1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass[12pt]{emulateapj}
3: %\usepackage{natbib}
4: \shorttitle{A Compact ESE Toward AO~0235+164}
5: \shortauthors{C. E. Senkbeil et al.}
6: \bibpunct{(}{)}{;}{a}{,}{,}
7: \begin{document}
8: \bibliographystyle{plainnat}
9:
10: %\nonstopmode
11:
12: \title{A Compact Extreme Scattering Event Cloud Towards AO~0235+164}
13: \author{C. E. Senkbeil, S. P. Ellingsen, and J. E. J. Lovell}\affil{School of Mathematics and Physics, Private Bag 37, University of Tasmania, Hobart, TAS 7001 , Australia}
14:
15: \and
16: \author{J.-P. Macquart$^\dagger$}
17: \affil{NRAO, P.O. Box 0, Socorro, NM 87801, U.S.A. and Dept. of Astronomy, California Institute of Technology, Pasadena, CA 91125, U.S.A.}\altaffiltext{$\dagger$}{NRAO Jansky Fellow}
18: \and
19: \author{G. Cim\`o}
20: \affil{Joint Institute for VLBI in Europe, Postbus 2, 7990 AA Dwingeloo, Netherlands}
21: \and
22: \author{D. L. Jauncey}
23: \affil{CSIRO Australia Telescope National Facility, Epping New South Wales, Australia}
24:
25:
26: \begin{abstract}
27: We present observations of a rare, rapid, high amplitude Extreme
28: Scattering Event toward the compact BL-Lac AO~0235+164 at 6.65\,GHz. The ESE cloud is compact; we estimate its diameter between 0.09 and 0.9\,AU, and is at a distance of less than $3.6$\,kpc. Limits on the
29: angular extent of the ESE cloud imply a minimum cloud electron density of
30: $\sim 4 \times 10^3$\,cm$^{-3}$. Based on the amplitude and timescale of the ESE observed here, we suggest that at least one of the transients reported by \citet{bower2007} may be attributed to ESEs.
31: \end{abstract}
32:
33: \keywords{ISM: structure --- BL Lacertae objects: individual (AO0235+164) --- galaxies: active}
34:
35:
36:
37: \section{Introduction}
38:
39: The $z=0.940$ \citep{cohen1987} flat spectrum BL Lac AO~0235+164 has
40: been known to exhibit variability since its discovery \citep{macleod1976}
41: over a broad range of wavelengths and timescales (\cite{raiteri2006} and references therein). Two absorbing galaxy complexes exist along the line of sight to this source at
42: $z=0.524$ and $z=0.851$ \citep{burbidge1976,rieke1976}, hindering interpretation of
43: the spectrum and possibly contributing to the observed variability
44: (e.g. by microlensing by stars in the foreground system
45: \citep{ostriker1985}). The source is amongst the most compact radio
46: AGN: it exhibits long-term variability down to meter-wavelengths and
47: is slightly resolved by high angular resolution VLBI at 43\,GHz
48: \citep{frey2000,piner2006}.
49:
50: The extreme compactness of this source renders the interpretation of
51: its variability difficult. For instance, there is debate whether the
52: origin of the centimeter-wavelength intra-day variability in observed
53: AO~0235$+$164 is primarily intrinsic or due to interstellar
54: scintillation (ISS) \citep{kraus1999,lovell2003}, and on what
55: timescales each contributes. In addition, the source compactness
56: renders it highly susceptible to flux density deviations by the compact
57: intervening refracting structures in the local ISM that give rise to
58: Extreme Scattering Events (ESEs) \citep{fiedleretal87} because it is
59: easy for even a small cloud to subtend the angular extent of such a
60: small source.
61:
62: Of the three main causes of cm-wavelength variability, ESEs are the
63: most rare and certainly the least understood. Almost all
64: flat-spectrum AGN show variability \citep{altschuler1977}. The
65: MASIV survey found that $\sim 20$\% of this population exhibits
66: scintillation at any one time with over 56\% exhibiting ISS at any
67: stage over the course of a year \citep{lov2007,jauncey2007}. However, the estimated rate of ESE
68: events in compact AGN is only 0.013\,source$^{-1}$yr$^{-1}$ \citep{fiedleretal87}
69:
70: There are various plausible interpretations as to the origins of the
71: structures that give rise to ESEs. These include intrinsically
72: turbulent ionized clouds, purely refractive (gaussian) lenses, and
73: primarily neutral clouds enveloped by a thin ionized sheath that
74: responsible for the cloud's refractive properties
75: \citep{fiedleretal94,romani87,clegg98,ww98}. These models pose a number
76: of challenges; the electron densities implied by the plasma lens
77: models requires the clouds to be $\sim 10^3$ times overpressured with
78: respect to the ambient ISM, while the neutral-cloud interpretation
79: implies that such structures would contain a large fraction of the
80: baryonic dark matter content of the Milky Way.
81:
82: In this paper we report an unusually short-timescale ESE in
83: AO~0235$+$164. Our observations are reported in section 2, while in
84: section 3 we show that both intrinsic variability and interstellar
85: scintillation are incapable of explaining this event, and discuss the
86: physical properties of the ESE necessary to give rise to the observed
87: lightcurve. Our conclusions are presented in section 4.
88:
89:
90:
91: \section{Observations}\label{sec:obs}
92: AO~0235+164 has been monitored quasi-continuously since 2003 as part
93: of the Continuous Single-dish Monitoring of Intraday variability at
94: Ceduna (COSMIC) program \citep{mcculloch2005}. The
95: observations were conducted with the University of Tasmania Ceduna 30-m antenna at a center
96: frequency of 6.65~GHz with a 300\,MHz bandwidth. The flux density was
97: sampled by scanning across the source in forward and reverse directions in both Right Ascension and Declination. The flux density measurements are obtained from
98: the height of the gaussian profile of the source signal above the
99: system temperature baseline. All the scans are scaled to a noise diode, which in turn is calibrated against 3C~227, which at
100: this frequency has a flux density of 1.99\,Jy \citep{baars1977}. Measured pointing offsets are used to correct the amplitude in orthogonal scans to
101: minimise the impact of inaccurate pointing on the measured
102: amplitude. A correction is made to account for gain dependence of the antenna due to
103: the distortion with respect to elevation. The four
104: fitted amplitudes are averaged together to constitute a single calibrated sample.
105:
106: The flux density monitoring data from Ceduna are known to suffer variations that are systematic in nature and predominantly diurnal in timescale, being related to changes in air temperature at the observatory. These errors scale with source flux density and affect source and calibrator equally (as they are approximately the same brightness). The amplitude of this systematic effect for the present case is 3\%, a factor of $\sim 40$ times smaller than the variability in AO~0235+164 discussed here. The calibrator data shown in Figure~\ref{fig:entirelc} demonstrates that this effect is negligible. Moreover, the variations occur on a much longer timescale than the diurnal systematic errors.
107:
108: Figure~\ref{fig:entirelc} demonstrates that the source exhibited variations on timescales of days and months to years during the period 2003 to 2005. During the
109: five day interval beginning 2005 July 20 the source exhibited a rapid, short timescale,
110: increase in flux density from $1$ to $2.3\,$Jy, followed by a fall to
111: $1\,$Jy over the next five days. Over the following four days the
112: flux density recovered to a value of $1.6\,$Jy, close to the long term
113: mean flux density of the source (Figure~\ref{fig:mainlc}).
114:
115: This event is atypical of the variability commonly observed in this
116: source and, indeed, all seven of the variable flat-spectrum quasars
117: regularly observed in the COSMIC program. The peak amplitude of the
118: flux density excursion exceeds the underlying mean by 44\% and occurs
119: on the exceedingly short timescale of only $\sim 4$\,days. This
120: behaviour contrasts markedly with the long-term intrinsic variability
121: evident in this source, as shown in Figure~\ref{fig:entirelc}. This
122: event is a one-off in the COSMIC dataset; no similar events have been
123: recorded in any of the other sources monitored in this program.
124:
125: \section{Analysis}\label{sec:anyl}
126: \subsection{Origin of the Event}
127: The main clue as to the physical origin of this event lies in the fact
128: that it contains an excursion well {\it below} the average source flux
129: density. Of the possible causes of this event -- scintillation
130: of a compact component, an intrinsic flare, microlensing by stars in an
131: intervening galaxy, or an extreme scattering
132: event (ESE) in our Galaxy -- only an ESE is capable of affecting the entire source
133: brightness in such a way as to cause both the large positive and
134: negative fluctuations observed.
135:
136: In the month prior to the event, the lightcurve is flat, indicating that
137: the flux density was dominated by quiescent emission originating from a
138: region of the source which, we argue here, is too large to exhibit
139: intrinsic variations on day timescales. If variations on a timescale of
140: $\sim 40\,$days are to be attributed to intrinsic variability, causality
141: requires that the emission region encompass a size no larger than $\sim
142: 0.03\,{\cal D}\,(1+z)^{-1}\,$pc, where ${\cal D}$ is the Doppler factor
143: associated with the motion of the emission region towards the observer.
144: However, the absence of interstellar scintillation in this source places a
145: lower limit on the source size which is incompatible with this limit.
146: Specifically, the absence of ISS implies an apparent angular size of $\ga
147: 20\,\mu$as. Using cosmological parameters of $H_0$ = 71
148: kms$^{-1}Mpc^{-1}$, $\Omega_M$ = 0.27, $\Omega_{vac}$ = 0.73, the linear
149: size of the emitting region must therefore exceed $0.16\,{\cal D}$\,pc.
150: This size is not sufficiently small to allow causally-connected intrinsic
151: variations in the source on the timescale observed.
152:
153:
154: An interstellar scintillation-based explanation is unsatisfactory for the same
155: reason. The onset of scintillation would require the appearance of a
156: new source component compact enough to exhibit ISS. However, since
157: ISS can only affect the flux density of this new compact component,
158: and not the extended quiescent emission, the lightcurve can never dip
159: below the level of the quiescent emission.
160:
161: Scattering due to the stellar wind of a nearby star along the line of
162: sight to the source is also not a plausible explanation. We searched several
163: Wolf-Rayet, O and B star catalogues through Vizier and found no nearby
164: stars capable of generating bubbles of dense ionized matter with the
165: properties required for this event.
166:
167:
168: Microlensing by stars in the intervening galaxies at $z=0.524$ and $z=0.851$ is similarly unable to explain the magnitude of the flux density excursion. The $50\,\mu$as source size makes it impossible for microlensing to appreciably alter the entire source flux density because lensing only affects a region of size comparable to the Einstein radius. Even for the closer of the two systems, the Einstein radius of a star of mass $m$ is only $\approx 1.56 \,(m/M_\sun)^{1/2}\,\mu$as (assuming $H_0 = 70\,$km\,s$^{-1}$Mpc$^{-1}$). Futhermore, we note that the observed dip in the source flux density during the event is at variance with the magnification profile expected from a microlensing event. A more detailed argument against the likelihood of microlensing in AO 0235$+$164 is presented in \citet{kraus1999}.
169:
170: We are thus forced to conclude that only an event completely extrinsic
171: to the source is capable of affecting the source's emission, which is
172: too large to exhibit ISS and too large to be causally connected to an
173: intrinsic flare. The most plausible explanation, therefore, is that
174: the flux density excursion is caused by an ESE. We also note
175: that the lightcurve exhibits the large increase followed by a
176: subsequent sharp decrease which is characteristic of ESE events. This
177: is consistent with the passing of the cloud across the line of sight
178: in which the radiation is first focused on to the observer and then,
179: as the observer passes the ESE boundary, the radiation is steered out
180: of the line of sight into the neighbouring focusing region.
181:
182: \subsection{ESE Cloud Properties}
183: The duration of the ESE places constraints on the transverse length
184: scale of the ionising material. The absence of data prior to and
185: following the high-amplitude variations places uncertainty on the
186: total length of the ESE event. We argue that the event had ceased by
187: day 947. This is supported by the 8 GHz data from Medicina \citep{bach2007}
188: that agree well with the Ceduna data where they overlap and, following
189: the period where rapid changes were observed, are consistent with the
190: flux density having returned to the underlying slow intrinsic
191: change. The earliest possible date for the commencement of the event
192: is day 916. Thus the maximum likely event duration is
193: $\Delta T=[947-916]$ days, while the minimum possible duration is 16
194: days assuming that the event is temporally symmetric about its peak,
195: as is observed in other ESEs (e.g. \citet{fiedleretal87}).
196:
197: The transverse extent of the cloud is
198: \begin{eqnarray}
199: r = 0.18 \left( \frac{v}{20\,{\rm km\,s}^{-1}} \right)
200: \left( \frac{\Delta T}{16\,{\rm days}} \right)\, {\rm AU},
201: \end{eqnarray}
202: where $v$ is the speed of the cloud transverse to the line of sight,
203: plausibly in the range $10-50\,$km\,s$^{-1}$. Taking the uncertainties
204: in both $\Delta T$ and $v$ into account, the estimated linear
205: scale of the cloud is in the range 0.09-0.90\,AU.
206:
207: Since the event significantly alters the entire flux density of the
208: source, the angular extent of the ionized cloud must be comparable to
209: or greater than the angle subtended by the core of the source. Space
210: VLBI observations of AO~0235+156 place an upper limit on the core size
211: at 4.8~GHz of $50\,\mu$as \citep{frey2000}. This, combined with the
212: favoured cloud size of 0.18\,AU requires a cloud distance of less than
213: 3.6~kpc.
214:
215:
216: The electron density of an ESE cloud must be much higher than the
217: ambient medium. A simple estimate of the electron density is obtained by considering refraction by a cloud with a gaussian density profile:
218: \begin{eqnarray}
219: n_e=n_0\exp \left[ \frac{-(r^2+z^2)}{2R_0^2} \right],
220: \end{eqnarray}
221: where $n_0$ is the maximum electron density, $R_0$ is the cloud
222: radius, $z$ is the distance along the line of sight, and $r$ is the distance from the cloud center transverse to the line of sight. The bending
223: angle, $\alpha$, of the cloud must be comparable to the
224: angular size of the source so that,
225: \begin{eqnarray}
226: \Delta \theta \approx \alpha = \frac{1}{k}\frac{\partial \phi}{\partial r},
227: \end{eqnarray}
228: where $k$ is the wavenumber and $\phi$ is the phase delay imparted by the cloud. These
229: assumptions lead to a relationship between $n_0$ and $\alpha$ as a function of distance, $r$, from the cloud center:
230: \begin{eqnarray}
231: \alpha= (2 \pi)^{1/2}\, n_0 \,r_e \lambda \, \frac{r}{R_0} \exp \left(- \frac{r^2}{2 R_0^2} \right),
232: %n_0 \frac{q_e^2 }{2 c k^2 m_e R_0 \epsilon_0} \sqrt{2 \pi} r \exp \left[- \frac{r^2}{2 R_0^2} \right]
233: \end{eqnarray}
234: where $r_e$ is the classical electron radius. The maximum bending angle occurs at $r=R_0$; this requires a minimum cloud electron density of $4.1 \times 10^3\,$cm$^{-3}$, comparable to or higher than density estimates of previous ESE clouds \citep{fiedleretal87}.
235:
236:
237:
238:
239: The COSMIC program, to date, observed seven sources over a period
240: of four years, during which time one ESE was detected. Our observation
241: of a single event in 35 source years of monitoring is not inconsistent with
242: the ESE rate inferred by \citet{fiedleretal87}. This supports our
243: interpretation of this phenomenon as an ESE.
244:
245:
246: There have been a number of recent reports of short-duration transients at
247: high Galactic latitude \citep{bower2007,niinuma2007,matsumura2007}. It is interesting to
248: speculate whether short duration ESEs, similar to, or perhaps even more extreme than the event we have observed in
249: AO 0235+164 could be responsible for these events. \citet{bower2007} used
250: archival VLA calibration observations to investigate the presence of radio
251: transients in a high-galactic latitude field for a period of about 20
252: minutes on average once every 7 days for more than 20 years. They
253: identified 10 radio transient events where there is no persistent emission
254: stronger than a few $\mu$Jy. However, in addition to these events their
255: data shows four cases where a source detected in the deep radio images was
256: detected in only a single epoch 20 minute observation. Comparing the peak
257: and mean flux densities, the amplification observed in these events is of
258: the order of 2-5, larger than typical ESEs, but not implausibly so. These
259: events may then be due to ESEs with similar timescales to the AO 0235$+$168
260: event ($<$ 20 days). However, for the transients of \citet{bower2007}, and those
261: detected in the Nasu 1.4 GHz observations \citep{niinuma2007,matsumura2007} which have no detected persistent
262: radio emission the implied amplifications are of the order of 10-100 or more
263: and cannot plausibly explained by ESEs.
264:
265: \section{Conclusion}
266:
267: We observed a rapid 16 day timescale flux density variation in AO~0235+164 as part of the 6.7\,GHz COSMIC program. The presence of a flux density excursion below the underlying mean flux density of the source is consistent with an ESE cloud passing through the line of sight.
268:
269: Assuming an ESE cloud velocity of 20\,km\,s$^{-1}$, we estimate the cloud to have a linear scale size ranging from 0.09 -- 0.90\,AU and distance less than 3.6\,kpc. We estimate the minimum electron density for the cloud to produce an ESE is $4 \times 10^3$\, cm$^{-3}$.% higher density than previously detected ESEs by \citet{fiedleretal87}.
270:
271: The detection and characterisation of this ESE required the dense flux density
272: monitoring of the COSMIC program. The rapidity of this ESE suggests that perhaps other rapid ESEs may be overlooked in other monitoring programs due to undersampling of the events.
273:
274: \acknowledgements
275: This research has been supported by the ARC grant number DP0342500 at the University of Tasmania.
276: %Financial support for the COSMIC project is provided by the Australian Research Council.
277: We thank Bev Bedson for her vital contribution to the operation of the Ceduna observatory and the
278: COSMIC project.
279: We thank Stefan Dieters for searching the archives of satellite based observatories for events coincident with the ESE detection.
280: The National Radio Astronomy Observatory is a Facility of the National Science Foundation operated under co-operative agreement by Associated Universities, Inc.
281:
282:
283:
284:
285: \begin{thebibliography}{99}
286:
287: \bibitem[Altschuler \& Wardle(1977)]{altschuler1977} Altschuler,
288: D.~R., \& Wardle, J.~F.~C.\ 1977, \mnras, 179, 153
289:
290: \bibitem[Baars et al.(1977)]{baars1977} Baars, J.~W.~M., Genzel,
291: R., Pauliny-Toth, I.~I.~K., \& Witzel, A.\ 1977, \aap, 61, 99
292:
293: \bibitem[Bach et al.(2007)]{bach2007} Bach, U., et al.\ 2007, \aap, 464, 175
294:
295: \bibitem[Burbidge et al.(1976)]{burbidge1976} Burbidge, E.~M.,
296: Caldwell, R.~D., Smith, H.~E., Liebert, J., \& Spinrad, H.\ 1976, \apjl,
297: 205, L117
298:
299:
300: \bibitem[Bower et al.(2007)]{bower2007} Bower, G.~C., Saul, D.,
301: Bloom, J.~S., Bolatto, A., Filippenko, A.~V., Foley, R.~J., \& Perley, D.\
302: 2007, \apj, 666, 346
303:
304: \bibitem[Clegg et al.(1998)]{clegg98} Clegg, A.W., Fey, A.L. \& Lazio, T.J.W., \apj, 496, 253
305:
306: \bibitem[Cohen et al.(1987)]{cohen1987} Cohen, R. D., Smith, H. E., Junkkarinen, V. T., Burbidge, E. M. 1987, ApJ 318, 577
307:
308: \bibitem[Condon et al.(1979)]{condon1979} Condon, J. J., Ledden, J. E., O'Dell, S.L. \& Dennison B. 1979, AJ, 84, 1
309:
310: \bibitem[Fiedler et al.(1987)]{fiedleretal87} Fiedler, R.L. et al. 1987, \apjs, 65, 319
311:
312: \bibitem[Fiedler et al.(1994)]{fiedleretal94} Fiedler, R.L. et al. 1994, \apj, 430, 581
313:
314: \bibitem[Frey et al.(2000)]{frey2000} Frey, S., et al.\ 2000,
315: \pasj, 52, 975
316:
317: \bibitem[Jauncey et al.(2007)]{jauncey2007} Jauncey, D. L., et al, in press.
318:
319: \bibitem[Kraus et al.(1999)]{kraus1999} Kraus, A., et al.\ 1999,
320: \aap, 344, 807
321:
322: \bibitem[Lovell et al.(2003)]{lovell2003} Lovell, J. E. J., et al.\ 2003, \apj, 126, 1699
323:
324: \bibitem[Lovell et al.(2007)]{lov2007} Lovell, J.E.J.,et al.\ 2007, ASP Conference Series, 365, 279
325:
326: \bibitem[MacLeod et al.(1976)]{macleod1976} MacLeod, J.~M., Andrew,
327: B.~H., \& Harvey, G.~A.\ 1976, \nat, 260, 751
328:
329: \bibitem[Macquart \& Jauncey(2002)]{mac2002} Macquart, J.-P., \& Jauncey, D. L. 2002, ApJ, 572, 786
330:
331: \bibitem[Matsumura et al.(2007)]{matsumura2007} Matsumura, N., et
332: al.\ 2007, \aj, 133, 1441
333:
334: \bibitem[McCulloch et al.(2005)]{mcculloch2005} McCulloch, P., Ellingsen, S.P., Jauncey, D.L., Carter, S.J.B., Cim\`o, G., Lovell, J.E.J. \& Dodson, R.G. 2005, \aj, 129, 2034
335:
336: \bibitem[Niinuma et al.(2007)]{niinuma2007} Niinuma, K., et al.\
337: 2007, \apjl, 657, L37
338:
339: \bibitem[Ostriker \& Vietri(1985)]{ostriker1985} Ostriker, J.~P., \&
340: Vietri, M.\ 1985, \nat, 318, 446
341:
342: \bibitem[Piner et al.(2006)]{piner2006} Piner, B.~G., Bhattarai,
343: D., Edwards, P.~G., \& Jones, D.~L.\ 2006, \apj, 640, 196
344:
345: \bibitem[Raiteri et al.(2006)]{raiteri2006} Raiteri, C.~M., et al.\
346: 2006, \aap, 459, 731
347:
348: \bibitem[Rieke et al.(1976)]{rieke1976} Rieke, G.~H., Grasdalen,
349: G.~L., Kinman, T.~D., Hintzen, P., Wills, B.~J., \& Wills, D.\ 1976, \nat,
350: 260, 754
351:
352:
353: \bibitem[Romani et al.(1987)]{romani87} Romani, R.W., Blandford, R.D. \& Cordes, J.M. 1987, \nat, 328, 324
354:
355: \bibitem[Walker \& Wardle(1998)]{ww98} Walker, M.A. \& Wardle, M. 1998, \apj, 498, L125
356: \end{thebibliography}
357:
358: \clearpage
359:
360: \begin{figure}
361: \epsscale{1.3}
362: \centerline{\plotone{f1.eps}}
363: \caption{The 6.65\,GHz lightcurve of AO~0235$+$164 (points) since 2003 Jan 1 (MJD 52640). The corresponding lightcurve of the calibrator, 3C~277, is shown (crosses), with 1.6\,Jy subtracted from its flux densities for display purposes. The averaging time for each flux density sample is 12 hours.\label{fig:entirelc}}
364: \end{figure}
365:
366: \begin{figure*}
367: \epsscale{1.4}
368: \centerline{\plotone{f2.eps}}
369: \caption{The 6.65\,GHz lightcurve of AO~0235+164 during the rapid flux density variation in 2005, The averaging time for each flux density sample is 12 hours. The points marked by crosses denote 8\,GHz measurements reported by \citet{bach2007}. \label{fig:mainlc}}
370: \end{figure*}
371:
372: \end{document}
373: