0706.1518/ms.tex
1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass[preprint2]{aastex}
3: %\documentclass[preprint2,longabstract]{aastex}
4: %\documentclass[onecolumn]{emulateapj}
5: 
6: \usepackage{graphicx}
7: \usepackage{natbib}
8: 
9: \newcommand{\gcn}{GCN Circular}
10: \newcommand{\thisgrb}{GRB\,051022}
11: \newcommand{\swift}{\textit{Swift}}
12: \newcommand{\hete}{HETE-2}
13: \newcommand{\swcommand}[1]{\textsf{#1}}
14: \newcommand{\obsname}[1]{#1}
15: \newcommand{\nh}{\ensuremath{N_\mathrm{H}}}
16: \newcommand{\plusminus}[2]{\ensuremath{^{+#1}_{-#2}}}
17: \newcommand{\ebv}{\ensuremath{E_{B-V}}}
18: \newcommand{\chisqred}{\ensuremath{\chi^2_{\mathrm{red}}}}
19: \newcommand{\chisqdof}{\ensuremath{\chi^2/\mathrm{d.o.f.}}}
20: \newcommand{\num}{\ensuremath{\nu_{\mathrm{m}}}}
21: \newcommand{\nuc}{\ensuremath{\nu_{\mathrm{c}}}}
22: \newcommand{\nua}{\ensuremath{\nu_{\mathrm{a}}}}
23: \newcommand{\Fp}{\ensuremath{F_{\nu,\mathrm{max}}}}
24: \newcommand{\tjet}{\ensuremath{t_{\mathrm{j}}}}
25: \newcommand{\Eiso}{\ensuremath{E_{\mathrm{iso}}}}
26: \newcommand{\thetajet}{\ensuremath{\theta_{\mathrm{j}}}}
27: \newcommand{\Epromptjet}{\ensuremath{E_{\mathrm{p,jet}}}}
28: \newcommand{\Epromptiso}{\ensuremath{E_{\mathrm{p,iso}}}}
29: \newcommand{\Ejet}{\ensuremath{E_{\mathrm{jet}}}}
30: \newcommand{\epse}{\ensuremath{\varepsilon_{\mathrm{e}}}}
31: \newcommand{\epsB}{\ensuremath{\varepsilon_{\mathrm{B}}}}
32: \newcommand{\Epeak}{\ensuremath{E_{\mathrm{peak}}}}
33: 
34: \hyphenation{nijmegen arcminute fluence}
35: 
36: \shorttitle{\thisgrb\ as a prototype dark burst}
37: \shortauthors{Rol et al.}
38: 
39: 
40: \begin{document}
41: 
42: 
43: \title{\thisgrb: physical parameters and extinction of a prototype
44:   dark burst}
45: 
46: 
47: \author{
48: Evert~Rol\altaffilmark{1},
49: Alexander~van~der~Horst\altaffilmark{2},
50: Klaas~Wiersema\altaffilmark{2},
51: Sandeep~K.~Patel\altaffilmark{3,4},
52: Andrew~Levan\altaffilmark{5},
53: Melissa~Nysewander\altaffilmark{6},
54: Chryssa~Kouveliotou\altaffilmark{3,7},
55: Ralph~A.\,M.\,J.~Wijers\altaffilmark{2},
56: Nial~Tanvir\altaffilmark{1},
57: Dan~Reichart\altaffilmark{8},
58: Andrew~S.~Fruchter\altaffilmark{6},
59: John~Graham\altaffilmark{6,9},
60: Jan-Erik~Ovaldsen\altaffilmark{10},
61: Andreas~O.~Jaunsen\altaffilmark{10},
62: Peter~Jonker\altaffilmark{11,12,13},
63: Wilbert~van~Ham\altaffilmark{14},
64: Jens~Hjorth\altaffilmark{15},
65: Rhaana~L.\,C.~Starling\altaffilmark{1},
66: Paul~T.~O'Brien\altaffilmark{1},
67: Johan~Fynbo\altaffilmark{15},
68: David~N.~Burrows\altaffilmark{16},
69: Richard~Strom\altaffilmark{2,17}
70: }
71: 
72: 
73: \altaffiltext{1}{Department of Physics and Astronomy, University of
74:   Leicester, University Road, Leicester, LE1\,7RH, United Kingdom;
75:   evert.rol@star.le.ac.uk}
76: 
77: \altaffiltext{2}{Astronomical Institute, University of Amsterdam,
78:   Kruislaan 403, NL-1098~SJ Amsterdam, The Netherlands}
79: 
80: \altaffiltext{3}{National Space Science
81:   and Technology Center, 320 Sparkman Drive, Huntsville, AL-35805,
82:   USA}
83: 
84: \altaffiltext{4}{Optical Sciences Corporation, 6767 Old Madison Pike,
85:   Suite 650, Huntsville, AL, 35806}
86: 
87: \altaffiltext{5}{Department of Physics, University of Warwick,
88:   Coventry, CV4 7AL, UK}
89: 
90: \altaffiltext{6}{Space Telescope Science Institute, 3700 San Martin
91:   Drive, Baltimore, MD-21218}
92: 
93: \altaffiltext{7}{NASA Marshall Space Flight Center}
94: 
95: \altaffiltext{8}{Department of Physics and Astronomy, University of
96:   North Carolina at Chapel Hill, Campus Box 3255, Chapel Hill,
97:   NC-27599, USA}
98: 
99: \altaffiltext{9}{Department of Physics and Astronomy, Johns Hopkins
100:   University, 3400 North Charles St., Baltimore, MD-21218}
101: 
102: \altaffiltext{10}{Institute of Theoretical Astrophysics, University of
103:   Oslo, P.O.Box 1029, Blindern, N-0315 Oslo, Norway} 
104: 
105: \altaffiltext{11}{SRON, Netherlands Institute for Space Research,
106:   Sorbonnelaan 2, NL-3584~CA Utrecht, The Netherlands}
107: 
108: \altaffiltext{12}{Harvard--Smithsonian Center for Astrophysics, 60
109:   Garden Street, Cambridge, MA-02138, Massachusetts, USA}
110: 
111: \altaffiltext{13}{Astronomical Institute, Utrecht University, P.O.Box
112:   80000, 3508 TA, Utrecht, The Netherlands}
113: 
114: \altaffiltext{14}{Department of Astrophysics, Radboud University
115:   Nijmegen, P.O.Box 9010, NL-6500~GL Nijmegen, The Netherlands}
116: 
117: \altaffiltext{15}{Dark Cosmology Centre, Niels Bohr Institute,
118:   University of Copenhagen, Juliane Maries Vej 30, DK-2100 Copenhagen
119:   \O, Denmark}
120: 
121: \altaffiltext{16}{Penn State University, State College, PA 16801, USA}
122: 
123: \altaffiltext{17}{ASTRON, P.O. Box 2, NL-7990~AA Dwingeloo, Netherlands}
124: 
125: 
126: 
127: \begin{abstract}
128:   \thisgrb\ was undetected to deep limits in early optical
129:   observations, but precise astrometry from radio and X-ray showed
130:   that it most likely originated in a galaxy at $z\approx0.8$.  We
131:   report radio, optical, near infra-red and X-ray observations of
132:   \thisgrb.  Using the available X-ray and radio data, we model the
133:   afterglow and calculate the energetics of the afterglow, finding it
134:   to be an order of magnitude lower than that of the prompt emission.
135:   The broad-band modeling also allows us to precisely define various
136:   other physical parameters and the minimum required amount of
137:   extinction, to explain the absence of an optical afterglow. Our
138:   observations suggest a high extinction, at least $2.3$ magnitudes in
139:   the infrared ($J$) and at least 5.4 magnitudes in the optical ($U$) in
140:   the host-galaxy restframe. Such high extinctions are unusual for
141:   GRBs, and likely indicate a geometry where our line of sight to the
142:   burst passes through a dusty region in the host that is not directly
143:   co-located with the burst itself.
144: 
145: \end{abstract}
146: 
147: 
148: \keywords{gamma rays: bursts --- dust, extinction}
149: 
150: 
151: 
152: \section{Introduction} \label{section:intro}
153: 
154: Dark gamma-ray bursts (GRBs) --- at the most basic level those without
155: optical afterglows --- are a long-standing issue in GRB observations.
156: Although in many cases the non-detection of an afterglow at optical
157: wavelengths may simply be due to an insufficiently deep search, or one
158: which takes place at late times \citep[e.g.][]{fynbo2001:aa369:373}, a
159: subset of GRBs with bright X-ray afterglows remains undetected despite
160: prompt and deep optical searches \citep[e.g.][]{groot1998:apj493:27}
161: and directly implies suppression of the optical light.
162: 
163: There are several plausible explanations for this, the most likely
164: being that the burst is at high redshift, such that the Ly-alpha break
165: has crossed the passband in question, or that there is high extinction
166: in the direction of the GRB. Examples of both have been found, with a
167: small number of GRBs at $z>5$ appearing as $V$ and $R$ band dropouts
168: \citep[e.g.][]{jakobsson2006:aa447:897, haislip2006:nat440:181} and
169: some GRB afterglows appearing very red at lower redshift, due to
170: effects of extinction
171: \citep[e.g.][]{levan2006:apj647:471,rol2007:mnras374:1078}.
172: 
173: Identification of GRBs at very high redshifts is the key to using them
174: as cosmological probes.  The proportion of bursts exhibiting
175: high dust extinction is also interesting from the point of view of
176: estimating the proportion of star formation that
177: is dust enshrouded, as well as understanding the environments
178: which favor GRB production \citep{trentham2002:mnras334:983,mnras2004:352:1073}.
179: 
180: The detection and follow-up of dark bursts at other
181: wavelengths is essential, as it enables 1) the modeling of the
182: afterglow, deriving estimates of the extinction and energies involved,
183: potentially providing information about the direct burst environment, 2)
184: pinpointing the burst position in the host, to enable late-time
185: high resolution imaging and the detection of dust enhanced regions in
186: the host, and 3) determination of the properties of the GRB host itself,
187: such as the SFR and average host-galaxy extinction.
188: 
189: 
190: The High Energy Transient Explorer 2 mission (HETE-2;
191: \citealt{ricker2003:aipc662:3}) detected and located an unusually
192: bright gamma-ray burst \citep{olive2005:gcn4131}
193: %(GRB) 
194: with its three main instruments, the French Gamma Telescope (FREGATE),
195: the Wide field X-ray monitor (WXM) and the Soft X-ray Camera, (SXC), on
196: October 22, 2005. A 2.5 arcminute localization was sent out within
197: minutes, enabling prompt follow-up observations
198: \citep[e.g.][]{torii2005:gcn4130,schaefer2005:gcn4132}; a
199: target-of-opportunity observation was also performed with \swift.
200: Details of the HETE-2 observations can be found in
201: \citet{nakagawa2006:pasjl58:35}.
202: 
203: The \swift\ observations resulted in the detection of a single fading
204: point source inside the SXC error region, which was consequently
205: identified as the X-ray afterglow of \thisgrb\
206: \citep{racusin2005:gcn4141}. However, optical and near infra-red (nIR)
207: observations failed to reveal any afterglow to deep limits, while
208: radio and millimeter observations with the Very Large Array (VLA), the
209: Westerbork Synthesis Radio Telescope (WSRT) and the Plateau de Bure
210: Interferometer detected the radio counterpart
211: \citep{cameron2005:gcn4154,vanderhorst2005:gcn4158,bremer2005:gcn4157}.
212: The position coincides with its likely host galaxy
213: \citep{berger2005:gcn4148} at a redshift of $z = 0.8$
214: \citep{gal-yam2005:gcn4156}.
215: 
216: In this paper, we describe our X-ray, optical, nIR and radio
217: observations of \objectname[GRB 051022]{\thisgrb}.  The outline of the
218: paper is as follows: in Section \ref{section:obs} we describe our
219: observations, data reduction and initial results. In Section
220: \ref{section:analysis}, we analyze these results and form our
221: afterglow picture, which is discussed in Section
222: \ref{section:discussion}. Our findings are summarized in Section
223: \ref{section:conclusions}.
224: 
225: In the following, we have used $F \propto \nu^{-\beta} t^{-\alpha}$ in
226: our definition of $\alpha$ and $\beta$. We assume a cosmology with
227: $H_0 = 71\, \mathrm{km} \mathrm{s}^{-1} \mathrm{Mpc}^{-1}$, $\Omega_M =
228: 0.27$ and $\Omega_{\Lambda} = 0.73$. All quoted errors in this paper
229: are 1 sigma (68\%) errors.
230: 
231: 
232: \section{Observations and data reduction} \label{section:obs}
233: 
234: 
235: \subsection{X-ray observations} \label{section:obs:X-ray}
236: 
237: X-ray observations were performed with the \swift\ X-Ray Telescope
238: (XRT) and the Chandra X-ray Observatory (CXO).
239: 
240: The XRT started observing the afterglow of \thisgrb\ 3.46 hours after
241: the \hete\ trigger, for a total effective integration time of 137~ks
242: between October 22 and November 6.
243: 
244: Observations were performed in Photon Counting (PC) mode, the most
245: sensitive observing mode. We reduced the data using the \swift\
246: software version 2.6 in the HEAsoft package version 6.2.0. Data were
247: obtained from the quick-look site and processed from level 1 to level
248: 2 FITS files using the \swcommand{xrtpipeline} tool in its standard
249: configuration.  The first two orbits (until $2.1 \times 10^4$ seconds
250: post burst) show pile-up and were therefore extracted with an annular
251: rather than circular region, with an inner radius of 19 and 12\arcsec\
252: for orbits 1 and 2, respectively, and an outer radius of 71\arcsec.
253: Orbits 3~--~7 ($2.4 \times 10^4$~--~$4.9 \times 10^4$ seconds) were
254: extracted with a circular region of 71\arcsec\ radius, and later
255: orbits were extracted using a 47\arcsec\ radius circle instead. The
256: data for the light curve were extracted between channels 100 and 1000,
257: corresponding to 1 and 10 keV, respectively; while the commonly used
258: range is 0.3~--~10 keV, the large absorption prevents the detection of
259: any data from the source below 1 keV.  Otherwise, the procedure is
260: similar to that described in \citet{evans2007:aap469:379}.
261: 
262: 
263: Observations with the CXO started on October 25, 2005, 21:14:20, 3.34
264: days after the HETE trigger, for a total integration time of 20~ks
265: \citep{patel2005:gcn4163}.  Data were reduced in a standard fashion
266: with the CIAO package.
267: 
268: We performed astrometry by matching X-ray sources with an optical
269: $R$-band image that was astrometrically calibrated to the 2MASS
270: catalog. Our CXO position is RA, Dec = 23:56:04.115, +19:36:24.04
271: (J2000), with positional errors of 0.33\arcsec\ and 0.12\arcsec\ for
272: the Right Ascension and Declination, respectively. This puts the
273: afterglow within 0.5\arcsec\ of the center of its host galaxy.
274: 
275: We modeled the XRT spectra with an absorbed power law in XSpec
276: \citep{arnaud1996:proc:17}, using data from the first seven orbits. A
277: good fit ($\chisqdof = 87.2/99$) was obtained with a resulting
278: spectral energy index of $\beta = 1.00 \pm 0.12$ and excess absorption
279: (at $z = 0.8$ and for assumed Galactic abundances) of $\nh = (2.82 \pm
280: 0.46) \times 10^{22}\ \mathrm{cm}^{-2}$ on top of the estimated
281: Galactic absorption at this position ($\nh = 4.06 \times 10^{20}\
282: \mathrm{cm}^{-2}$, \citealt{dickey1990:araa28:215}). The CXO data are
283: fully in agreement with these values, showing no change in the
284: spectrum over time between 0.3 and 3.3 days after the burst. The
285: absorption measured is far less than that measured by the HETE team in
286: their prompt data, $\nh = (8.8\plusminus{1.9}{1.8}) \times
287: 10^{22}\,\mathrm{cm}^{-2}$ \citep{nakagawa2006:pasjl58:35}. This could
288: indicate a change in absorption between the early (prompt)
289: measurements and those at the time of the XRT observations. For the
290: prompt emission spectrum, however, the values found by Konus-Wind
291: \citep{golenetskii2005:gcn4150} are rather different than those found
292: by HETE-2, and may be the result of the lower energy cut-off for
293: FREGATE compared to Konus-wind. Alternatively, the fact that these
294: spectra are an average over the whole emission period may also result
295: in incorrect model parameters. In the two last cases, the \nh\ in the
296: prompt emission could be as low as the XRT value and still produce an
297: equally well fit, but with slightly different model parameters.
298: 
299: For the XRT data, \citet{butler2005:gcn4165} and
300: \citet{nakagawa2006:pasjl58:35} find a value somewhat higher than our
301: value ($4.9 \times 10^{22}\,\mathrm{cm}^{-2}$ and $5.3 \times
302: 10^{22}\,\mathrm{cm}^{-2}$ respectively, when scaled by $(1+z)^3$,
303: \citealt{gunn1965:apj142:1633}). This difference could be explained by
304: a different count-binning or an updated XRT calibration used in our
305: modeling.
306: 
307: The XRT light curve count rates have been converted to 1--10~keV
308: fluxes using the results from our spectral modeling and calculating
309: the ratio of the flux and count rate at the logarithmic center of the
310: orbits. The 1~--~10 keV CXO flux was derived using the actual spectral fit.
311: 
312: A broken power law fit to the X-ray light curve results in $\alpha_1 =
313: 1.16 \pm 0.06$, $\alpha_2 = 2.14 \pm 0.17$ and a break time of
314: $110\plusminus{21}{23}$ ks, or around 1.27 days. The difference between
315: $\alpha_1$ and $\alpha_2$, and the fact that the spectral slope does
316: not change across the break (the CXO measurement is past the break),
317: are highly indicative that the observed break in the light curve is a jet
318: break. In Section \ref{section:broadband-modeling}, we perform full
319: modeling of the afterglow using the fireball model, indeed resulting
320: in a jet-break time \tjet\ that agrees reasonably well with the break time as
321: determined from only the X-rays.  We point out that our value for
322: \tjet\ is different than that cited in \citet{racusin2005:gcn4169},
323: largely because their measurement of \tjet\ was based on a preliminary
324: XRT light curve.
325: 
326: 
327: 
328: 
329: \subsection{Optical and near infra-red observations} 
330: \label{section:obs:optical}
331: 
332: Observations were obtained in $Z$ and $R$-band with the William
333: Herschel Telescope (WHT) using the Auxiliary Port and the Prime Focus
334: Imaging Camera, respectively, in $r^{\prime}i^{\prime}z^{\prime}$ with
335: the Gemini South telescope using the GMOS instrument, in $JHK_s$ with
336: the Wide Field Camera on the United Kingdom InfraRed Telescope
337: (UKIRT), in $BVRI$ with the DFOSC instrument on the Danish 1.54m
338: telescope and in $J$ and $K_s$ with the Southern Astrophysical
339: Research (SOAR) telescope using OSIRIS. The optical data were reduced
340: in a standard fashion using the \swcommand{ccdproc} package within the
341: IRAF software \citep{tody1986:proc:733}, whereas the SOAR data were
342: reduced using the \swcommand{cirred} package within IRAF. The UKIRT
343: data were reduced using the standard pipeline reduction for WFCAM.
344: 
345: 
346: Photometric calibration was done using the calibration provided by
347: \citet{henden2005:gcn4184} for Johnson-Cousins filters. For the
348: $r^{\prime}i^{\prime}z^{\prime}$ GMOS filters, we converted the
349: magnitudes of the calibration stars provided by Henden to the Sloan
350: filter system using the transformations provided by
351: \citet{jester2005:aj130:873}, and verified by the published GMOS zero
352: points. The WHT $Z$-band was calibrated using the spectroscopic
353: standard star SP2323+157.  Calibration of the infrared $JHK$
354: magnitudes was done using the 2MASS catalog
355: \citep{skrutskie2006:aj131:1163}.
356: 
357: No variable optical source was found at the position of the X-ray and
358: radio afterglow. For the early epoch images ($<1$ day post burst), we
359: estimated a limiting magnitude by performing image subtraction between
360: this and a later image using the ISIS image subtraction package
361: \citep{alard2000:aaps144}. To this end, artificial low signal-to-noise
362: sources were added onto the images, with a Gaussian PSF matched in
363: size to the seeing (some artificial sources were added on top of
364: existing sources, e.g. galaxies, some on the background sky). We
365: determined our upper limit to be the point where we could retrieve
366: 50\% of the artificial sources in the subtracted image. This assumes that
367: the change in brightness of any point source on top of the host
368: galaxy is sufficient to be seen in such a subtracted image. With the
369: difference in time between the epochs, this seems a reasonable
370: assumption (for example, for a source fading with a shallow power law like
371: slope of $F \propto t^{-0.5}$, the magnitude difference between the
372: two WHT $Z$-band observations is $\approx 0.6$ magnitudes).
373: 
374: Photometry of the host galaxy has been performed using aperture
375: photometry, with an aperture 1.5 times the seeing for each image,
376: estimated from the measured FWHM of the PSF for point sources in the
377: images.
378: 
379: Table \ref{table:obslog} shows the log of our optical/nIR
380: observations, while Table \ref{table:limmags} shows the upper limits
381: for any optical/nIR afterglow.
382: 
383: 
384: \begin{deluxetable}{lccccc}
385: %\tabletypesize{\scriptsize}
386:   \tablecaption{Overview of optical observations\label{table:obslog}}
387: %\tablewidth{\columnwidth}
388: \tablehead{
389: \colhead{Start date} &
390: \colhead{$\Delta T$ (average)} &
391: \colhead{exposure time} &
392: \colhead{filter} &
393: \colhead{seeing} &
394: \colhead{telescope \& instrument} \\
395: &
396: \colhead{(days)} &
397: \colhead{(seconds)} &
398: &
399: \colhead{(arcsec)} &
400: }
401: \startdata
402: 2005-10-22T23:25:14  &  0.4287   &  1800  &  $Z$      &  0.8  &  WHT + API        \\
403:                                                       
404: 2005-10-23T00:22:33  &  0.4684   &  1620  &  $J$      &  1.2  &  SOAR + OSIRIS    \\
405: 2005-10-23T00:56:00  &  0.4917   &  1620  &  $K_s$    &  1.3  &  SOAR + OSIRIS    \\
406:                                                       
407: 2005-10-23T00:48:03  &  0.5144   &  1920  &  $i^{\prime}$     &  0.6  &  Gemini South + GMOS   \\
408: 2005-10-23T01:07:53  &  0.5288   &  1920  &  $r^{\prime}$     &  0.6  &  Gemini South + GMOS   \\
409: 2005-10-23T01:27:46  &  0.5426   &  1920  &  $z^{\prime}$     &  0.5  &  Gemini South + GMOS   \\
410:                                                       
411: %2005-10-23T05:55:52 &  0.7076   &  1080  &  $Z$      &  1.3  &  UKIRT + WFCAM    \\
412: 2005-10-23T06:31:03  &  0.7525   &  720   &  $J$      &  1.4  &  UKIRT + WFCAM    \\
413: 2005-10-23T06:36:39  &  0.7526   &  360   &  $H$      &  1.3  &  UKIRT + WFCAM    \\
414: 2005-10-23T06:47:59  &  0.7604   &  360   &  $K$      &  1.3  &  UKIRT + WFCAM    \\
415:                                                       
416: 2005-10-23T21:15:57  &  1.3389   &  1200  &  $Z$      &  1.0  &  WHT + API        \\
417:                                                       
418: 2005-10-24T09:35:10  &  1.8467   &  720   &  $K$      &  0.3  &  UKIRT + WFCAM    \\
419:                                                       
420: 2005-10-25T01:34:03  &  2.5181   &  1602  &  $K_s$    &  1.3  &  SOAR + OSIRIS    \\
421: 2005-10-25T02:13:18  &  2.5454   &   720  &  $J$      &  1.2  &  SOAR + OSIRIS    \\
422:                                                       
423: 2005-10-25T02:22:02  &  2.5698   &  1920  &  $r^{\prime}$     &  1.1  &  Gemini South + GMOS   \\
424: 2005-10-25T02:39:59  &  2.5792   &  1440  &  $z^{\prime}$     &  1.2  &  Gemini South + GMOS   \\
425:                                                       
426: 2005-10-26T00:36:58  &  3.4785   &  1800  &  $R$      &  1.4  &  WHT+PFIP         \\
427: 
428: 2005-10-26T02:48:06  &  3.5695   &   600  &  Gunn $i$ &  1.4  &  DK1.54m + DFOSC  \\
429: 2005-10-26T03:23:35  &  3.5942   &   600  &  $R$      &  1.9  &  DK1.54m + DFOSC  \\
430: 2005-10-27T01:01:04  &  4.4952   &   600  &  $B$      &  2.3  &  DK1.54m + DFOSC  \\
431: 2005-10-27T02:59:20  &  4.5773   &   600  &  $R$      &  1.6  &  DK1.54m + DFOSC  \\
432: 2005-10-27T02:00:48  &  4.5367   &   600  &  $V$      &  1.8  &  DK1.54m + DFOSC  \\
433: 2005-10-28T02:18:38  &  5.5491   &   600  &  $i$      &  1.4  &  DK1.54m + DFOSC  \\
434: 2005-10-30T02:32:59  &  7.5590   &   600  &  $B$      &  1.8  &  DK1.54m + DFOSC  \\
435: 2005-10-30T04:18:30  &  7.6323   &   600  &  $U$      &  1.8  &  DK1.54m + DFOSC  \\
436: 2005-10-30T01:33:57  &  7.5180   &   600  &  $V$      &  1.4  &  DK1.54m + DFOSC  \\
437: 2005-10-31T03:19:05  &  8.5910   &   600  &  $B$      &  1.0  &  DK1.54m + DFOSC  \\
438: 2005-10-31T01:03:40  &  8.4970   &   600  &  $R$      &  1.0  &  DK1.54m + DFOSC  \\
439: 2005-10-31T02:10:02  &  8.5431   &   600  &  $V$      &  1.0  &  DK1.54m + DFOSC  \\
440: 2005-11-01T01:52:57  &  9.5312   &   600  &  $R$      &  0.9  &  DK1.54m + DFOSC  \\
441: 2005-11-02T02:04:47  &  10.539   &   600  &  $V$      &  1.2  &  DK1.54m + DFOSC  \\
442: 2005-11-03T01:10:34  &  11.502   &   600  &  $B$      &  1.2  &  DK1.54m + DFOSC  \\
443: 2005-11-07T01:25:30  &  15.512   &   600  &  Gunn $i$      &  1.4  &  DK1.54m + DFOSC  \\
444: 2005-11-08T01:40:48  &  16.523   &   600  &  Gunn $i$      &  1.4  &  DK1.54m + DFOSC  \\
445: \enddata
446: \end{deluxetable}
447: 
448: \begin{deluxetable}{ccccc}
449: \tablecaption{Limiting magnitudes\label{table:limmags}}
450: \tablehead{
451: \colhead{filter} &
452: \colhead{limiting magnitude\tablenotemark{a}} &
453: \colhead{$\Delta T$ (average)} &
454: \colhead{frequency} &
455: \colhead{specific flux\tablenotemark{b}} \\
456: &
457: &
458: \colhead{days} &
459: \colhead{Hz} &
460: \colhead{$\mu$Jy}
461: }
462: \startdata
463: $K_s$    &  $>20.0$   &  0.4917    &    $1.40 \cdot 10^{14}$   &   $< 6.82 $  \\
464: $J$      &  $>20.3$   &  0.4684    &    $2.40 \cdot 10^{14}$   &   $<12.3  $  \\
465: $Z$      &  $>22.9$   &  0.4287    &    $3.43 \cdot 10^{14}$   &   $< 2.66 $  \\
466: $z^{\prime}$     &  $>23.5$   &  0.5426    &    $3.36 \cdot 10^{14}$   &   $< 1.53 $  \\
467: $r^{\prime}$     &  $>25.3$   &  0.5288    &    $4.76 \cdot 10^{14}$   &   $< 0.305$  \\
468: \enddata
469: \tablenotetext{a}{See text for the definition of the limiting
470:   magnitude.}  \tablenotetext{b}{Specifc fluxes have been corrected for a
471:   Galactic extinction value of $E_{B-V} = 0.04$
472:   \citep{schlegel1998:apj500:525}, and converted from magnitudes using
473:   the calibration by \citet{tokunaga2005:pasp117:421} for the $JK_s$
474:   filters; the other filters are on the magnitude AB-system
475:   \citep{oke1983:apj226:713}}
476: \end{deluxetable}
477: 
478: \subsection{Radio observations} \label{section:obs:radio}
479: 
480: Radio observations were performed with the WSRT at 8.4 GHz, 4.9 GHz
481: and 1.4 GHz.  We used the Multi Frequency Front Ends
482: \citep{tan1991:aspc19:42} in combination with the IVC+DZB back
483: end\footnote{See sect. 5.2 at
484:   \url{http://www.astron.nl/wsrt/wsrtGuide/node6.html}} in continuum mode,
485: with a bandwidth of 8x20 MHz. Gain and phase calibrations were
486: performed with the calibrators 3C~286 and 3C~48, although at one 8.4
487: GHz measurement 3C~147 was used. Reduction and analysis were performed
488: using the MIRIAD software
489: package\footnote{\url{http://www.atnf.csiro.au/computing/software/miriad}}.
490: The observations are detailed in Table \ref{table:radiolog}. In our
491: modeling described in section \ref{section:broadband-modeling} we
492: have also used the VLA radio detection at 8.5 GHz from
493: \citet{cameron2005:gcn4154}.
494: 
495: \begin{deluxetable}{lcccc}
496: %\tabletypesize{\scriptsize}
497:   \tablecaption{Overview of \obsname{WSRT} radio
498:     observations\label{table:radiolog}}
499: %\tablewidth{\columnwidth}
500: \tablehead{
501: \colhead{Start date} &
502: \colhead{$\Delta T$ (average)} &
503: \colhead{integration time} &
504: \colhead{frequency} &
505: \colhead{specific flux} \\
506: &
507: \colhead{(days)} &
508: \colhead{(hours)} &
509: \colhead{(GHz)} &
510: \colhead{($\mu$Jy)}
511: }
512: 
513: \startdata
514: 2005-11-04T18:14:24  &  13.37   &  4.0   &  8.5   &  38 $\pm$ 132 \\
515: 2005-11-08T14:19:41  &  17.19   &  7.0   &  8.5   &  28 $\pm$  97 \\
516: 2005-10-23T15:20:10  &   1.19   &  5.0   &  4.9   & 281 $\pm$  32 \\
517: 2005-10-24T15:17:17  &   2.22   &  6.2   &  4.9   & 342 $\pm$  34 \\
518: 2005-10-25T15:12:58  &   3.30   &  5.4   &  4.9   & 143 $\pm$  30 \\
519: 2005-10-28T18:33:08  &   6.40   &  8.5   &  4.9   &  91 $\pm$  28 \\
520: 2005-10-30T18:00:00  &   8.32   &  5.8   &  4.9   & 138 $\pm$  28 \\
521: 2005-11-01T18:00:00  &  10.38   &  8.9   &  4.9   & 169 $\pm$  28 \\
522: 2005-11-04T17:31:12  &  13.37   &  4.6   &  4.9   &  70 $\pm$  34 \\
523: 2005-10-25T15:56:10  &  3.33    &  5.4   &  1.4   &   8 $\pm$  78 \\
524: \enddata
525: \end{deluxetable}
526: 
527: 
528: 
529: \section{Analysis} \label{section:analysis}
530: 
531: \subsection{Broadband modeling}
532: \label{section:broadband-modeling}
533: 
534: 
535: We have performed broadband modeling of the X-ray and radio
536: measurements, using the methods presented in \citet{vanderhorst2007:submitted}.  In our modeling we assume a purely
537: synchrotron radiation mechanism.
538: 
539: The relativistic blastwave causing the afterglow accelerates electrons
540: to relativistic velocities, which gives rise to a broadband spectrum
541: with three characteristic frequencies: the peak frequency \num,
542: corresponding to the minimum energy of the relativistic electrons that
543: are accelerated by the blastwave, the cooling frequency \nuc,
544: corresponding to the electron energy at which electrons lose a
545: significant fraction of their energy by radiation on a timescale that
546: is smaller than the dynamical timescale, and the self-absorption
547: frequency \nua, below which synchrotron self-absorption produces
548: significant attenuation.  The broadband spectrum is further
549: characterized by the specific peak flux \Fp\ and the slope $p$ of the electron
550: energy distribution.
551: 
552: The dynamics of the relativistic blastwave determine the temporal
553: behavior of the broadband synchrotron spectrum, i.e. the light curves
554: at given frequencies.  At first the blastwave is extremely
555: relativistic, but is decelerated by the surrounding medium.  When the
556: Lorentz factor $\Gamma$ of the blastwave becomes comparable to
557: $\thetajet^{\,-1}$, where \thetajet\ is the opening angle of the jet, the
558: jet starts to spread sideways. At that time, \tjet, the temporal
559: behavior of the broadband spectrum changes \citep[see
560: e.g.][]{rhoads1997:apj487:487}.
561: 
562: We fit our data to six parameters: \nuc, \num, \nua, \Fp, $p$ and
563: \tjet.  From these parameters and the redshift of the burst, $z =
564: 0.8$, we can find the physical parameters governing the blastwave and
565: its surroundings: the blastwave isotropic equivalent energy \Eiso, the
566: jet opening angle \thetajet, the collimation corrected blastwave
567: energy \Ejet, the fractional energy densities behind the relativistic
568: shock in electrons and in the magnetic field, \epse\ and \epsB\
569: respectively, and the density of the surrounding medium. The meaning
570: of the latter parameter depends on the density profile of the surrounding
571: medium. For a homogeneous circumburst medium, we simply determine the density
572: $n$. For a massive stellar wind, where the density is proportional to
573: $R^{\,-2}$ with $R$ the distance to the GRB explosion center, we
574: obtain the parameter $A_{\ast}$, which is the ratio of the mass-loss
575: rate over the terminal wind velocity of the GRB progenitor.
576: 
577: Our modeling results are shown in Table \ref{table:fitresults}, for
578: both the homogeneous external medium and the stellar wind environment.
579: The light curves for the best fit parameters are shown in Figure
580: \ref{figure:resultsall}.  We have performed Monte Carlo simulations
581: with synthetic data sets in order to derive accuracy estimates of the
582: best fit parameters, which are also given in the table.  It is evident
583: from the results that our six fit parameters are reasonably well
584: constrained in both cases for the circumburst medium.  The derived
585: physical parameters are also well constrained, except for \epse\ and
586: \epsB. The values we find for both the isotropic and the collimation
587: corrected energy, are similar to those found for other bursts; this is
588: also true for $p$. See e.g. \citet{panaitescu2001:apj560:49} and
589: \citet{yost2003:apj597:459}.  The jet opening angle and the density of
590: the surrounding medium are quite small, but both not unprecedented.
591: The jet-break time \tjet\ is somewhat smaller than estimated in
592: Section \ref{section:obs:X-ray}, but both estimates have relatively
593: large errors, likely because of the lack of (X-ray) data around the
594: jet-break time.
595: 
596: With the absence of optical light curves, it is not possible to
597: discriminate between the two different circumburst media.  This is
598: mainly due to the fact that the X-ray band lies above both \num\ and
599: \nuc, in which case the slopes of the light curves do not depend on
600: the density profile of the circumburst medium (even at 0.15 days,
601: back-extrapolating \nuc\ from Table \ref{table:fitresults} results in
602: its value being below the X-ray band). The \chisqred\ is somewhat
603: better for the stellar wind case, but the homogeneous case cannot be
604: excluded.  From the X-ray light curve, however, one can conclude that
605: the density profile of the medium does not change between
606: approximately 0.15 and 12 days after the burst.  If there were a
607: transition from a stellar wind to a homogeneous medium, the X-ray flux
608: has to rise or drop significantly, unless the densities are the
609: fine-tuned at the transition point \citep{peer2006:apj643:1036}.  From
610: the fact that the medium does not change during the X-ray
611: observations, one can draw conclusions on the distance of the wind
612: termination shock of the massive star: if one assumes that the medium
613: is already homogeneous at $\approx\,0.15$ days, the wind termination
614: shock position is at $R_{\mathrm{w}}\lesssim 9.8\cdot
615: 10^{17}\,\mathrm{cm}$ (0.32 pc); if the circumburst medium is a
616: stellar wind up to $\approx 12$ days after the burst, $R_{\mathrm{w}}
617: \gtrsim 1.1\cdot 10^{19}\,\mathrm{cm}$ (3.7 pc).
618: 
619: 
620: \begin{deluxetable}{lll}
621:   \tablecaption{Results of broadband modeling for both a homogeneous
622:     external medium and a massive stellar wind.  The best fit
623:     parameters are shown together with accuracy estimates from Monte
624:     Carlo simulations with synthetic data sets.  The characteristic
625:     frequencies of the synchrotron spectrum and the specific peak flux are
626:     given at $t_{\mathrm{j}}$.  
627: \label{table:fitresults}
628: }
629: \tablehead{
630: \colhead{Parameter} &
631: \colhead{Homogeneous} &
632: \colhead{Stellar wind}
633: }
634: \startdata
635: \nuc(\tjet) & 
636: $(1.45\plusminus{1.12}{0.23}) \cdot 10^{17}$ Hz                  
637: & $(2.84\plusminus{0.32}{1.30}) \cdot 10^{17}$ Hz \\
638: 
639: \num(\tjet)     
640: & $(3.50\plusminus{2.26}{1.47}) \cdot 10^{11}$ Hz                  
641: & $(2.90\plusminus{2.03}{1.15}) \cdot 10^{11}$ Hz \\
642: 
643: \nua(\tjet)     
644: & $(4.56\plusminus{2.85}{3.08}) \cdot 10^{9}$ Hz                   
645: & $(2.68\plusminus{2.17}{1.60}) \cdot 10^{9}$ Hz \\
646: 
647: \Fp(\tjet)      
648: & $888\plusminus{52}{109}$ $\mu$Jy                                 
649: & $694\plusminus{30}{240}$ $\mu$Jy \\ 
650: 
651: $p$             
652: & $2.06\plusminus{0.19}{0.05}$                                     
653: & $2.10\plusminus{0.08}{0.09}$ \\
654: 
655: \tjet           
656: & $0.96\plusminus{0.40}{0.28}$ days                                
657: &$1.06\plusminus{0.41}{0.11}$ days \\
658: 
659: \thetajet       
660: & $3.39\plusminus{2.02}{2.27}$ deg                                
661: & $2.30\plusminus{1.09}{0.85}$ deg \\
662: 
663: \Eiso           
664: & $(5.23\plusminus{1.13}{1.69}) \cdot 10^{52}$ erg                 
665: & $(28.2\plusminus{31.0}{10.4}) \cdot 10^{52}$ erg \\
666: 
667: \Ejet         
668: & $(0.917\plusminus{0.655}{0.512}) \cdot 10^{50}$ erg                 
669: & $(2.27\plusminus{2.25}{0.79}) \cdot 10^{50}$ erg \\
670: 
671: \epse           
672: & $0.247\plusminus{1.396}{0.212}$                                  
673: & $0.0681\plusminus{0.3951}{0.0348}$ \\
674: 
675: \epsB           
676: & $(7.63\plusminus{42.57}{6.30}) \cdot 10^{-3}$                    
677: & $(8.02\plusminus{28.18}{7.17}) \cdot 10^{-3}$ \\
678: 
679: $n$             
680: & $(1.06\plusminus{9.47}{1.04}) \cdot 10^{-2}$ $\mathrm{cm}^{-3}$  
681: &  \nodata  \\
682: 
683: $A_{\ast}$$^a$  
684: & \nodata                                                    
685: & $(2.94\plusminus{6.98}{2.11}) \cdot 10^{-2}$  \\
686: 
687: \chisqred       
688: & $1.9$ 
689: & $1.5$ \\
690: 
691: \enddata
692: \tablenotetext{a}{The parameter $A_{\ast}$ is a measure
693:     for the density in the case of a stellar wind environment, being
694:     the ratio of the mass-loss rate over the terminal wind velocity,
695:     and here given in units of $10^{-5}$ Solar masses per year divided
696:     by a wind velocity of 1000 km/s \citep[see][]{vanderhorst2007:submitted}. }
697: \end{deluxetable}
698: 
699: \placefigure{figure:resultsall}
700: \begin{figure*}
701: \includegraphics[width=\textwidth]{fig1}
702: \caption{Fit results for a homogeneous circumburst medium (left panel)
703:   and a massive stellar wind (right panel).  The solid and dash-dotted
704:   lines are the best model fits, and the dotted and dashed lines
705:   indicate the predicted rms scatter due to interstellar
706:   scintillation; see the appendix for further details. Also included in the figure (and modeling) is the
707:   reported VLA 8.5~GHz detection
708:   \citep[][left-most point in the 8.5~GHz subplot]{cameron2005:gcn4154}. \label{figure:resultsall}}
709: \end{figure*}
710: 
711: 
712: \subsection{The non-detection of the optical afterglow} \label{section:OA}
713: 
714: It is quickly seen that \thisgrb\ falls into the category of the
715: so-called ``dark bursts''. Using, for example, the quick criterion
716: proposed by \citet{jakobsson2003:apj617:21}, we find $\beta_{OX} <
717: -0.05$ at 12.7 hours after the burst using the Gemini $r^{\prime}$
718: band observation, well below the proposed limit of $\beta_{OX} < 0.5$.
719: A more precise criterion would combine the available spectral and
720: temporal parameters of the X-ray afterglow, allow all valid
721: combinations, and from that infer the range of possible optical
722: magnitudes from the X-rays \citep[see e.g.][]{rol2005:apj624:868}.
723: This is, in fact implied in our previous modeling: the modeled
724: specific fluxes corresponding to the band and epoch of our optical and
725: nIR upper limits are listed in Table \ref{table:compupperlimit} (see
726: also Table \ref{table:limmags}).
727: 
728: While the values in this table are given for local extinction, not
729: K-corrected to $z = 0.8$, it is
730: immediately obvious that our $K$-band observations put a 
731: stringent constraint on the required extinction directly surrounding
732: the burst.
733: 
734: \begin{deluxetable}{cccccccc}
735:   \tablecaption{Upper limits compared to model specific flux calculations.  The
736:     inferred lower limits on the extinction are given in the observers
737:     frame. The \ebv\ values are given for a Galactic extinction curve
738:     ($R_V = 3.08$), and are for illustrative purposes; see the
739:     comments at the end of Section \ref{section:OA}.
740: \label{table:compupperlimit}}
741: \tablehead{
742: \colhead{filter} &
743: \colhead{upper limit} &
744: \multicolumn{3}{c}{homogeneous density profile} &
745: \multicolumn{3}{c}{stellar wind density profile} \\
746: &
747: & 
748: \colhead{modeled specific flux} &
749: \colhead{extinction} &
750: \colhead{\ebv} &
751: \colhead{modeled specific flux} &
752: \colhead{extinction} &
753: \colhead{\ebv} \\
754: &
755: \colhead{($\mu$Jy)} &
756: \colhead{($\mu$Jy)} &
757: \colhead{(mag.)}&
758: \colhead{} &
759: \colhead{($\mu$Jy)} &
760: \colhead{(mag.)}&
761: \colhead{}
762: }
763: 
764: \startdata
765: $K_s$    &   $< 6.82 $  &  $93.1$  &  2.84  &  7.74   &  $57.2$  &  2.31  &  6.29 \\
766: $J$      &   $<12.3  $  &  $117 $  &  2.44  &  2.71   &  $74.1$  &  1.95  &  2.16 \\
767: $Z$      &   $< 2.66 $  &  $103 $  &  3.97  &  2.58   &  $67.8$  &  3.52  &  2.29 \\
768: $r^{\prime}$     &   $< 0.305$  &  $74.5$  &  5.97  &  2.17   &  $44.4$  &  5.41  &  1.97 \\
769: $z^{\prime}$     &   $< 1.53 $  &  $87.7$  &  4.40  &  2.97   &  $51.9$  &  3.83  &  2.59 \\
770: 
771: \enddata
772: \end{deluxetable}
773: 
774: To estimate the amount of local extinction in the host galaxy, we have
775: modeled the nIR to X-ray spectrum around 0.5~days after the burst,
776: considering 3 different extinction curves: those of the Milky Way
777: (MW), the Large Magellanic Cloud (LMC) and the Small Magellanic Cloud
778: (SMC), from \citet{pei1992:apj395:130}, with $R_V$ of 3.08, 3.16 and
779: 2.93, respectively.
780: 
781: 
782: For this, we used the unabsorbed XRT flux obtained from the spectral
783: fit to orbits 3~--~7 (which do not contain piled-up data), and fixed
784: the energy spectral slope in the X-rays at $\beta = 1$ (also from the
785: X-ray spectral fit). The optical specific fluxes were scaled to the
786: logarithmic mid-observation time of the X-ray observations with an
787: assumed $\alpha = 1.16$ decline. This estimated optical decay is
788: derived from the pre-break X-ray decay value, allowing for the cooling
789: break between the two wavelength regions, and averaging the two
790: possible values for $\alpha_X - \alpha_{\mathrm{opt}}$ (-0.25 and
791: 0.25).  We can further put the most stringent constraint on the broken
792: power law spectral shape, by setting the spectral break just below the
793: X-rays, at $1.8 \times 10^{17}$~Hz, which follows from our previous
794: broad-band modeling. Our results indicate that, for the aforementioned
795: extinction curves, a local extinction of $\ebv \approx 7$ (for all
796: three extinction curves) is necessary to explain the $K$-band upper
797: limit.
798: 
799: We can relate the resulting \nh\ from our X-ray spectral fits to any
800: local \ebv, using the relations found in \citet{predehl1995:aa293},
801: \citet{fitzpatrick1985:apj299} and \citet{martin1989:aa215:219} for
802: $N(HI)/\ebv$, and adjusting the metallicity in our X-ray absorption
803: model accordingly. We obtain $\ebv = 7.5, 1.54$ and 0.84 for a MW, LMC
804: and SMC extinction curve respectively, with the MW value showing the
805: best agreement with our findings for optical extinction (both
806: \citealt{nakagawa2006:pasjl58:35} and \citealt{butler2005:gcn4170}
807: find \ebv\ values roughly twice as high here, for a MW extinction
808: curve only, since their \nh\ estimate is larger than ours). This,
809: obviously, depends on the assumption that the MW (or otherwise, LMC or
810: SMC) extinction curves are valid models to compare with our observed
811: data here. Since these data happen to originate from just one sight
812: line in a galaxy, this may very well not be the case.  Further, even
813: if the extinction curve is correct, the actual value of $R_V$ may be
814: rather different for the host galaxy. Finally, the \ebv~--~\nh\
815: relations show a rather large scatter, especially at higher column
816: densities, nor is the \nh\ always derived using X-ray spectroscopy.
817: Our above results are therefore approximations, which are useful to
818: compare with other (GRB host) studies, but should be taken with the
819: necessary caution.
820: 
821: 
822: \subsection{The host galaxy of GRB\,051022}
823: \label{section:host}
824: 
825: 
826: \begin{deluxetable}{ccc}
827:   \tablecaption{Measured host galaxy magnitudes\label{table:hostgalaxy}}
828: \tablehead{
829: \colhead{filter} &
830: \colhead{magnitude} &
831: \colhead{magnitude error}
832: }
833: \startdata
834: $K$                   &  18.40  &  0.04  \\ 
835: $K_s$                 &  18.36  &  0.09  \\
836: $H$                   &  19.42  &  0.09  \\
837: $J$                   &  19.92  &  0.05  \\
838: $Z$\tablenotemark{a}  &  21.41  &  0.05  \\
839: $z^{\prime}$           &  21.30  &  0.04  \\
840: $i^{\prime}$           &  21.77  &  0.01  \\
841: $r^{\prime}$           &  22.04  &  0.01  \\
842: $R$                   &  21.84  &  0.09  \\
843: $V$                   &  22.30  &  0.04  \\
844: $B$                   &  22.75  &  0.02  \\
845: $U$                   &  $>$ 21.3\tablenotemark{b} & \nodata
846: \enddata
847: \tablenotetext{a}{AB magnitude}
848: \tablenotetext{b}{5-$\sigma$ upper limit}
849: \end{deluxetable}
850: 
851: 
852: Using the optical data described above, we fit the SED of the host of
853: GRB\,051022 using the HyperZ program\footnote{See
854:   \url{http://webast.ast.obs-mip.fr/hyperz}} developed by
855: \citet{bolzonella2000:aap363:476}. The photometry of the host has been
856: performed using \swcommand{apphot} within IRAF, in an aperture 1.5
857: times the estimated seeing in the different exposures. The results are
858: reported in Table \ref{table:hostgalaxy} (see also
859: \citealt{ovaldsen2007:apj662:294}). The range of photometric
860: magnitudes reported in this paper provides one of the most complete
861: broadband optical datasets of a GRB host galaxy to date.  We fit using
862: the eight synthetic galaxy templates provided within HyperZ at the
863: redshift of the host, and find that the host galaxy is a blue compact
864: galaxy of type irregular, with a dominant stellar population age of
865: $\approx 20$~Myr, similar to other long GRB hosts
866: \citep{christensen2005:apjl631:29}.  A moderate amount of extinction
867: of $A_V \approx 1$~mag is required to fit the SED, with an SMC-type
868: extinction curve providing a best fit, and the luminosity of the host
869: is approximately 1.5 $L_*$ (assuming $M_{*,B} = -21$); these findings
870: are in full agreement with \citet{castro-tirado2006:aipc836:79}.  The
871: amount of extinction in the line of sight towards the GRB required to
872: suppress the optical light of the afterglow to the observed limits is
873: clearly higher than the $A_V$ value found from the host SED: $A_V =
874: 4.4$ magnitudes towards the GRB, estimated from blueshifting our
875: measured (observer frame) $z^{\prime}$ band extinction to $z = 0.8$.
876: The host galaxy SED extinction is, however, an average value derived
877: from the integrated colors of the host.
878: 
879: The host of GRB\,051022 is located in a field crowded with galaxies of
880: various Hubble types.  We perform photometry on several galaxies close
881: to the GRB host (within 1 arcminute) to investigate the possibility
882: that the high star formation rate seen in the optical
883: (\citealt{castro-tirado2006:aipc836:79} report an SFR of
884: $\approx\,20$M$_\odot$yr$^{-1}$) is induced by a recent interaction
885: with one of the neighboring galaxies. As formation of high mass stars
886: has also been observed to occur in dusty regions in merging systems
887: \citep[see e.g.][]{lin2007:apjl660:51}, this could help to explain the
888: excess optical extinction towards GRB\,051022. We performed HyperZ
889: fits to these galaxies, and find that none of them is well fit by a
890: photometric redshift of $z \approx 0.8$. Particularly the two galaxies
891: closest to the GRB host galaxy are not compatible with a redshift 0.8,
892: and show best fits with photometric redshifts of $z \approx
893: 0.2$~--~0.25.  Out of the sample of six galaxies close to the GRB host
894: we find that four have best-fit photometric redshifts in the range
895: 0.20~--~0.25, making it unlikely that a possible overdensity of
896: galaxies near the host galaxy is due to a cluster or galaxy group at
897: the host redshift.
898: 
899: 
900: \section{Discussion} \label{section:discussion}
901: 
902: The issue of non-detected (``dark'') GRB afterglows has received
903: significant interest ever since the discovery of the first GRB
904: afterglow, starting with the non-detection of GRB\,970828 to very deep
905: limits \citep{groot1998:apj493:27,odewahn1997:iauc6735}. For this
906: particular afterglow, its non-detection has been attributed to a
907: dust-lane in its host galaxy \citep{djorgovski2001:apj562}. Dust
908: extinction as the cause of the non-detection of the optical afterglow
909: has been inferred in the case of several other GRBs, notably those
910: with a precise X-ray or radio position, where one can pinpoint the
911: afterglow position on top of its host galaxy (e.g. GRB\,000210,
912: \citealt{piro2002:apj577}). 
913: 
914: 
915: Optical drop-outs due to high redshift will also result in dark
916: bursts, but are harder to confirm, since it would require at least one
917: detection in a red band, to detect the Ly$\alpha$ break. Otherwise, it
918: becomes indistinguishable from dust extinction.
919: 
920: Other explanations of afterglow non-detections include the intrinsic
921: faintness of the afterglow. For HETE-2 detected GRBs, this has been
922: inferred for e.g. GRB\,020819 \citep{jakobsson2005:apj629:45}. For
923: Swift bursts, where rapid and accurate X-ray positions are often
924: available, this is a major cause of non-detections
925: \citep{berger2005:apj634:501}, largely attributed to a higher average
926: redshift.
927: 
928: In our case here, the host galaxy has been detected at a relatively
929: modest redshift, which almost automatically points to the dust
930: extinction scenario. The radio and X-ray detections even allow us to
931: accurately model the necessary amount of extinction between us and the
932: GRB.
933: 
934: 
935: \subsection{The burst environment} \label{section:environment}
936: 
937: The issue of the role of dust extinction in the lines of sight towards
938: GRBs is still very much an open one. While clear signs of dust
939: depletion are seen in several afterglow spectra, the $A_V$ values that
940: are predicted from these depletion measures are generally much higher
941: than the observed ones, that can be found from the continuum shape
942: \citep{savaglio2004:apj614:293}. Recently, selected samples of GRB
943: afterglows were homogeneously analyzed for X-ray and optical
944: extinction, showing dust to gas ratios different from Galactic and
945: Magellanic cloud values
946: \citep{starling2007:apj661:787,schady2007:mnras377:273}. 
947: \citet{galama2001:apj549} and \citet{stratta2004:apj608:846} had
948: already found dust (optical) to gas (X-ray) ratios to be lower than
949: the Galactic value (in all cases, however, there is a bias in
950: these samples to optically and X-ray detected afterglows).  Comparison
951: of neutral hydrogen columns and metallicities of afterglow lines of
952: sight with X-ray extinction values \citep{watson2007:apjl600:101}
953: showed that the absorption probed by these two wavelength regimes is
954: generally located at different positions in the host. In all these
955: cases there may be significant biases against bursts with low apparent
956: magnitudes, preventing optical spectroscopy, which are hard to
957: quantify.
958: 
959: 
960: 
961: In the case of \thisgrb\, there is a significant discrepancy between
962: the extinction for the host as a whole and that along the line of
963: sight to the burst, or at least along our line of sight towards the
964: burst. This is perhaps not too surprising if one assumes, for example,
965: that the burst occurred inside a Giant Molecular Cloud (GMC).
966: \citet{jakobsson2006:aal460:13} compared the GRB $N$(HI) distribution
967: to that of modeled GRBs located inside Galactic-like GMCs. They found
968: that the two distributions are incompatible, and possibly GRBs are
969: more likely to occur inside clouds with a lower $N$(HI), or
970: alternatively, outside the actual GMC. (Note that their study
971: concentrates on bursts with $z > 2$, where the Ly-$\alpha$ absorption
972: is visible in the optical wavebands; it is also biased towards
973: optically detected afterglows). A GMC could therefore actually be
974: positioned in front of the GRB, where the required optical and X-ray
975: extinction is easily achieved. This agrees with the findings by
976: \citet{prochaska2007:astro-ph/0703365}, who analyzed several
977: GRB-Damped\,Lyman\,Alpha spectra and from observed depletion levels
978: infer that the gas is not located directly near the GRB (e.g. its
979: molecular cloud) but further out. The specific case of GRB\,060418
980: confirmed this through time-resolved high resolution spectroscopy,
981: showing that the observed metal lines originate past 1.7 kpc from the
982: burst itself \citep{vreeswijk2007:aap468:83}. In fact, X-ray radiation
983: from the burst could easily destroy grains out to 100 pc
984: \citep{waxman2000:apj537:796,fruchter2001:apj563:597,drain2002:apj569:780}
985: and permit the afterglow radiation to penetrate the surrounding
986: molecular cloud. Dust extinction is therefore likely to occur further
987: out, perhaps to several kiloparsecs.
988: 
989: It is interesting to find a non-SMC type of extinction curve from the
990: combination of X-ray and optical absorption (though not completely
991: ruled out): in most cases modeled, an SMC extinction curve fits the
992: optical--X-ray spectra best
993: \citep{starling2007:apj661:787,schady2007:mnras377:273}, presumably
994: attributable to the absence of the 2175 \AA\ feature
995: \citep{savage1979:araa17:73} and the low dust to gas ratio. Our
996: findings indicate that the extinction along the line of sight to the
997: GRB will generally be different than one of the three assumed
998: extinction curves.  Local small scale density variations in clouds,
999: such as found by from infrared studies in the Taurus region and from
1000: simulations \citep{padoan2006:apj649:807}, could cause this fairly
1001: easily.
1002: 
1003: 
1004: \subsection{Energetics}
1005: 
1006: Our modeling provides us with a detailed set of parameters of the
1007: afterglow energetics, including \Ejet, the energy of the afterglow.
1008: For the prompt emission energy, we use the data from the Konus-Wind
1009: measurements \citep{golenetskii2005:gcn4150}. We calculate a prompt
1010: isotropic energy of $4.39\plusminus{0.29}{0.18} \times 10^{53}$ erg in
1011: the 20 keV~--~20 MeV observer frame, and, by applying a K-correction
1012: \citep[as in e.g.][]{bloom2001:aj121}, $\Epromptiso =
1013: 10.4\plusminus{0.7}{0.4} \times 10^{53}$ erg in the 1~--~$10^5$ keV
1014: rest frame. The collimation corrected energy depends on the assumed
1015: density profile of the surrounding medium: for a homogeneous medium, we
1016: obtain $\Epromptjet = 18.2 \times 10^{50}$ erg, and for a wind-like
1017: medium, $\Epromptjet = 8.38 \times 10^{50}$ erg.  With $\Epeak =
1018: 918\plusminus{66}{59}$ keV in the burst rest frame, we find that the
1019: \Epeak~--~\Epromptjet\ relation \citep{ghirlanda2004:apj616:331}
1020: somewhat underestimates the \Epeak\ when calculated from \Epromptjet:
1021: $\Epeak \approx 740$ keV for a homogeneous medium, and $\approx 430$
1022: keV for a wind medium (the difference between our chosen cosmology and
1023: that used by \citealt{ghirlanda2004:apj616:331} amounts to only a
1024: 0.3\% difference in \Eiso).  These estimates, however, come with a few
1025: caveats: \emph{1)} the \Epeak\ from the Konus-Wind data is calculated
1026: using an exponential cut-off model, not the Band function
1027: \citep{band1993:apj413:281}. Since the Band function includes the case
1028: of an exponential cut-off model (with $\beta = -\infty$, this should,
1029: however, pose no problem in estimating the actual \Epeak), \emph{2)}
1030: our break time, and therefore the jet-opening angle, are calculated
1031: from the full modeling of the afterglow, which effectively means
1032: derived from the available X-ray and radio data. Further, the original
1033: Ghirlanda relation was derived using optical break times. Recent
1034: efforts show that estimating jet-break times from X-ray light curves
1035: may not lead to the same results
1036: \citep[e.g.][]{panaitescu2006:mnras369:2059}, and \emph{3)} the
1037: relatively large error on the jet opening angle estimate allows for a
1038: relatively large range in collimation corrected energies. We have
1039: simply used here our best value, but an \Epeak\ value of 1498 keV
1040: derived from \Ejet\ can still be accommodated within our errors. (We
1041: note that, with a different \Epeak\ estimate and an incorrect value
1042: for the jet-break time, \citealt{nakagawa2006:pasjl58:35} still found
1043: their results to lie on the Ghirlanda relation).  The break time
1044: problem can be avoided by looking only at the \Epeak\ -- \Epromptiso\
1045: relation \citep{amati2002:aa390:81,amati2006:mnras372:233}. From this,
1046: we estimate $\Epeak \approx 924$ keV, nicely in agreement with the
1047: value found directly from the spectra fit.
1048: 
1049: Comparing the prompt emission energy (\Epromptjet) and afterglow blast
1050: wave kinetic energy (\Ejet), we find their ratio to be $\Epromptjet /
1051: \Ejet = 3.7$ in the case of a wind-like circumburst medium, while for
1052: a homogeneous medium, $\Epromptjet / \Ejet = 20$.  These ratios are
1053: similar to those found for other bursts \citep[e.g.][Figure
1054: 3]{berger2003:nature426:154}.
1055: 
1056: \thisgrb\ is also one of the brightest bursts observed by HETE, with a
1057: prompt 30--400 keV fluence of $S = 1.31 \times 10^{-4}$ erg cm$^{-2}$
1058: \citep{nakagawa2006:pasjl58:35}. In fact, compared to the sample of 35
1059: FREGATE bursts analyzed by \citet{barraud2003:aa400}, \thisgrb\ has
1060: the largest fluence, even topping the relatively close-by GRB\,030329
1061: (\citealt[$S = 1.2 \times 10^{-4}$ erg
1062: cm$^{-2}$]{vanderspek2004:apj617:1251}; note that for \thisgrb, its
1063: redshift is close to the median redshift of HETE-2 detected GRBs and
1064: therefore distance effects will play a very minor role).
1065: \citet{rol2005:apj624:868} noted this potential correlation of fluence
1066: with the non-detection of a GRB afterglow for the small subset of
1067: genuinely dark bursts in their sample: the truly dark bursts all have
1068: a much higher than average fluence (although this is for a relatively
1069: small sample only).  Potentially, this could point to an external
1070: origin for the prompt emission, instead of being due to internal
1071: shocks: a large amount of dust may result in more matter that will
1072: radiate, while at the same time the radiation will be suppressed at UV
1073: and optical wavelengths.  This would indicate an origin of the
1074: extinction quite close to the burst instead, in contrast to previous
1075: findings for other bursts, as discussed in Section
1076: \ref{section:environment}. These latter bursts, however, were all
1077: optically selected to obtain spectroscopy, and may therefore show
1078: different surroundings than \thisgrb.  Unfortunately, with the small
1079: sample size of genuine dark bursts a firm conclusion on this
1080: correlation is not possible, but remains something to watch for in
1081: future dark bursts.
1082: 
1083: 
1084: 
1085: \section{Conclusions} \label{section:conclusions}
1086: 
1087: 
1088: \thisgrb\ is a prototypical dark burst, with the local extinction
1089: exceeding $2.3$ magnitudes in $J$ and 5.4 magnitudes in $U$, in the
1090: host-galaxy restframe, with the exact limits depending on the
1091: circumburst density profile.  The extinction curve derived from an
1092: X-ray~--~optical spectral fit points towards a Galactic type of
1093: extinction curve, although it is likely that this is more or less a
1094: coincidence: the host galaxy itself is best modeled with an SMC-like
1095: extinction curve, with a modest amount of extinction, $A_V \approx
1096: 1$~mag.  The large optical absorption towards the afterglow of
1097: GRB\,051022 is therefore probably the effect of an unfortunate
1098: position in the host where the line of sight crosses dense regions
1099: within the host.
1100: 
1101: The X-ray and radio afterglow data allow for a full solution of the
1102: blastwave model, although we unfortunately cannot distinguish between
1103: the density profile (homogeneous or wind-like) of the circumburst
1104: medium.  We estimate a collimation-corrected energy in the afterglow
1105: emission of 0.92~--~2.3 $\times 10^{50}$ erg, while the energy in
1106: prompt emission (1~--~$10^5$~keV rest frame) is 8.4~--~18 $\times
1107: 10^{50}$~erg. Aside from the large optical extinction, the afterglow
1108: otherwise appears as an average afterglow, with no outstanding
1109: properties. The potentially interesting point here is that the
1110: 30-400~keV fluence of the prompt emission is one of the largest ever
1111: detected in the HETE-2 sample.
1112: 
1113: In the era of Swift GRBs, dust-extincted bursts can actually be
1114: found in optical/nIR thanks to the rapid availability of precise
1115: positions: examples are found where the burst is relatively bright
1116: early on at optical/nIR wavelengths, while the afterglow proper (post
1117: few hours) often can go undetected
1118: \citep[e.g.][]{oates2006:mnras372:3270,perley2007:astro-ph/0703538}.
1119: This allows targeted follow-up of such dark bursts, i.e. determining
1120: the host galaxy (and the bursts precise position therein) and a
1121: redshift measurement. In our case, a precise CXO and radio position
1122: pinpointed the host galaxy, but such data may not always be available.
1123: High resolution late-time observations of the host, at the location of
1124: the GRB, may then reveal whether the burst indeed occurred inside a
1125: dense host region.
1126: 
1127: 
1128: 
1129: \acknowledgments
1130: 
1131: {\small We thank the referee for a careful reading of the manuscript
1132:   and constructive comments.  We thank Kim Page and Andy Beardmore for
1133:   useful discussions regarding the XRT data analysis. ER and RLCS
1134:   acknowledge support from PPARC. KW and RAMJW acknowledge support of
1135:   NWO under grant 639.043.302. The authors acknowledge funding for the
1136:   Swift mission in the UK by STFC, in the USA by NASA and in Italy by
1137:   ASI. The Dark Cosmology Centre is funded by the Danish National
1138:   Research Foundation. The William Herschel Telescope is operated on
1139:   the island of La Palma by the Isaac Newton Group in the Spanish
1140:   Observatorio del Roque de los Muchachos of the Instituto de
1141:   Astrof\'{\i}sica de Canarias.  The United Kingdom Infrared Telescope
1142:   is operated by the Joint Astronomy Centre on behalf of the U.K.
1143:   Particle Physics and Astronomy Research Council. The data reported
1144:   here were obtained as part of the UKIRT Service Programme. The
1145:   Westerbork Synthesis Radio Telescope is operated by ASTRON
1146:   (Netherlands Foundation for Research in Astronomy) with support from
1147:   the Netherlands Foundation for Scientific Research (NWO). Support
1148:   for this work was provided by the National Aeronautics and Space
1149:   Administration through Chandra Award Number 1736937 issued by the
1150:   Chandra X-ray Observatory Center, which is operated by the
1151:   Smithsonian Astrophysical Observatory for and on behalf of the
1152:   National Aeronautics Space Administration under contract NAS8-03060.
1153:   This publication makes use of data products from the Two Micron All
1154:   Sky Survey, which is a joint project of the University of
1155:   Massachusetts and the Infrared Processing and Analysis
1156:   Center/California Institute of Technology, funded by the National
1157:   Aeronautics and Space Administration and the National Science
1158:   Foundation. This research has made use of data obtained from the
1159:   High Energy Astrophysics Science Archive Research Center (HEASARC),
1160:   provided by NASA's Goddard Space Flight Center. }
1161: 
1162: 
1163: 
1164: \appendix
1165: 
1166: \section{Interstellar scintillation in the radio modeling}
1167: 
1168: 
1169: The 4.9 GHz measurements show scatter around the best fit light curve,
1170: which can be accounted for by interstellar scintillation (ISS).  In
1171: Figure \ref{figure:resultsall} we have indicated the predicted rms
1172: scatter due to ISS.  We have calculated the scattering measure from
1173: the \citet{cordes2002:astro-ph/0207156} model for the Galactic
1174: distribution of free electrons: $SM=2.04\cdot
1175: 10^{-4}\,\mathrm{kpc}\,/\mathrm{m}^{-20/3}$.  The radio specific flux will be
1176: modulated when the source size is close to one of the three
1177: characteristic angular scales, i.e. for weak, refractive or
1178: diffractive ISS.  From \citet{walker1998:mnras294:307}, we calculate
1179: the transition frequency between weak and strong ISS,
1180: $\nu_0=9.12\,\mathrm{GHz}$, and the angular size of the first Fresnel
1181: zone, $\theta_{{\mathrm{F}}_0}=0.994\,\mu\mathrm{as}$.  Our
1182: measurements were all performed at frequencies below $\nu_0$, i.e. in
1183: the strong ISS regime, which means that only refractive and
1184: diffractive ISS modulate the specific flux significantly.  We calculate the
1185: evolution of the source size in the extreme relativistic phase
1186: ($\theta_\mathrm{s}=R/\Gamma$) and after the jet-break
1187: ($\theta_\mathrm{s}=R \theta_{\mathrm{j}}$), and compare this source
1188: size with the diffractive angular scale
1189: $\theta_\mathrm{d}=\theta_{{\mathrm{F}}_0}(\nu_0/\nu)^{-6/5}=0.0701\cdot
1190: \nu_{\mathrm{GHz}}^{6/5}\,\mu\mathrm{as}$ and the refractive angular
1191: scale
1192: $\theta_\mathrm{r}=\theta_{{\mathrm{F}}_0}(\nu_0/\nu)^{11/5}=128\cdot
1193: \nu_\mathrm{GHz}^{-11/5}\,\mu\mathrm{as}$ to calculate the modulation
1194: index $m_\mathrm{p}$.  In the case of diffractive ISS the modulation
1195: index is $1$, and in the case of refractive ISS
1196: $m_\mathrm{p}=(\nu_0/\nu)^{-17/30}=0.286\cdot
1197: \nu_{\mathrm{GHz}}^{17/30}$.  Because of the expansion of the
1198: blastwave the angular source size exceeds one of the characteristic
1199: angular scales at some point in time.  Then the modulation will begin
1200: to quench as $m_\mathrm{p}(\theta_\mathrm{d}/\theta_\mathrm{s})$ in
1201: the case of diffractive ISS, and as
1202: $m_\mathrm{p}(\theta_\mathrm{r}/\theta_\mathrm{s})^{7/6}$ in the case
1203: of refractive ISS.
1204: 
1205: 
1206: \bibliographystyle{apj}
1207: \bibliography{references}
1208: 
1209: \end{document}
1210: