Within $1$\,pc: Forward and Reverse Structures}
593: \label{sect:smaller1mas}
594: Numerical simulations (\citealt{Alo03}) show that an abrupt
595: perturbation of the fluid density at the jet injection point
596: during a short time propagates downstream, evolves spreading
597: asymmetrically along the jet and finally splits into two distinct
598: regions. Both of these two regions have enhanced energy density
599: with respect to the underlying jet, and they emit synchrotron
600: radiation. The leading (forward shock) and the following region (reverse shock)
601: have higher and lower Lorentz factors, respectively,
602: than the underlying jet.
603: Thus, they should separate with time as they propagate downstream in
604: the jet.
605:
606: Component F matches the description of a backward moving wave
607: associated with the major injection into the jet of 3C\,111 after
608: the flux-density outburst of 1996. It follows the trail of
609: component E but at a lower speed. If
610: component F is identified with a reverse shock and component E with a forward
611: shock, it is possible then to compute the size of the shocked
612: region \citep{Per07}. In
613: 1996.82 and 1997.19, E and F were both very bright and separated by
614: only $\sim 0.3$\,pc in projected distance. During these two
615: epochs, F was 300\,mJy to 500\,mJy brighter than the leading
616: component E. Following \citet{Alo03}, a backward shock can be
617: brighter than a forward shock if the latter is beamed in a cone
618: smaller than the viewing angle due to its larger speed. We have
619: examined the Doppler factors of components E and F for the range of
620: possible viewing angles (see Appendix~\ref{sect:inclination})
621: and the measured
622: velocities and conclude that this alone cannot explain
623: the brightness difference between component E and F
624: because the difference in apparent speed is not large enough.
625: \citet{Jor05} point out that backward
626: shocks can be brighter than forward shocks as long as the
627: disturbance is prolonged and there is a continuous supply of
628: particles entering from the underlying jet through the shock region. Within
629: half a year, between 1997.19 and 1997.66, F lost about half of
630: its brightness. This extraordinarily fast dimming of the backward
631: shock can be caused by the lack of input of particles from behind,
632: i.e., a lower plasma ejection rate after the primary injection
633: possibly due to a depletion of the inner accretion disk \citep{Mar02,Mar06} which feeds
634: the plasma injection.
635:
636: Component F can also be interpreted as a rarefaction propagating
637: backwards in the reference frame of the ejected blob of gas. A
638: rarefaction is produced when the blob is overpressured
639: with respect to the jet, as this overpressure causes the front to
640: accelerate in the jet, thus leaving a rarefied region between the
641: head of the blob (forward shock) and its rear part, which is still
642: slower (it moves with the injection velocity). In this case, the
643: emission in component F could be associated to the denser and
644: overpressured gas in the blob which has still not been rarefied.
645: This gas would cease to emit as soon as it reaches the
646: rarefied region, which may also explain the sudden decrease in
647: brightness of this component. An extended discussion on the nature
648: of component F and the evolution of its brightness will be given
649: in \citet{Per07}.
650:
651:
652: %E and F are traveling with $\sim 1$\,mas\,yr$^{-1}$ and
653: %$\sim 0.64$\,mas\,yr$^{-1}$, respectively. If the velocity of the
654: %underlying flow is in the middle between the speed of the
655: %forward shock and the backward shock and if the backward shock passes the
656: %material from the underlying flow with a relative inward speed of $\sim 0.19$\,mas\,yr$^{-1}$,
657: %this implies that the injection of material
658: %after the start of the outburst persisted for 7--8 weeks.
659:
660:
661:
662: \subsection{Between $2$\,pc and $4$\,pc: Expansion and Acceleration}
663: It is not \emph{a-priori} clear with which post-split-up
664: component the original feature E should be identified after 1999.
665: A natural identification would be the leading component E1 but that
666: requires an acceleration of this component (see Fig.~4) from
667: $\beta_{\rm app,E}=3.26\pm0.07$ to $\beta_{\rm app,E1}=5.5\pm0.1$
668: between 1998.18 and 1999.38. This may be interpreted in terms of
669: an expansion of the jet in a rarefied medium. Taking an angle to
670: the line of sight of $19^\circ$ (see
671: Appendix~\ref{sect:inclination}), the component would be accelerated
672: from $\beta=0.956$ ($\gamma\sim3.4$) to $\beta=0.995$
673: ($\gamma\sim10.3$). The increase of velocity is less at smaller
674: viewing angles.
675: %Identifying component H with component H1 also
676: %implies an acceleration of the former at about the same distance
677: %to the source, although, in this case, the flux density decay of
678: %component H1 with respect to the trailing components H2, H3 and H4
679: %remains to be understood.
680: %Thus, components may
681: %enter a region with lower density or a steeper density gradient
682: %between 2\,mas and 4\,mas from the core.
683:
684: An alternative model for the acceleration and brightening of component E
685: would be a change of the jet inclination to the line of sight from about 24$^\circ$ to about 11$^\circ$
686: at this location in the jet as observed in the case of the quasar 3C\,279 \citep{Hom03}.
687: However, we see no significant change in jet position angle in the sky which would be expected
688: to accompany such a large change in jet direction. Moreover, subsequent components,
689: particularly G, do not show the same kind of large acceleration in this region.
690:
691: Direct identification of component E with component E1 is not
692: straightforward in the frame of expansion, as component E1 in
693: epoch 1999.38 was smaller than component E in 1998.18 (see
694: Table~2). However, component E3 in epoch 1999.38 is larger than
695: component E in 1998.18. We can interpret this as component E
696: including components E1 and E3 (and maybe E4). These components
697: would be indistinguishable in our observations before 1999.38.
698: In fact, \cite{Jor05}
699: monitored 3C\,111 between 1998 and 2001 with
700: the VLBA at 43GHz. They find an emission complex, that can be
701: identified with our component E, that gradually stretches out as
702: it travels from $\sim$ 2 mas from the core in 1998 to roughly between $5$\,mas and $8$\,mas
703: from the core in 2001. Their leading component C1 can be
704: identified with our component E1, their component c2 with E2 and
705: their c1 with E3.
706: At their higher angular resolution, Jorstad et al.\nocite{Jor05} can separate
707: components C\,1 and c\,1 already in early 1998.
708: In agreement with our analysis at 15\,GHz, they
709: detect c2 (E2) about a year after they detect c1 (E3).
710: They do not detect a component
711: corresponding to E4 but this may be an effect of partially resolving out the
712: jet structure at their higher observing frequency, particularly in later epochs.
713: It is
714: further interesting to note that the observed speeds at both frequencies agree well.
715: For E1(C1), $\mu_{\rm app, 2\,cm} = 1.69 \pm 0.04$\,mas\,yr$^{-1}$ at 15 GHz and
716: $\mu_{\rm app, 7\,mm} = 1.77 \pm 0.06$\,mas\,yr$^{-1}$ at 43 GHz;
717: for E2(c2), $\mu_{\rm app, 2\,cm} = 1.29 \pm 0.06$\,mas\,yr$^{-1}$ at 15 GHz and
718: $\mu_{\rm app, 7\,mm} = 1.23 \pm 0.04$\,mas\,yr$^{-1}$ at 43 GHz;
719: and for E3(c1),
720: $\mu_{\rm app, 2\,cm} = 1.22 \pm 0.05$\,mas\,yr$^{-1}$ at 15 GHz and
721: $\mu_{\rm app, 7\,mm} = 1.07 \pm 0.02$\,mas\,yr$^{-1}$ at 43 GHz.
722: The discrepancy in the speeds measured for E3 and c1 seems to be due to a
723: slight acceleration of E3 after 2002. A fit to the 15 GHz data of
724: E3 between 1999 and 2002 alone yields a slower speed of $\sim$ 1.0
725: mas\,yr$^{-1}$ similar to the speed of c1 in the same time period at
726: 43\,GHz. In their work, Jorstad et al.\nocite{Jor05} do not report acceleration of components
727: from 2\,mas to 4\,mas. However, this is likely due to the fact
728: that their observations started in early 1998, thus missing the
729: first observations of component E presented in this paper, when
730: its speed has been measured to be smaller.
731:
732: \subsection{Between $2$\,pc and $6$\,pc: Recollimation of the Jet}
733: \label{sect:recollimation}
734: Inspection of Fig.~9 shows that the back-extrapolation of the
735: brightness temperature of component E from regime II to regime I is at
736: least two orders of magnitude too high if this
737: extrapolation is based on the gradient given by E1.
738: The low brightness temperature of component E in regime I cannot be
739: explained by opacity effects because the radio-light curve in
740: Fig.~1 shows that the source was optically thin
741: from 1997 on. Moreover, if we identify component E with E1, it is
742: Doppler-deboosted from epoch 1998.18 to 1999.38 due to the
743: acceleration and a relatively large viewing angle; thus, we are
744: not able to explain the increase in brightness temperature in
745: terms of Doppler boosting.
746: However, compact sub-components may have larger brightness temperatures, so that the
747: $T_{\rm b}$ values plotted in Fig.~9 for E in regime I inward of about $3$\,mas
748: may represent lower limits for
749: compact components already embedded in the unresolved structure.
750:
751: Not only E/E1 but also components G and H show an increase in
752: total flux density several milliarcseconds downstream. Compared to
753: E/E1, these somewhat weaker components exhibit their flux-density
754: maxima at somewhat larger distances from the core (compare Fig.~6
755: and Fig.~7).
756: This can be explained if the gas in the components travels through a
757: mild standing shock in a recollimation region.
758: This effect has been
759: observed in numerical simulations of parsec \citep{Gom97}
760: and kiloparsec scale jets \citep{Per07b}.
761: The
762: material in the components is expected to be overpressured with
763: respect to its environment, thus expanding into it. After the
764: initial expansion, the components become underpressured with
765: respect to the underlying flow. The resulting recollimation leads
766: to the formation of a shock, whose strength depends on the initial
767: degree of overpressure of the material in the component. This
768: process explains the increase in flux density and brightness
769: temperature as due to compression of the gas in the recollimation.
770: In Figs.~6 and 7, we see that the flux density of component E
771: increases closer to the core
772: than for component G and H, which is consistent with the former being slower
773: than the latter, thus recollimating earlier \citep[see][]{Per07b}.
774: It also
775: explains why we see a significant acceleration only in the faster
776: expanding, brighter component E/E1.
777:
778: Finally, after this mild recollimation, the fluid becomes
779: overpressured with respect to its environment, thus further
780: expanding and accelerating downstream.
781:
782: %\subsection{An Alternative Scenario: Elastic Deflection between 2\,pc and 4\,pc}
783: %The acceleration of component E can also be interpreted on the
784: %basis of an elastic-deflection event that would produce an
785: %apparent acceleration of the component due to the jet approaching
786: %the observer's line of sight.
787: %Such an event has been seen in the jet of quasar 3C\,279 \citep{Hom03}
788: %where a prominent change in the (projected) trajectory of a bright jet component
789: %could be explained by an intrinsically small change of the jet-angle to the
790: %line of sight.
791: %However, in the case of 3C\,111, a large change in the angle
792: %to the line of sight is implied: from about $24^{\circ}$ to about $11^{\circ}$ \citep{Kad05}. Considering the lack of a pronounced change in the
793: %component position angle, i.e., the observed angle projected onto
794: %the the plane of the sky, this deflection would have to happen
795: %nearly perpendicular to the projection plane,
796: %which is clearly an unlikely scenario.
797: %Moreover, such a large deflection should produce a strong shock with the
798: %ambient medium resulting in a strong enhancement of the brightness
799: %in this region.
800: %And finally, subsequent components, particularly G, do
801: %not show the same kind of large acceleration in this region.
802: %The
803: %elastic-deflection scenario, however, is capable of explaining the
804: %"bump" in the brightness-temperature gradient between 3\, and 5\,mas
805: %(see Fig.~9) via differential Doppler boosting associated with
806: %the change in the angle to the line of sight. A de-projection
807: %according to the elastic-deflection scenario results to a smooth
808: %brightness-temperature gradient of approximately $-3.75$ \citep{Kad05}.
809:
810:
811: \subsection{Near $3$\,pc: The Role of the External Medium}
812: \label{sect:polarization}
813: The polarization behavior of components H and I can be
814: understood in terms of
815: an interaction between the jet and the external medium at a
816: distance of ~3.3\,mas ($= 3.3$\,pc) in the jet.
817: Assuming no Faraday rotation, the
818: EVPA of component H within approximately 3.3\,mas from the core
819: indicates a transverse magnetic field order
820: as might be expected for a transverse shock propagating down the
821: jet. The change in the
822: fractional polarization, its north-south gradient, and the rotation of the EVPA
823: suggest that a contact surface persists at the
824: southern boundary of the jet beam at a
825: distance of approximately 3.3\,mas downstream the jet core.
826: If the bulk jet material flows faster than
827: the flow at the southern boundary, the magnetic field is stretched
828: through shear.
829: Our overall picture then is of an originally transverse
830: shock interacting with the jet on the southern side of the
831: jet at ~ 3.3\,mas from the core. The interaction changes
832: the component's magnetic field through some combination
833: of oblique shock and differential flow resulting in a
834: magnetic field approximately parallel to the jet axis in
835: the later epochs.
836: No strong shock is needed at this location
837: in the jet
838: but this region may be identified with the recollimation
839: region (see Sect.~\ref{sect:recollimation}) at about the
840: same position in the jet).
841: In this picture, the jet-medium interaction
842: may form an effective nozzle which accelerates the jet on
843: one edge relative to the other.
844:
845: An alternative explanation for the observed polarization structure and dynamics of 3C\,111
846: can be found by considering an inhomogeneous external Faraday screen. Such a screen could
847: produce the observed differential rotation of the EVPA while a component travels
848: through a given region along the jet. \cite{Zav02} observed 3C\,111 with the VLBA
849: and produced a Faraday rotation-measure map between 8\,GHz and 15\,GHz. They find strong
850: Faraday rotation, $\sim 730$\,rad\,m$^{-2}$, at the same distance from the core
851: (3.3\,mas) where our observations show the swing of the EVPA
852: of the component H and steeply decreasing Faraday rotation further downstream.
853: However, we note that $730$\,rad\,m$^{-2}$ translates to 17$^\circ$ of rotation at 15\,GHz
854: which alone is not enough to explain the change in EVPA that we observe while
855: component H travels through this region. On the other hand, the steep decrease
856: of the Faraday rotation measured up- and downstream of this region by \citet{Zav02} again
857: agrees with a change of the external gas density at this point, which in turn
858: may be identified with the pressure gradient responsible for the
859: component expansion and accelleration.
860:
861: A combination of inhomogeneous Faraday rotation and an interaction between the jet plasma and its ambient medium appears most likely to explain our observations of the varying linear polarization structure;
862: however, both explanations point to the role of the external medium, either through a discrete interaction or a rapid decrease in external gas pressure, in shaping the jet flow downstream of this location.
863:
864: \subsection{Between $3$\,pc and $5$\,pc: Formation of Trailing Components}
865: \label{sect:3-5mas}
866: The components E2, E3, and E4 can be interpreted as
867: trailing components forming in the wake of the leading E1 which is
868: identical with the original component E. This scenario is
869: attractive because the basic concept of trailing components as
870: introduced by \citet{Agu01} predicts the formation of
871: trailing features in the wake of the initial perturbation in the
872: jet flow. Such a behaviour has first been found both associated with bright
873: sub- and superluminal jet components in Centaurus A and 3C 120 \citep{Tin01,Gom01}.
874: \citet{Jor05} report trailing components in four additional sources
875: (3C\,273, 3C\,345, CTA\,102 and 3C\,454.3) and in 3C\,111 (see below).
876:
877: The interaction of the external medium
878: with a strong shock pinches the surface of the jet, leading to the
879: production of pinch-body mode Kelvin- Helmholtz instabilities: the
880: trailing features. Hence, a single strong superluminal shock
881: ejection from the jet nozzle may lead to the production of a
882: multiple set of emission features through this mechanism. The
883: trailing features have a characteristic set of properties, which
884: make them recognizable with high resolution VLBI:
885: they form in the wake of strong components
886: instead of being ejected from the core of VLBI jets, they are
887: related to oblique shocks, they are always slower than the leading
888: feature, and (if the underlying jet has a certain opening angle)
889: they should be generated with a wide range of apparent speeds
890: (from almost stationary near the core to superluminal further
891: downstream). Moreover, \citet{Agu01} showed that the separation
892: between the trailing components increases downstream due to their
893: motion down a pressure gradient.
894:
895: All this is in agreement with what we observe in the trailing
896: components of E\,1 and with our interpretation of an expansion of
897: the jet in a density decreasing ambient medium.
898: For the time range covered by their observations (1998.23 to 2001.28),
899: \citet{Jor05} also identified the
900: {trailing phenomenology} in this source.
901:
902:
903:
904: The
905: north-south gradients detected in the linear-polarization emission
906: in the region where the trailing features are formed, is in agreement with an
907: oblique shock structure. The steep brightness-temperature
908: gradients of the trailing components indicate that the particle
909: and magnetic field density associated with these components evolve
910: in a different way compared to the ``normal" jet flow. These
911: shocked regions may be more overpressured with respect to
912: their environment, making them expand rapidly. This fast
913: expansion implies a larger positive value of $l$, which, however,
914: is compensated by an even larger (negative) value of $n$ and $b
915: (1-\alpha)$ in equation~\ref{eq:s}, resulting in a very steep
916: brightness temperature gradient (regime II).
917:
918: Pinching modes of the Kelvin-Helmholtz instability were shown to
919: couple to the trailing components observed in the simulations in
920: \citet{Agu01}. In the case of components E2-E4, the distance
921: between them ranges from 0.7-0.8 mas at the first epochs in which
922: they are observed, to almost 2.0 mas in the latest epochs. Taking
923: into account that: a) their FWHM is of the same order (see Table
924: 2); b) that these wavelengths have to be corrected for
925: geometrical and relativistic effects, resulting in a maximum
926: intrinsic wavelength of $\sim 0.7\,\rm{mas}$, and c) that the size
927: of the components can be of the order or smaller than the jet
928: radius (\citealt{Per07a}), this implies coupling of the pinching to
929: wavelengths of the order or smaller than the jet radius.
930: \citet{Per07} have shown that resonant Kelvin-Helmholtz
931: instabilities associated to high-order body modes appear in
932: sheared jets at these wavelengths. These modes have larger growth
933: rates than low-order body modes or surface modes, and their growth
934: brings the jet to a final quasi-steady state in which it remains
935: well-collimated and generates a hot shear-layer which shields the
936: core of the jet from the ambient medium. Interestingly, the jet in
937: 3C~111 is known to be well-collimated up to kiloparsec scales.
938: Further research in this direction is needed in order to check the
939: influence of the resonant modes in the long term evolution of this
940: jet.
941:
942: A by-product of the interpretation of these components as
943: Kelvin-Helmholtz instabilities is the fact that it allows us to
944: put constraints to the velocity of the jet. We can regard the wave
945: speed as the minimum speed of the jet flow, as KH modes have an
946: upper limit in their wave speeds that is precisely the velocity of
947: the flow in which they propagate \citep{Per06}. The upper limit is given by the
948: speed of E1, interpreted as a shock wave, that has to be thus
949: faster than the underlying flow. In this picture, we would have
950: the structure E1 moving with Lorentz factor $\gamma\sim 8.3$
951: through a jet with Lorentz factor $8.3>\gamma_j\geq 4.6$ in the
952: accelerated region (post 1999.38), where the lower limit is given
953: by the Lorentz factor of component E2, the fastest of the three
954: trailing components identified here.
955:
956:
957: \section{Summary and Conclusions}
958: \label{sect:conclusions}
959: In this paper, we have investigated the parsec-scale
960: jet kinematics and the interaction of the jet with its ambient
961: medium in the broad-line radio galaxy 3C\,111. Our analysis has
962: demonstrated that a variety of processes influence the jet
963: dynamics in this source: a plasma injection into the jet beam
964: associated with a major flux-density outburst leads to the
965: formation of multiple shocks that travel at different speeds
966: downstream and interact with each other and with the ambient
967: medium. The primary perturbation causes the formation of a forward
968: and a backward shock (or rarefaction). The latter fades away so
969: fast that is likely to remain undetected in minor ejections. A
970: separate work by \citet{Per07} focuses on the nature and
971: characteristics of these initial components. Several parsecs
972: downstream, the jet plasma enters a region of rapidly decreasing
973: external pressure, expands into the jet ambient medium and
974: accelerates. In the following, the plasma gets recollimated and
975: trailing features are formed in the wake of the leading component.
976:
977: A particularly interesting aspect of the source 3C\,111 in the
978: light of this and other recent works is that it is one of
979: the very rare non-blazar gamma-ray bright AGN. Besides Centaurus\,A
980: \citep{Sre99} and the possible identification of NGC\,6251
981: with the EGRET source 3EG\,J1621+8203 \citep{Muk02}, 3C\,111
982: is the only AGN whose jet-system is inclined at a relatively large
983: angle to the line of sight and that has a reliable EGRET identification:
984: \citet{Sgu05} reconsidered the possible identification
985: of the EGRET source 3EG\,J0416+3650 with 3C\,111, which was first suggested by
986: \citet{Har99} but considered unlikely because of the poor
987: positional coincidence. Very recently, R.\,C.~Hartman \& M.~Kadler (in prep.)
988: found that 3EG\,J0416+3650
989: is composed out of at least two distinct components.
990: One of them is the dominant source above 1\,GeV and is in excellent positional agreement with the
991: location of 3C\,111. Compared to blazars, the large inclination angle
992: and the relatively small distance of 3C\,111 allow us to resolve
993: structures along the jet that are as small as parsecs in deprojection
994: and which would be heavily blended with adjacent features in blazar jets.
995: As demonstrated in this paper, VLBA observations of 3C\,111 probe
996: a variety of physically
997: different regions in a relativistic extragalactic jet such as a compact
998: core, superluminal jet components, recollimation shocks and regions
999: of interaction between the jet and its surrounding medium, which are all
1000: possible sites of gamma-ray production.
1001: From early 2008 on, the gamma-ray satellite GLAST \citep{Lot07} is going to monitor
1002: the sky.
1003: If
1004: detected by GLAST, 3C\,111 may become a key source in the quest for an understanding
1005: of the origin of gamma-rays from extragalactic jets.
1006: In addition, the combination of GLAST and VLBA
1007: data with spectral data at intermediate
1008: wavelengths (optical, IR, X-ray) may allow a better determination of
1009: jet parameters and relativistic beaming effects than in most blazars
1010: because of the higher linear resolution offered by this nearby and only weakly projected
1011: jet system.
1012:
1013:
1014: Our observations of 3C\,111 are qualitatively in remarkable agreement
1015: with numerical relativistic hydrodynamic structural and emission
1016: simulations of jets such as the ones presented by \cite{Agu01} and
1017: \cite{Alo03}. Further progress is being made in the transition from
1018: two-dimensional to three-dimensional simulations of relativistic jets
1019: and in the development of new methods considering magnetic fields
1020: \citep[e.g.,]{Lei05,Miz07,Roc08}, the equation of state for relativistic gases \citep{Per07b}, and radiative processes \citep[e.g.,][and Mimica et al. in preparation]{Mim04,Mim07}. But so far neither observational data nor simulations have reached an adequate level of detail and completeness in order to allow us a quantitative direct comparison of
1021: numerical models and observed relativistic jet structure and evolution.
1022: In particular, it is not
1023: feasible
1024: today to fit iteratively the parameters of relativistic magneto-hydro-dynamical (RMHD) jet simulations to match the brightness distribution observed for any individual source.
1025: The main reasons for this are a) the immense computational power required to conduct a realistic (i.e., sufficiently detailed) modern 3D jet simulation and b) the highly non-linear nature of RMHD plasmas and their evolution. Simulation results depend critically on the starting conditions like the exact velocity, composition, and transversal structure of the flow, the structure and strength of the magnetic field and the jet environment. Future development of computational power will allow us to use larger resolutions to decrease the numerical viscosities, and to implement nonlinear and microphysics processes into simulations. VLBA observations are capable of putting hard quantitative constraints on the input parameters for RMHD jet simulations if they are densely sampled over several years. Polarimetric observations at multiple radio frequencies may allow the
1026: effects of jet-intrinsic magnetic-field variations and external Faraday-screen inhomogenities or temporal variations to be disentangled. Such data at 15\,GHz are on the way, e.g., as part of the next phase of the MOJAVE program, in which rapidly evolving sources like 3C\,111 are being observed every two months.
1027:
1028:
1029:
1030: \acknowledgments
1031: We would like to acknowledge the support of the rest of the MOJAVE Team, who have contributed
1032: to the data used in this paper, in particular we would like to thank Christian Fromm for his help
1033: with the production of the figures for the paper.
1034: We thank Dharam Vir Lal and Silke Britzen for their careful reading of the manuscript and their comments.
1035: We also thank the referee for his
1036: very constructive suggestions, which have helped to improve this paper.
1037: We are grateful to Greg Taylor, who provided complementary 2\,cm VLBA data for an additional
1038: epoch.
1039: The MOJAVE project is supported under National Science Foundation
1040: grant 0406923-AST.
1041: The
1042: Very Long Baseline Array is operated by the National Radio Astronomy Observatory, a
1043: facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.
1044: UMRAO is partially supported by a series of grants from the NSF, most
1045: recently AST-0607523, and by funds
1046: from the University of Michigan.
1047: MK has been supported in part by a Fellowship of the
1048: International Max Planck Research School for Radio and Infrared Astronomy and in part by
1049: an appointment to the NASA Postdoctoral Program at the Goddard Space Flight Center,
1050: administered by Oak Ridge Associated Universities through a contract with NASA.
1051: MP acknowledges support in part by the Spanish \emph{Direcci\'on
1052: General de Ense\~nanza Superior} under grant AYA2004-08067-C03-01
1053: and in part by a postdoctoral fellowship of the \emph{Generalitat Valenciana} (\emph{Beca Postdoctoral
1054: d'Excel$\cdot$l\`encia}).
1055: YYK is a Research Fellow of the Alexander von Humboldt
1056: Foundation and was supported in part by the Russian Foundation
1057: for Basic Research (project 05-02-17377).
1058: DCH was supported by grants from Research Corporation and the National Science Foundation (AST-0707693)
1059: IA has been supported in part by an I3P contract with the Spanish Consejo Superior de Investigaciones Cient\'{i}ficas and in part by a contract with the German Max-Planck-Institut f\"ur Radioastronomie (through the ENIGMA network, contract HPRN-CT-2002-00321), which
1060: were partially funded by the EU.
1061:
1062: \appendix
1063: \section{The Jet Inclination Angle}
1064: \label{sect:inclination}
1065: The 1996 radio outburst of 3C\,111 puts strong constraints on the angle to the line of sight for this source,
1066: if one assumes that a similarly bright component as E has been ejected in the counterjet, as well. Due to differential
1067: Doppler boosting, the flux density ratio between the jet- and counter-jet emission is
1068: \begin{equation}
1069: \frac{S_{\rm J}}{S_{\rm CJ}} = \left(\frac{1+\beta \cos{\theta}}{1-\beta \cos{\theta}}\right)^{2-\alpha} \quad .
1070: \end{equation}
1071: Thus, for a given jet to counter-jet ratio $x = \frac{S_{\rm J}}{S_{\rm CJ}}$
1072: \begin{equation}
1073: \beta \cos{\theta} = \frac{x-1}{x+1} \quad .
1074: \end{equation}
1075: With $\alpha = 0.3$, $S_{\rm J} = 3.4$\,Jy (components E and F in 1997.19), $S_{\rm CJ} < 10$\,mJy and $\beta < 1$, $\theta < 21^\circ$.
1076: For a realistic jet speed of, e.g., $\beta = 0.956$ ($\gamma = 3.4$), the angle to the line
1077: of sight is: $\theta = 19^\circ$.
1078: An estimate close to this value
1079: can be derived from the variability Doppler factor measured by \citet{Lae99} and the apparent superluminal jet speed.
1080: As outlined in detail in \citet{Coh07}, this leads to a value of $\theta \sim 15^\circ$.
1081:
1082: It is important to note that this calculation implicitly assumes symmetry between the jet and counter-jet, which
1083: in projection does not have to be the case if the counter-jet is covered by an obscuring torus
1084: as it is well-established for systems at larger inclination angles \citep[e.g., NGC\,1052: see ][]{Kad04}. Indeed,
1085: Faraday rotation measurements towards the 3C\,111 pc-scale jet (\citealt{Zav02}; see also Sect.~\ref{sect:polarization})
1086: and X-ray spectral observations
1087: (\citealt{Lew05}) suggest substantial amounts
1088: of obscuring material. Free-free absorption could also substantially lower the counter-jet radio emission and allow for
1089: larger jet angles to the line of sight.
1090:
1091: An independent lower limit on the inclination angle of $\theta > 21^\circ$ was given by \cite{Lew05} assuming that the
1092: deprojected size of the largescale 3C\,111 double-lobe structure is smaller than $500 h^{-1}$\,kpc. This discrepancy
1093: implies that either 3C\,111 is unusually large or there is a misalignement between the large-scale jet-axis and the parsec-scale
1094: jet axis inclination to the line of sight, although the projected position angles of the large-scale jet ($63^\circ$) and the
1095: parsec-scale jet ($\sim 65^\circ$) are almost the same.
1096:
1097: \clearpage
1098: \section{Image-Plane vs. $(u,v)$-plane Model Fitting}
1099:
1100: It is interesting to compare our results from this very detailed analysis
1101: of one individual object with the results of the kinematical-survey
1102: analysis of \cite{Kel04}, who investigated the speeds of 110 extragalactic jets, including
1103: 3C\,111, based on the VLBA 2\,cm Survey data between 1994 and 2001.
1104: \cite{Kel04} made the component identification in the image plane and
1105: represented the evolving structure of component E and its trailing features
1106: by only one component
1107: %In total, they consider three moving components: between
1108: %1995 and 1996 and identify two components between 1\,mas and
1109: %5\,mas from the core to which they assign a ``poor'' quality code.
1110: %In this work, this region has been represented by four different components
1111: %(A, B, C, D) of which two (B, C) could be traced over four epochs
1112: %and the model fitting was done in the $(u,v)$-domain.
1113: %We find $(5.7 \pm 0.1) c$ and $(4.6 \pm 0.3) c$ for components B and C while
1114: %\cite{Kel04} report $(4.2 \pm 1.0) c$ and $(2.9 \pm 1.7) c$.
1115: %While we believe that,
1116: %in the framework of our overall kinematical model, the representation of
1117: %this region with four different components is the most consistent one, it needs to
1118: %be noted that the 1995/1996 data do not fully constrain any multi-epoch
1119: %model, so that the differences between this work and \cite{Kel04}
1120: %in this range should not be over-interpreted.
1121: %More insightful is the comparison of our model of the post-1996 ejecta,
1122: %involving the components
1123: whereas, in this work we distinguish the sub-components E, F, E1, E2, E3, and E4.
1124: %with the one
1125: %component that \cite{Kel04} used to represent this structure,
1126: %which is assigned an ``excellent'' quality code
1127: %and which is found to travel with a speed of
1128: Formally, the speed of $(4.9 \pm 0.2) c$ found by \citet{Kel04} is in good agreement
1129: with the speed of our leading component, E1, so
1130: the much simpler model derived from image-plane analysis represents the
1131: fastest moving structure. The acceleration (with respect to the ejecta's
1132: smaller velocity prior to mid 1999), as well as the additional components that
1133: we interpret as a backward shock and trailing jet features become visible
1134: only after a more complicated model fitting of the data in the
1135: $(u,v)$-domain. The necessarily less-complex model used in a survey
1136: analysis like the one conducted by \cite{Kel04} is only part of the reason for
1137: this discrepancy. Image-plane analysis makes it very difficult to interpret
1138: a moving feature that changes its structure in a complex way and that
1139: has no clear persistent brightness maxima. In addition, $(u,v)$-plane
1140: fitting in general achieves higher angular resolution so that the
1141: two bright, but closely separated components E and F could not be distinguished
1142: in in early to mid 1996. Figure~\ref{fig:projplots} demonstrates that even in
1143: October 1996 when E and F are separated by only 0.3\,mas and are located within
1144: 1\,mas from the core, a one-component model
1145: clearly fails to represent this compact structure.
1146:
1147: \bigskip
1148:
1149: \begin{figure*}[h]
1150: \centering
1151: \includegraphics[width=\linewidth]{f11}
1152: \caption{Visibility amplitudes projected radially along
1153: P.A. 70$^\circ$ for the 1996.82 observation of 3C\,111.
1154: The double-peak indicates the presence of a bright structure
1155: within less than a milliarcsecond of the core.
1156: The top left panel shows a model (black) which was fitted to the data (green) consisting of
1157: one model component for the newly ejected jet feature and the residuals
1158: of this model are shown in the bottom left panel. Up to about $600$\,mJy of
1159: correlated flux density is missing from the model. The right panels show the same data
1160: fitted by a model consisting of two components (corresponding to E and F).
1161: }
1162: \label{fig:projplots}
1163: \end{figure*}
1164:
1165:
1166:
1167: \clearpage
1168: \begin{thebibliography}{}
1169: %
1170: \bibitem[Agudo et al.(2001)]{Agu01} Agudo, I., G\'omez, J.,
1171: Mart{\'{\i}}, J.~M., et al.\
1172: 2001, \apj, 549, L183
1173: %
1174: \bibitem[Alef et al.(1998)]{Ale98} Alef, W., Preuss, E., Kellermann,
1175: K.~I., Gabuzda, D.\ 1998, in Radio Emission from Galactic and Extragalactic Compact Sources, ASP Conf. Ser. 144, 129
1176: %
1177: \bibitem[Aller, Aller, \& Hughes(2003)]{All03} Aller, M.~F., Aller,
1178: H.~D., Hughes, P.~A.\ 2003, in
1179: Radio Astronomy at the Fringe, Zensus, J.~A., Cohen, M.~H., Ros, E. (eds.),
1180: ASP Conference Ser. 300, 159
1181: %
1182: \bibitem[Aloy et al.(2003)]{Alo03} Aloy, M.-{\'A}.,
1183: Mart{\'{\i}}, J.-M., G{\'o}mez, J.-L., Agudo, I., M{\"u}ller, E., \&
1184: Ib{\'a}{\~n}ez, J.-M.\ 2003, \apjl, 585, L109
1185: %
1186: \bibitem[Blandford \& Konigl(1979)]{Bla79} Blandford, R.~D.,
1187: \& Konigl, A.\ 1979, \apj, 232, 34
1188: %
1189: \bibitem[Cara \& Lister(2007)]{Car07} Cara, M., \& Lister,
1190: M.~L.\ 2007, ArXiv Astrophysics e-prints, arXiv:astro-ph/0702449
1191: %
1192: \bibitem[Cohen et al.(1971)]{Coh71} Cohen, M.~H., Cannon, W.,
1193: Purcell, G.~H., Shaffer, D.~B., Broderick, J.~J., Kellermann, K.~I., \&
1194: Jauncey, D.~L.\ 1971, \apj, 170, 207
1195: %
1196: \bibitem[Cohen et al.(1977)]{Coh77} Cohen, M.~H., et al.\
1197: 1977, \nat, 268, 405
1198: %
1199: \bibitem[Cohen et al.(2007)]{Coh07} Cohen, M.~H., Lister,
1200: M.~L., Homan, D.~C., Kadler, M., Kellermann, K.~I., Kovalev, Y.~Y., \&
1201: Vermeulen, R.~C.\ 2007, \apj, 658, 232
1202: %
1203: \bibitem[Goetz et al.(1987)]{Goe87}
1204: Goetz, M.~M.~A., Preuss, E., Alef, W., \& Kellermann, K.~I.\ 1987, \aap,
1205: 176, 171
1206: %
1207: \bibitem[G\'omez et al.(1997)]{Gom97} G\'omez, J.~L., Mart\'{\i},
1208: J.~M.~A., Marscher, A.~P., Ibanez, J.~M.~A., \& Alberdi, A.\ 1997, \apjl,
1209: 482, L33
1210: %
1211: \bibitem[G\'omez et al.(2001)]{Gom01} G\'omez, J., Marscher, A.~P.,
1212: Alberdi, A., Jorstad, S.~G., Agudo, I.\ 2001, \apjl, 561, L161
1213: %
1214: \bibitem[G{\'o}mez(2005)]{Gom05} G{\'o}mez, J.\ 2005, Future
1215: Directions in High Resolution Astronomy, 340, 13
1216: %
1217: \bibitem[Hartman et al.(1999)]{Har99} Hartman, R.~C., et al.\
1218: 1999, \apjs, 123, 79
1219: %
1220: \bibitem[Hartman \& Kadler(2007)]{Har07} Hartman, R.~C. \& Kadler, M.\ 2007, \apj, submitted
1221: %
1222: \bibitem[Homan et al.(2006)]{Hom06} Homan, D.~C., et al.\
1223: 2006, \apjl, 642, L115
1224: %
1225: \bibitem[Homan \& Lister(2006)]{Hom06b} Homan, D.~C., \&
1226: Lister, M.~L.\ 2006, \aj, 131, 1262
1227: %
1228: \bibitem[Homan et al.(2003)]{Hom03} Homan, D.~C., Lister,
1229: M.~L., Kellermann, K.~I., Cohen, M.~H., Ros, E., Zensus, J.~A., Kadler, M.,
1230: \& Vermeulen, R.~C.\ 2003, \apjl, 589, L9
1231: %
1232: \bibitem[Homan et al.(2002)]{Hom02} Homan, D.~C., Ojha, R.,
1233: Wardle, J.~F.~C., Roberts, D.~H., Aller, M.~F., Aller, H.~D., \& Hughes,
1234: P.~A.\ 2002, \apj, 568, 99
1235: %
1236: \bibitem[Hughes et al.(1985)]{Hug85} Hughes, P.~A., Aller,
1237: H.~D., \& Aller, M.~F.\ 1985, \apj, 298, 301
1238: %
1239: \bibitem[Jorstad et al.(2001)]{Jor01} Jorstad, S.~G.,
1240: Marscher, A.~P., Mattox, J.~R., Wehrle, A.~E., Bloom, S.~D., \& Yurchenko,
1241: A.~V.\ 2001, \apjs, 134, 181
1242: %
1243: \bibitem[Jorstad et al.(2005)]{Jor05} Jorstad, S.\,G., Marscher, A.\,P.,
1244: Lister, M.\,L., et al.\ 2005, \aj, 130, 1418
1245: %
1246: \bibitem[Kadler(2005)]{Kad05} Kadler, M.\ 2005, Ph.\,D. Thesis, Rheinische Friedrich-Wilhelms-Universit\"at Bonn, Bonn, Germany
1247: %
1248: \bibitem[Kadler et al.(2004)]{Kad04} Kadler, M., Ros, E., Lobanov, A.\,P., Falcke, H., Zensus, J.\,A.\ 2004, \aap, 426, 481
1249: %
1250: \bibitem[Kellermann et al.(1998)]{Kel98} Kellermann, K.~I., Vermeulen, R.~C.,
1251: Zensus, J.~A., \& Cohen, M.~H.\ 1998, \aj, 115, 1295
1252: %
1253: \bibitem[Kellermann \& Moran(2001)]{Kel01} Kellermann, K.~I.,
1254: \& Moran, J.~M.\ 2001, \araa, 39, 457
1255: %
1256: \bibitem[Kellermann et al.(2004)]{Kel04} Kellermann, K.~I., Lister,
1257: M.~L., Homan, D.~C., et al.\ 2004, \apj, 609, 539
1258: %
1259: %\bibitem[Kharb et al.(2003)]{Kha03} Kharb, P., Gabuzda, D.,
1260: %Alef, W., Preuss, E., \& Shastri, P.\ 2003, New Astronomy Review, 47, 621
1261: %
1262: \bibitem[Kovalev et al.(2005)]{Kov05} Kovalev, Y.~Y., Kellermann, K.~I., Lister, M.~L.,
1263: et al.\ 2005, \aj, 130, 2473
1264: %
1265: \bibitem[L\"ahteenm\"aki \& Valtaoja(1999)]{Lae99} L\"ahteenm\"aki, A.
1266: \& Valtaoja, E.\ 1999, \apj, 521, 493
1267: %
1268: \bibitem[Leismann et al.(2005)]{Lei05} Leismann, T.,
1269: Ant{\'o}n, L., Aloy, M.~A., M{\"u}ller, E., Mart{\'{\i}}, J.~M., Miralles,
1270: J.~A., \& Ib{\'a}{\~n}ez, J.~M.\ 2005, \aap, 436, 503
1271: %
1272: \bibitem[Lewis et al.(2005)]{Lew05} Lewis, K.\,T., Eracleous, M.,
1273: Gliozzi, M., et al.\ 2005, \apj, 622, 816
1274: %
1275: \bibitem[Linfield \& Perley(1984)]{Lin84} Linfield, R. \& Perley, R.\ 1984, \apj, 279, 60
1276: %
1277: \bibitem[Lister \& Homan(2005)]{Lis05} Lister, M.~L., \&
1278: Homan, D.~C.\ 2005, \aj, 130, 1389
1279: %
1280: \bibitem[Lott et al.(2007)]{Lot07} Lott, B., Carson, J.,
1281: Ciprini, S., Dermer, C.~D., Giommi, P., Madejski, G., Lonjou, V., \&
1282: Reimer, A.\ 2007, American Institute of Physics Conference Series, 921, 347
1283: %
1284: \bibitem[Marscher \& Gear(1985)]{Mar85} Marscher, A.~P., \&
1285: Gear, W.~K.\ 1985, \apj, 298, 114
1286: %
1287: \bibitem[Marscher et al.(2002)]{Mar02} Marscher, A.~P.,
1288: Jorstad, S.~G., G{\'o}mez, J.-L., Aller, M.~F., Ter{\"a}sranta, H., Lister,
1289: M.~L., \& Stirling, A.~M.\ 2002, \nat, 417, 625
1290: %
1291: \bibitem[Marscher (2006)]{Mar06} Marscher, A.~P.\ 2006, AN 327, 217
1292: %
1293: \bibitem[Mukherjee et al.(2002)]{Muk02} Mukherjee, R.,
1294: Halpern, J., Mirabal, N., \& Gotthelf, E.~V.\ 2002, \apj, 574, 693
1295: %
1296: \bibitem[Mimica et al.(2004)]{Mim04} Mimica, P., Aloy, M.~A.,
1297: M{\"u}ller, E., \& Brinkmann, W.\ 2004, \aap, 418, 947
1298: %
1299: \bibitem[Mimica et al.(2007)]{Mim07} Mimica, P., Aloy, M.~A.,
1300: M\"uller, E.\ 2007, \aap, 466, 93
1301: %
1302: \bibitem[Mizuno et al.(2007)]{Miz07} Mizuno, Y., Hardee, P.,
1303: \& Nishikawa, K.-I.\ 2007, \apj, 662, 835
1304: %
1305: %\bibitem[Pauliny-Toth \& Kellermann(1966)]{Pau66}
1306: %Pauliny-Toth, I.~I.~K., \& Kellermann, K.~I.\ 1966, \apj, 146, 634
1307: %
1308: \bibitem[Perucho et al.(2006)]{Per06} Perucho, M., Lobanov, A.P.,
1309: Mart\'{\i}, J.M., Hardee, P.E.\ A\&A 456, 493, 2006
1310: %
1311: \bibitem[Perucho et al.(2007)]{Per07} Perucho, M., Hanasz, M., Mart\'{\i},
1312: J.M., Miralles, J.A.\ 2007, Phys. Rev. E, 75, 056312
1313: %
1314: \bibitem[Perucho \& Lobanov(2007)]{Per07a} Perucho, M., \& Lobanov,
1315: A.~P.\ 2007, \aap, 469, L23
1316: %
1317: \bibitem[Perucho \& Mart\'{\i}(2007)]{Per07b} Perucho, M., \&
1318: Mart\'{\i}, J.M.\ 2007, MNRAS, 382, 526
1319: %
1320: \bibitem[Piner et al.(2007)]{Pin07} Piner, B.~G., Mahmud, M.,
1321: Fey, A.~L., \& Gospodinova, K.\ 2007, \aj, 133, 2357
1322: %
1323: \bibitem[Preuss, Alef, \& Kellermann(1988)]{Pre88} Preuss, E., Alef, W.,
1324: Kellermann, K.~I.\ 1988, in
1325: IAU Symp.~129: The Impact of VLBI on Astrophysics and Geophysics, 105
1326: %
1327: \bibitem[Roca-Sogorb et al.(2008)]{Roc08} Roca-Sogorb, M., Perucho, M.,
1328: Gómez, J.L., Martí, J.M., Antón, L., Aloy, M.A., Agudo, I.\ 2008, in
1329: proceedings of Extragalactic Jets: Theory and Observation from Radio to
1330: Gamma-Ray, eds: T.A.~Rector, D.S.~de~Young, ASP Conference Series, in
1331: press
1332: %
1333: %\bibitem[Savolainen et al.(2002)]{Sav02} Savolainen, T.,
1334: %Wiik, K., Valtaoja, E., Jorstad, S.~G., \& Marscher, A.~P.\ 2002, \aap,
1335: %394, 851
1336: %
1337: \bibitem[Scheck et al.(2002)]{Sch02} Scheck, L., Aloy, M.~A.,
1338: Mart{\'{\i}}, J.~M., G{\'o}mez, J.~L., M\"uller, E.\ 2002, \mnras,
1339: 331, 615
1340: %
1341: \bibitem[Sguera et al.(2005)]{Sgu05} Sguera, V., Bassani, L.,
1342: Malizia, A., Dean, A.~J., Landi, R., \& Stephen, J.~B.\ 2005, \aap, 430,
1343: 107
1344: %
1345: \bibitem[Shepherd(1997)]{She97} Shepherd, M.~C.\ 1997,
1346: Astronomical Data Analysis Software and Systems VI, 125, 77
1347: %
1348: \bibitem[Sreekumar et al.(1999)]{Sre99} Sreekumar, P.,
1349: Bertsch, D.~L., Hartman, R.~C., Nolan, P.~L., \& Thompson, D.~J.\ 1999,
1350: Astroparticle Physics, 11, 221
1351: %
1352: \bibitem[Ter\"asranta et al.(2004)]{Ter04} Ter\"asranta, H., Achren, J.,
1353: Hanski, M., et al.\ 2004, \aap, 427, 769
1354: %
1355: \bibitem[Tingay, Preston, \& Jauncey(2001)]{Tin01} Tingay, S.~J., Preston, R.~A., \& Jauncey, D.~L.\ 2001, \aj, 122, 1697
1356: %
1357: \bibitem[Vermeulen \& Cohen(1994)]{Ver94} Vermeulen, R.~C.,
1358: \& Cohen, M.~H.\ 1994, \apj, 430, 467
1359: %
1360: \bibitem[Whitney et al.(1971)]{Whi71} Whitney, A.~R., et al.\
1361: 1971, Science, 173, 225
1362: %
1363: \bibitem[Zavala \& Taylor(2002)]{Zav02} Zavala, R.~T., \&
1364: Taylor, G.~B.\ 2002, \apjl, 566, L9
1365: %
1366: \bibitem[Zensus(1997)]{Zen97} Zensus, J.~A.\ 1997, \araa, 35,
1367: 607
1368: %
1369: \bibitem[Zensus et al.(2002)]{Zen02} Zensus, J.~A., Ros, E.,
1370: Kellermann, K.~I., et al.\ 2002, \aj, 124, 662
1371: %
1372: \end{thebibliography}
1373:
1374: \clearpage
1375: \input{tab1}
1376: \clearpage
1377: \LongTables
1378: \input{tab2}
1379: \clearpage
1380: \input{tab3}
1381: \end{document}
1382: