1: %\documentclass[12pt,preprint]{aastex}
2: %\documentstyle[12pt,aasms4,graphics]{article}
3: \documentclass{emulateapj}
4: %\documentstyle[emulateapj,graphics]{article}
5: \usepackage{psfig,natbib,amsmath}
6:
7: % User defs
8: \newcommand{\etal}{et~al.}
9: \newcommand{\citepeg}[1]{\citep[{e.g.,}][]{#1}}
10: \newcommand{\mnhi}{N_{\rm HI}}
11: %\def\baselinestretch{1.0}
12: \begin{document}
13:
14: \title{GRB 071003: Broadband Follow-up Observations of a Very Bright Gamma-Ray Burst in a Galactic Halo}
15:
16: \def\berk{1}
17: \def\lick{2}
18: \def\uva{3}
19: \def\jansky{4}
20: \def\sloan{5}
21: \def\mich{6}
22: \def\davis{7}
23: \def\caltech{8}
24: \def\chicago{9}
25: \def\nrao{10}
26: \def\keck{11}
27: \def\arizona{12}
28: \def\aerospace{13}
29: \def\boeing{14}
30: \def\rice{15}
31: \def\benedict{16}
32:
33: \author{D.~A.~Perley\altaffilmark{\berk},
34: W. Li\altaffilmark{\berk},
35: R.~Chornock\altaffilmark{\berk},
36: J.~X.~Prochaska\altaffilmark{\lick},
37: N.~R.~Butler\altaffilmark{\berk},
38: P.~Chandra\altaffilmark{\uva,\jansky},
39: L.~K.~Pollack\altaffilmark{\lick},
40: J.~S.~Bloom\altaffilmark{\berk,\sloan},
41: A.~V.~Filippenko\altaffilmark{\berk},
42: H.~Swan\altaffilmark{\mich},
43: F.~Yuan\altaffilmark{\mich}
44: C.~Akerlof\altaffilmark{\mich},
45: M.~W.~Auger\altaffilmark{\davis},
46: S.~B.~Cenko\altaffilmark{\caltech},
47: H.-W.~Chen\altaffilmark{\chicago},
48: C.~D.~Fassnacht\altaffilmark{\davis},
49: D.~Fox\altaffilmark{\caltech},
50: D.~Frail\altaffilmark{\nrao},
51: E.~M.~Johansson\altaffilmark{\keck},
52: D.~Le~Mignant\altaffilmark{\lick,\keck},
53: T.~McKay\altaffilmark{\mich},
54: M.~Modjaz\altaffilmark{\berk},
55: W.~Rujopakarn\altaffilmark{\arizona},
56: R.~Russell\altaffilmark{\aerospace},
57: M.~A.~Skinner\altaffilmark{\boeing},
58: G.~H.~Smith\altaffilmark{\lick},
59: I.~Smith\altaffilmark{\rice},
60: M.~A.~van~Dam\altaffilmark{\keck}, and
61: S.~Yost\altaffilmark{\benedict}
62: }
63: \altaffiltext{\berk}{Department of Astronomy,
64: University of California, Berkeley, CA 94720-3411.}
65: \altaffiltext{\lick}{Department of Astronomy and Astrophysics, UCO/Lick Observatory;
66: University of California, 1156 High Street, Santa Cruz, CA 95064.}
67: \altaffiltext{\uva}{Department of Astronomy, University of Virginia, P.O. Box 400325,
68: Charlottesville, VA 22904.}
69: \altaffiltext{\jansky}{Jansky Fellow, National Radio Astronomy Observatory.}
70: \altaffiltext{\sloan}{Sloan Research Fellow.}
71: \altaffiltext{\mich}{University of Michigan, Randall Laboratory of Physics,
72: 450 Church St., Ann Arbor, MI, 48109-1040.}
73: \altaffiltext{\davis}{Department of Physics, University of California, 1 Shields Avenue, Davis, CA 95616.}
74: \altaffiltext{\caltech}{Division of Physics, Mathematics, and Astronomy, MS 105-24,
75: California Institute of Technology, Pasadena, CA 91125.}
76: \altaffiltext{\chicago}{Department of Astronomy and Astrophysics, University of Chicago,
77: 5640 S. Ellis Ave, Chicago, IL 60637.}
78: \altaffiltext{\nrao}{National Radio Astronomy Observatory, P.O. Box O, Socorro, NM 87801.}
79: \altaffiltext{\keck}{W. M. Keck Observatory, 65-1120 Mamalahoa Highway, Kamuela, HI 96743.}
80: \altaffiltext{\arizona}{Steward Observatory, Tucson, AZ 85721.}
81: \altaffiltext{\aerospace}{The Aerospace Corporation, Mail Stop M2-266, PO Box 92957, Los Angeles, CA 90009-29957.}
82: \altaffiltext{\boeing}{The Boeing Company, 535 Lipoa Parkway, Suite 200, Kihei, HI 96753.}
83: \altaffiltext{\rice}{Department of Physics and Astronomy, Rice University, 6100 South Main, MS108, Houston, TX 77251-1892.}
84: \altaffiltext{\benedict}{College of St. Benedict, St. Joseph, MN 56374.}
85: \email{(dperley,wli)@astro.berkeley.edu}
86:
87: \slugcomment{To appear in ApJ 2008 November 10}
88:
89: \begin{abstract}
90: The optical afterglow of long-duration GRB 071003 is
91: among the brightest yet to be detected from any GRB, with $R \approx
92: 12$ mag in KAIT observations starting
93: 42~s after the GRB trigger, including filtered detections during
94: prompt emission. However, our high S/N ratio afterglow
95: spectrum displays only extremely weak absorption lines at what we
96: argue is the host redshift of $z=1.60435$ --- in contrast to the three
97: other, much stronger \ion{Mg}{2} absorption systems observed at lower
98: redshifts. Together with Keck adaptive optics observations which fail
99: to reveal a host galaxy coincident with the burst position, our
100: observations suggest a halo progenitor and offer a cautionary tale
101: about the use of \ion{Mg}{2} for GRB redshift determination. We
102: present early through late-time observations spanning the
103: electromagnetic spectrum, constrain the connection between the prompt
104: emission and early variations in the light curve (we observe no
105: correlation), and discuss possible origins for an unusual, marked
106: rebrightening that occurs a few hours after the burst: likely either a
107: late-time refreshed shock or a wide-angle secondary jet. Analysis of
108: the late-time afterglow is most consistent with a wind environment,
109: suggesting a massive star progenitor. Together with GRB~070125, this
110: may indicate that a small but significant portion of star formation in
111: the early universe occurred far outside what we consider a normal
112: galactic disk.
113: \end{abstract}
114:
115: \keywords{gamma rays: bursts --- gamma-ray bursts: individual: 071003}
116:
117: \section{Introduction}
118: % S 1
119:
120: Concurrent observations of long-wavelength afterglow and ongoing
121: gamma-ray burst (GRB) activity should, in principle, yield important
122: constraints on the nature of the physical processes of the emission
123: (eg., \citealt{Kobayashi2000}). However, as a GRB typically lasts less
124: than 100~s, it is challenging for large ground-based optical/infrared
125: follow-up facilities to react to a GRB alert quickly and take data
126: during the prompt phase. Multi-color observations, which provide vital
127: information on the emission mechanism, are even more difficult to
128: obtain during the prompt phase because of the added overhead
129: associated with changing filters. Nevertheless, due to the coordinated
130: efforts of recent space missions (\textit{HETE-II}, \citealt{Ricker+2003};
131: \textit{Swift}, \citealt{Gehrels+2004}) to detect GRBs and various
132: ground-based optical follow-up programs, observations during the prompt
133: phase of GRBs are no longer uncommon --- the optical afterglows (OAs)
134: of several dozen GRBs have been observed
135: \citepeg{Akerlof+1999,Vestrand+2006,Yost+2007} during gamma-ray
136: emission, and multi-color optical data have been obtained in a handful
137: of cases \citepeg{Blake+2005,Nysewander+2007}.
138:
139: Observations of GRBs in the past several years have also revealed
140: a rich demography in OA behavior. Some OAs have monotonic power-law
141: decays \citepeg{Li+2003a,Laursen+2003}, while others have plateau
142: \citepeg{Rykoff+2006} and rebrightening \citepeg{Wozniak+2006}
143: phases. Even among GRBs with relatively simple behavior, however,
144: short-timescale features not predicted in the basic shock models often
145: appear in sufficiently well-sampled data. Various modifications to
146: the standard picture have been proposed to explain such observations,
147: including the presence of a jet with single (e.g., \citealt{Sari+1999})
148: or multiple (e.g., \citealt{Berger+2003}) components, refreshed shocks
149: \citep{Zhang+2006}, central engine activity
150: \citep{Kocevski+2007,Chincarini+2007}, gravitational microlensing
151: \citep{Garnavich+2000}, and density irregularity in the GRB
152: environment \citep{Holland+2003}. Observationally, constraints on the
153: change in the afterglow color and the spectral energy distribution
154: (SED) play an important role in limiting the viability of models for a
155: particular GRB.
156:
157: The question of the nature of the GRB itself is intimately tied to the
158: question of its environment and origins. At intermediate to late
159: times, spectroscopy of the afterglow
160: \citepeg{Prochaska+2007,DElia+2007} and deep imaging of the host
161: environment \citep[e.g.,][]{Bloom+2002, Fruchter+2006} can help
162: establish the nature of the GRB's progenitor and environment,
163: connecting what we learn about the burst itself to the larger question
164: of its origins and place in the early universe.
165:
166: In this paper, we report on our photometric and spectroscopic
167: observations of GRB\,071003 with various telescopes from the prompt
168: phase to late times. In \S 2 we describe the observations, and in \S
169: 3 we present the reductions. The analysis of the light curves and
170: the constraints on the changes in the colors and SEDs are given in \S
171: 4. The conclusions, including the implications of the extremely
172: unusual spectrum of this event, are discussed in \S 5. We assume
173: $H_{\circ}=71$ km s$^{-1}$ Mpc$^{-1}$, $\Omega_{\rm M}=0.3$, and
174: $\Omega_{\Lambda}=0.7$ throughout the paper. These reports should be
175: considered the final analysis of our group's data on GRB\,071003,
176: superseding any previously announced results (e.g., in the GCN
177: Circulars).
178:
179: \section{Observations}
180: % S 2
181:
182: \subsection{BAT/XRT Observations}
183: % S 2.1
184:
185: On 2007 October 3, 07:40:55 UT (defined as $t=0$ in this paper; UT dates
186: are used throughout), a bright GRB triggered the Burst Alert Telescope (BAT)
187: onboard the \textit{Swift} satellite (trigger 292934
188: \citealt{GCN6837}). The first GCN notice was distributed within 16~s.
189: Unfortunately, \textit{Swift} was still returning to normal
190: observations after its 2007 August gyro failure, but it did slew to the
191: position after 22~ks and began observations using the X-Ray Telescope (XRT).
192:
193: We downloaded the \textit{Swift} BAT and XRT data from the {\it
194: Swift}~Archive\footnote{ftp://legacy.gsfc.nasa.gov/swift/data } and
195: quicklook data site.\footnote{\url
196: http://swift.gsfc.nasa.gov/cgi-bin/sdc/ql } The XRT and BAT spectra
197: were fitted using ISIS\footnote{\url http://space.mit.edu/CXC/ISIS }.
198:
199: The XRT data were processed with version 0.11.4 of the {\tt
200: xrtpipeline} reduction script from the HEAsoft~6.3.1\footnote{\url
201: http://heasarc.gsfc.nasa.gov/docs/software/lheasoft/ } software
202: release. We employ the latest (2007 December 4) XRT calibration
203: files. Our reduction of XRT data from cleaned event lists output by
204: {\tt xrtpipeline} to science-ready light curves and spectra is
205: described in detail by \cite{ButlerKocevski2007a}. We use the latest
206: calibration files from the 2007 September 24 BAT database release. We
207: establish the energy scale and mask weighting for the BAT event mode
208: data by running the {\tt bateconvert} and {\tt batmaskwtevt} tasks.
209: Spectra and light curves are extracted with the {\tt batbinevt} task,
210: and response matrices are produced by running {\tt batdrmgen}.
211: To produce the BAT spectra, we
212: apply the systematic error corrections to the low-energy BAT spectral
213: data as suggested by the BAT Digest Web site\footnote{\url
214: http://swift.gsfc.nasa.gov/docs/swift/analysis/bat\_digest.html}, and
215: fit the data in the 15--150 keV band. The spectral normalizations are
216: corrected for satellite slews using the {\tt batupdatephakw} task.
217:
218: \begin{figure}
219: \centerline{\includegraphics[width=3.5in,angle=0]{f1.eps}}
220: \caption[relation] {Light curve from the \textit{Swift} BAT of GRB
221: 071003, with optical photometry from KAIT and P60, and the optical
222: light curve model discussed in Section 4.3, overplotted. The GRB
223: is dominated by a complicated, spiky emission episode
224: in the first 30~s, but a pulse is also observed much later, at 150~s.
225: Optical data points (all from KAIT, except one $R$-band measurement from
226: the P60), by contrast, show a power-law decay at early times followed by a
227: slow-rising ``bump.'' Here the $V$ and $I$ filtered observations have
228: been offset to match the $R$ and unfiltered points based on the
229: relative colors at 2000~s.}
230: \label{fig:grblc}
231: \end{figure}
232:
233: The burst exhibits one dominant emission episode of duration $dt\approx
234: 30$~s, followed by a minor pulse $\sim 150$~s later of duration $\sim
235: 20$~s. The total duration is $T_{90}=148\pm1$~s,
236: \footnote{All uncertainties quoted in this paper are 1$\sigma$, except
237: where specified otherwise.}
238: placing it clearly into the long GRB class.
239: The primary pulse is resolved into multiple pulses.
240: The gamma-ray light curve is shown in Figure \ref{fig:grblc},
241: overplotted with early-time photometry from KAIT and P60
242: (discussed in \S 3.2 and \S 3.5, respectively).
243:
244: %Possibly move these next two paragraphs into analysis section
245:
246: The time-integrated BAT spectrum from $t=-10.3$ to $t=169$~s is
247: acceptably fitted ($\chi^2/\nu=47.64/55$, where $\nu$ is the number of
248: degrees of freedom) by a power-law model, with
249: photon index $\alpha = -1.3\pm 0.1$ and energy fluence
250: $S_{\gamma}=(1.7\pm 0.1) \times 10^{-5}$ erg cm$^{-2}$ (15--350 keV).
251: The main emission episode ($t=-1.4$~s to $t=22.8$~s) is harder
252: ($\alpha = -1.08\pm 0.03$, $S_{\gamma}=(1.51\pm0.03) \times 10^{-5}$
253: erg cm$^{-2}$, $\chi^2/\nu=56.71/55$), while the final pulse
254: ($t=131$--$169$~s) is softer ($\alpha = -1.8\pm 0.2$,
255: $S_{\gamma}=1.2^{+0.1}_{-0.2} \times 10^{-6}$ erg cm$^{-2}$,
256: $\chi^2/\nu=41.15/55$).
257:
258: X-ray observations with the XRT began 6.2 hr after the BAT trigger.
259: The X-ray light curve measured until $t\approx 5\times 10^{5}$~s is
260: well fitted by a power-law time decay $t^{-1.68\pm0.03}$. The
261: time-integrated spectrum is well fitted ($\chi^2/\nu=48.47/54$) by an absorbed
262: power-law model [photon index $\Gamma=2.14\pm 0.12$, unabsorbed
263: $F_X=(5.8\pm 0.4) \times 10^{-13}$ erg cm$^{-2}$ s$^{-1}$]. The
264: equivalent H column density, $N_{\rm H} = (2.2\pm 0.4) \times 10^{21}$
265: cm$^{-2}$, is marginally consistent with the expected Galactic column
266: density in the source direction, $N_{\rm H}=1.1\times 10^{21}$ cm$^{-2}$
267: \citep{Dickey1990}. Examining the X-ray hardness ratio
268: \citep[e.g.,][]{ButlerKocevski2007b}, there is no evidence for
269: spectral evolution during the XRT observation.
270:
271: \subsection{KAIT Observations}
272: % S 2.2
273:
274: The Katzman Automatic Imaging Telescope (KAIT) is a 0.76-m robotic
275: telescope at Lick Observatory that is dedicated to searching for and
276: observing supernovae and monitoring other variable or ephemeral
277: celestial phenomena. It is equipped with a Finger Lakes Instrument
278: (FLI) ProLine PL77 back-illuminated CCD camera having a resolution of
279: $0\farcs8$ pixel$^{-1}$ and a total field of view (FOV) of $\sim$
280: $6\farcm8\times6\farcm8$. More information on KAIT can be found in
281: \citet{Li+2000}, \citet{Filippenko+2001}, and \citet{Filippenko2005},
282: while the KAIT GRB alert system is described in detail by
283: \citet{Li+2003a}. Notable KAIT observations of GRBs include
284: GRB~021211 \citep{Li+2003b}, GRB~051111 \citep{Butler+2006},
285: GRB~060210 (\citealt{GCN4727}, Li et al. 2008 in preparation), and
286: GRB~080319B \citep{Bloom+2008}.
287:
288: Several improvements have been implemented for the KAIT GRB alert
289: system since the description given by \citet{Li+2003a}. An FLI PL77
290: camera has replaced the Apogee AP7 camera, offering a much faster
291: readout time (1.2~s for FLI vs. 11.0~s for Apogee). A new feature
292: has been incorporated into the software so the system can easily
293: terminate an ongoing exposure in preparation for the GRB response
294: sequence. Most importantly, a real-time image-processing pipeline has
295: been developed to compare the KAIT images to archival Digital Sky
296: Survey (DSS) images to identify new objects. Astrometry solutions are
297: derived for the KAIT images by matching the detected objects to the
298: USNO B1 catalog \citep{Monet+2003}, providing coordinates to any new
299: objects to a precision of $\sim$0$\arcsec$.2. Point-spread-function
300: (PSF) fitting photometry is also performed on new objects, and
301: calibrated to the red magnitudes of the stars in the USNO B1 catalog.
302: The image-processing results are displayed in real time on a
303: website.\footnote{{\url
304: http://hercules.berkeley.edu/grbdata/grbfinder.gif .}}
305:
306: \begin{figure*}
307: \centerline{\includegraphics[width=5.5in,angle=0]{f2.eps}}
308: \caption[relation] {Sequence of KAIT images for the OA of GRB
309: 071003. An $80\arcsec \times 80\arcsec$ section is shown for the
310: first and fifth unfiltered 5~s images and for a 20~s unfiltered image
311: that started at $t = 431$~s. The OA is the central object in the
312: circles. It is well detected in the early images and rapidly fades.
313: The image quality is poor owing to the very high airmass of the object.}
314: \label{fig:kaitseq}
315: \end{figure*}
316:
317: For GRB\,071003, the KAIT GRB alert program received the GCN socket
318: notice at $t = 16$~s. The system immediately terminated the ongoing
319: supernova search program and began to slew the telescope to the GRB
320: position. After slewing from close to meridian to an hour angle of
321: 4.2 hr, a sequence of 5$\times$5~s unfiltered images began at
322: $t = 42$~s. KAIT then switched to a sequence that alternated with
323: 20~s $V$, $I$, and unfiltered images. Finally, the sequence
324: converted to 20~s $I$ and unfiltered images. Because of the physical
325: west hour angle limit of 4.7 hr, KAIT only finished part of this
326: pre-arranged sequence. In total, 56 images were obtained in the $V$,
327: $I$, and unfiltered passbands from $t = 42$ to $1628$~s, with full
328: width at half-maximum intensity (FWHM) of $\sim$3$\arcsec$.
329:
330: Visual inspection of the image-processing results revealed a true new
331: object, first reported by our group \citep{GCN6838}, measured at 12.8
332: mag at a position of $\alpha$ = $20^h07^m24^s.12$, $\delta$ =
333: $+10^\circ56\arcmin51\arcsec.8$ (equinox 2000.0; approximate $1\sigma$
334: astrometric uncertainty 0\arcsec.3). Our candidate OA was
335: subsequently confirmed by observations from the automated Palomar
336: 60-inch (1.5 m) telescope (P60; \citealt{GCN6839}). Further preliminary analysis of OA
337: early-time behavior from the KAIT observations was reported by
338: \citet{GCN6844}. Figure \ref{fig:kaitseq} shows a sequence of the KAIT
339: images for the OA of GRB\,071003. An $80\arcsec \times 80\arcsec$
340: section is shown for the first and fifth unfiltered 5~s image and a
341: 20~s unfiltered image that started at $t = 431$~s. As seen in Figure
342: \ref{fig:kaitseq} and reported by several groups
343: \citep{GCN6838,GCN6839,GCN6840,GCN6844}, a bright ($R \approx 11$ mag)
344: foreground star is located 6$\arcsec$.5 west of the OA of GRB
345: 071003. As discussed in \S 3, the presence of this bright star
346: complicates the photometry for the OA, and various methods have been
347: used to minimize its contamination.
348:
349: \subsection{P60 Observations}
350: % S 2.3
351:
352: The Palomar 60-inch telescope (P60; \citealt{cfm+06}) automatically
353: responded to the \textit{Swift} trigger for GRB\,071003, beginning a
354: pre-programmed sequence of observations at 07:43:51 UT (176~s after
355: the trigger). Observations were taken in the Kron $R$, Sloan
356: $i^{\prime}$ and $z^{\prime}$, and Gunn $g$ filters at large airmass
357: ($> 2.5$). Individual images were reduced in real time by our
358: automated reduction pipeline. Manual inspection revealed a fading
359: point source \citep{GCN6839} in all four filters at the location
360: reported by \citet{GCN6838}.
361:
362: A second epoch of observations was manually scheduled for the night of
363: UT October 4. In an attempt to lessen the contamination of the nearby
364: bright saturated star, these observations were taken in the Johnson
365: $V$-band filter in relatively short (30~s) exposures. A sequence of
366: 30 images was obtained.
367:
368: \subsection{AEOS Observations}
369: % S 2.4
370:
371: The 3.6-m US Air Force Advanced Electro-Optical System (AEOS)
372: telescope, located at the Maui Space Surveillance System on Haleakala\footnote{Based on data from the Maui Space Surveillance System, which is operated by Detachment 15 of the U.S. Air Force Research Laboratory's Directed Energy Directorate.},
373: observed the OA of GRB 071003 with the AEOS Burst Camera (ABC,
374: \citealt{Swan+2006}). ABC has a back-illuminated $2048 \times 2048$
375: pixel EEV chip, with a scale of 0$\arcsec$.189 pixel$^{-1}$ and a FOV
376: of $\sim$ $6\farcm5\times6\farcm5$. Because there is no direct internet
377: access to AEOS, after \textit{Swift} detected the GRB, a FAX alert
378: was automatically sent to the AEOS control room, to initiate a series
379: of Target-of-Opportunity (ToO) observations.
380:
381: The AEOS observations of GRB 071003 are all unfiltered 10~s exposures.
382: The first batch of images started at $\sim$9 minutes after the BAT
383: trigger, and 238 images were observed until $t \approx 83$ minutes, all
384: with very good image quality (FWHM $\approx$ 0$\arcsec$.9). The second
385: batch of images started at $t \approx 205$ minutes, and 56 images were
386: observed until $t \approx 222$ minutes. Due to the large airmass for these
387: observations and the degraded seeing conditions, however, the images
388: have rather poor quality. We have tried various methods to measure the
389: brightness of the OA in these images but failed. Accordingly, only the
390: first batch of 238 images is analyzed in this study. Preliminary
391: analysis of the AEOS observations is reported by \citet{GCN6841}.
392:
393: \subsection{Keck I/Gemini-S observations}
394: % S 2.5
395:
396: In response to the detection of the OA of GRB 071003, we organized a
397: campaign to obtain spectroscopy and late-time photometry with the 10-m
398: Keck I and the 8-m Gemini-S telescopes. At $t \approx 2.6$ hr, we
399: attempted to observe the OA with the HIRES spectrograph at Keck I, but
400: the data are of poor signal-to-noise ratio (S/N) and no obvious lines
401: were detected \citep{GCN6843}. Just before the HIRES spectroscopy
402: started, we also obtained guider images for the OA, providing
403: important photometric coverage during a gap in the photometry obtained
404: elsewhere (see \S 3.4). The guider images have a scale of
405: 0$\arcsec$.37 pixel$^{-1}$ with a FOV of 53$\arcsec$.5 $\times$
406: 71$\arcsec$.3.
407:
408: On 2007 October 4, we observed the GRB\,071003 OA with the Low Resolution
409: Imaging Spectrometer (LRIS; \citet{Oke+1995}) on Keck I. Anticipating
410: significant fading of the OA, a series of deep 300~s images was taken
411: with the $g$ and $R$ filters under excellent seeing conditions (FWHM
412: $\approx$ 0$\arcsec$.5). Inspection of the images reveals that the OA
413: was still bright and saturated in most of the images. Consequently,
414: only a single image in each of the $g$ and $R$ bands, where the OA is
415: not saturated, is analyzed in this study. LRIS uses a beamsplitter to
416: separate the light between two arms, red and blue. Both the blue and
417: red cameras have a usable FOV of $\sim$ $6\farcm0\times7\farcm8$. The
418: red camera used a back-illuminated Tek $2048 \times 2048$ pixel chip
419: with a scale of 0$\arcsec$.215 pixel$^{-1}$, while the blue camera has
420: a mosaic of two $2048 \times 4096$ pixel Marconi chips with a scale of
421: 0$\arcsec$.135 pixel$^{-1}$.
422:
423: Encouraged by the brightness of the OA, we also performed LRIS
424: spectroscopy of the OA. A preliminary analysis of the spectrum is
425: reported by \citet{GCN6850}, and a more detailed analysis is presented
426: in \S 3.5.
427:
428: We performed more LRIS imaging for the OA of GRB\,071003 on 2007 October
429: 8, 9, 10, 11, and 15, using various combinations of $u$, $g$, $V$, and
430: $R$ filters. The presence of the very bright star presents a
431: significant challenge to extracting useful data on the OA, as its
432: diffraction spikes change positions and intensity according to the
433: time and seeing conditions of the observations. Unfortunately,
434: observations on 2007 October 8 were adversely affected by diffraction
435: spikes and poor seeing, and were not usable. The data taken on 2007
436: October 15 are seriously affected by clouds, and do not provide an
437: interesting limit to the brightness of the OA, so they are not used in
438: this study.
439:
440: We also triggered our TOO program (GS-2007B-Q-2; PI H.-W.~Chen) for
441: GRBs with the Gemini-S telescope and obtained $g$-, $r$-, $i$-, and
442: $z$-band images with the GMOS camera on 2007 Oct. 5 and 6. The GMOS
443: camera is equipped with three back-illuminated EEV $2048 \times 4608$
444: pixel chips. For our observations, the camera is used in a 2$\times$2
445: binning mode with a scale of 0$\arcsec$.146 pixel$^{-1}$ and a FOV of
446: $\sim$5$\arcmin$.5$\times$5$\arcmin$.5. Unfortunately, the 2007 October 5
447: images are badly affected by bleeding from the very bright star and
448: are not used in this study.
449:
450: As part of the efforts to follow the evolution of the OA of GRB
451: 071003, we also performed adaptive optics (AO) observations with Keck
452: I on 2007 October 19 \citep{GCN7010}. The details of the AO observations
453: can be found in \S 3.6.
454:
455: \subsection{Radio Observations}
456: % S 2.6
457:
458: GRB\,071003 was observed with the Very Large Array (VLA)\footnote{The
459: NRAO is a facility of the National Science Foundation, operated under
460: cooperative agreement by Associated Universities, Inc.} on various
461: occasions. We made the observation in the B configuration array. We
462: used VLA source 1950+081 as phase calibrator for 4.86 GHz (C) band
463: observations and 2001+104 for 8.46 GHz (X) band observations.
464: The data were analyzed using standard data reduction routines of the
465: Astronomical Image Processing System (AIPS). The first observation
466: took place on 2007 October 5 in the X band with flux density of $393 \pm 55$
467: $\mu$Jy. Since then we made six observations in the X band and three
468: observations in the C band (Table \ref{tab:vla}).
469:
470: %Our results are consistent with the declining power law behavior.
471:
472:
473: \section{Data Reduction}
474: % S 3
475:
476: The bright star in the neighborhood of the OA of GRB 071003 makes it a
477: challenge to measure reliable photometry from the data described in \S
478: 2. In this section we describe the methods used to minimize its
479: contamination.
480:
481: \subsection{Photometric Calibrations}
482: % S 3.1
483:
484: For photometric calibrations, the field of GRB 071003 was observed in
485: $B$, $V$, $R$, and $I$ on two photometric nights (2007 October 7 and 8)
486: at Lick Observatory, using both KAIT and the Lick Nickel 1-m
487: telescope. About a dozen Landolt standard-star fields (Landolt 1992)
488: were observed at different airmasses throughout each photometric
489: night. Photometric solutions to the Landolt standard stars yield a
490: scatter of $\sim$0.02 mag for all the filters. The GRB 071003 field
491: was also observed for several sets of $BVRI$ images with different
492: depth on both nights. The photometric solutions are used to calibrate
493: a set of local standard stars in the GRB 071003 field. Because the GRB
494: 071003 field is quite crowded, the number of calibrated local standard
495: stars is large (Table \ref{tab:calib}). A finder for a subset of 23
496: relatively bright local standard stars is in Figure \ref{fig:finder}.
497: As seen in Table \ref{tab:calib}, the local standard stars in the
498: field of GRB 071003 are well calibrated, with standard deviation of
499: the mean (SDOM) of $\sim$0.01 mag for all the $BVRI$ bands. We
500: refer to this calibration as the ``Lick calibration" throughout the
501: rest of the paper.
502:
503: \begin{figure}
504: \centerline{\includegraphics[width=3in,angle=0]{f3.eps}}
505: \caption[relation] {Finder chart for a subset of local standard
506: stars in the field of GRB 071003. The field of view is
507: 6$\arcmin$.6$\times$6$\arcmin$.6. North is up and east is to the
508: left. The displayed image is the unfiltered template taken with KAIT
509: on 2007 October 9.}
510: \label{fig:finder}
511: \end{figure}
512:
513:
514: Several Landolt standard-star fields were also observed with LRIS at
515: Keck I: in the $u$, $g$, and $R$ bands on 2007 October 9, and in the $V$
516: band on 2007 October 11. As the number of the observed standard-star
517: fields is small, it is not possible to derive a complete photometric
518: solution for either night. Since the GRB field was observed at similar
519: airmasses with some of the standard-star fields, we can treat the LRIS
520: filters as standard and derive the magnitudes for the local standard
521: stars via differential photometry. Unfortunately, this procedure
522: suggests that the 2007 Oct. 9 night was not photometric, as different
523: standard-star observations yield somewhat different zero points. The
524: 2007 October 11 night was photometric, but only the $V$-band standard stars
525: were observed.
526:
527: We elected to use the Lick calibration as the foundation for all the
528: photometric calibrations, except in the case of the $u$ band. The
529: Lick-calibrated magnitudes are in $BVRI$, and can be reliably
530: converted to the $g$, $r$, and $i$ bands using color transformation
531: equations \citep{Jester+2005}. The conversion to the $z$ band
532: \citep{Rodgers+2006} is somewhat problematic, and as a result we adopt
533: a relatively large uncertainty for the converted magnitudes. For the
534: $u$ band, only two standard-star fields were observed with LRIS on
535: 2007 Oct. 9, and they give a difference of 0.30 mag in the zero
536: points. We chose to calibrate the GRB 071003 field with the
537: standard-star field that is closer in time of GRB observation, but
538: we added an uncertainty of 0.30 mag to all the calibrated magnitudes. We
539: note that the true error for the $u$-band calibration may be higher than
540: 0.30 mag due to the nonphotometric conditions on 2007 October 9.
541:
542: \subsection{KAIT Data Reduction}
543: % S 3.2
544:
545: The KAIT data were automatically processed with bias and dark current
546: subtraction and flat-fielding. The PSF of the OA is seriously
547: affected by the bright star which is less than 10 pixels away in the
548: KAIT images. Consequently, normal PSF-fitting photometry cannot fit the
549: peak and background of the OA simultaneously to produce a reliable
550: measurement.
551:
552: We use image subtraction to remove the contamination of the bright
553: star. To generate template images for subtraction, KAIT imaged the
554: GRB 071003 field in the unfiltered mode and in the $V$ and $I$ filters
555: for the next several nights after the burst. To make sure the bright
556: star is not saturated, short (5~s) exposures were used,
557: and 50--100 images for each filter were acquired to ensure high S/N
558: in the combined images. As discussed in
559: \S 4, the GRB OA was still reasonably bright in the second night after
560: the burst, so we used the images obtained at 4--6~days after the burst
561: as the template for the field without significant OA contribution.
562: Our image subtraction code is based on the ISIS package
563: \citep{Alard+1998} as modified by B.\ Schmidt for the High-$z$ Supernova
564: Search Team \citep{schmidt98}. An illustration of the image
565: subtraction is presented in the top panels of Figure
566: \ref{fig:imagesub}.
567:
568: \begin{figure}
569: \centerline{\includegraphics[width=3in,angle=0]{f4.eps}}
570: \caption[relation] {Illustration of using image subtraction to remove
571: the contamination of the bright nearby star to the OA of GRB 071003.
572: The KAIT image subtraction code is demonstrated here. (\textit{a}) An
573: $80\arcsec \times 80\arcsec$ section of the original 20~s unfiltered
574: KAIT image of the OA taken at $t = 431$~s; (\textit{b}) the same section after
575: image subtraction of the central $50\arcsec \times 50\arcsec$ using an
576: unfiltered template image after the OA has faded; (\textit{c}) an $80\arcsec
577: \times 80\arcsec$ section of the combined unfiltered AEOS image at $t
578: = 5002.6$~s; and (\textit{d}) the same section after image subtraction of the
579: central $30\arcsec \times 30\arcsec$ using a hand-made template
580: image. See text for more details. }
581: \label{fig:imagesub}
582: \end{figure}
583:
584:
585: The Lick calibration was used to transform the KAIT instrumental
586: magnitudes to the standard Johnson $V$ and Cousins $I$ passbands, with
587: proper color terms measured from the photometric nights. We also find
588: that the combination of the KAIT optics and the quantum efficiency of
589: the FLI CCD camera makes the KAIT unfiltered observations mostly mimic
590: the $R$ band. During the two photometric nights, unfiltered
591: observations of the Landolt standard-star fields were also performed.
592: Analysis of these images indicates that the KAIT unfiltered magnitudes
593: can be effectively transformed to the $R$ band, with a relatively
594: large color term and an rms of $\sim$0.05 mag,
595: similar to the earlier results we reported
596: \citep{Li+2003a,Li+2003b,Butler+2006}.
597:
598: To increase the S/N, the late-time KAIT images of GRB 071003 were
599: combined into groups of three to eight images. The final KAIT photometry for
600: the GRB 071003 OA is listed in Table \ref{tab:kait}. The reported
601: error bars are the uncertainties in PSF-fitting photometry and those
602: in the calibration process, added in quadrature. A plot of the KAIT
603: photometry, along with measurements from other telescopes during the
604: same timespan (with BAT data overplotted and fitted by a chromatic model
605: described in \S~4.3) is presented in Figure \ref{fig:earlylc}.
606:
607: \begin{figure}
608: \centerline{\includegraphics[width=3.5in,angle=0]{f5.ps}}
609: \caption[relation] {Early-time light curve of the optical afterglow of
610: GRB~071003 using KAIT photometry, supplemented by observations from
611: P60 and AEOS. The gamma-ray light curve from the BAT is overplotted
612: in gray (scaled arbitrarily). A clearly additive ``bump'' at
613: 100--500~s is apparent. Photometric follow-up observations continued after
614: 2000~s with P60 and AEOS, as well as with Gemini and Keck in subsequent
615: nights; the complete 16-day optical light curve is presented in Figure
616: \ref{fig:lc}.}
617: \label{fig:earlylc}
618: \end{figure}
619:
620:
621:
622:
623: \subsection{AEOS Data Reduction}
624: % S 3.3
625:
626: The ABC images were processed using dark subtraction only. Because of
627: highly variable stray light and vignetting, we did not apply a flat
628: field to these images. We used SExtractor \citep{Bertin+1996} to find
629: all the sources in the images, from which we were able to determine
630: the astrometry.
631:
632: We employed the NN2 flux difference method (\citealt{Barris+2005};
633: hereafter the NN2 method) for constructing the AEOS light curve. The
634: NN2 method also uses image subtraction to measure the fluxes for a
635: variable source, but it does not designate one particular image as the
636: template. Instead, given $N$ total observations, the NN2 method solves
637: for the vector of fluxes from the individual images using the
638: antisymmetric matrix of flux differences from the $N(N-1)/2$ distinct
639: possible subtractions. Compared to the template image subtraction
640: method, the NN2 method takes all the available information from the
641: images into account, and is less susceptible to possible noise
642: associated with a single template image. To avoid a large number of
643: image subtractions, we combined the original 238 AEOS observations
644: into 39 images. For the first 228 images, each set of six consecutive
645: images is combined into one. The last 10 images are combined into a
646: single image. We compared the results from the NN2 method to those
647: from a traditional template image subtraction method (bottom panels of
648: Figure \ref{fig:imagesub}) and found them to be consistent with each
649: other.
650:
651: To calibrate the AEOS data to the standard photometry system, we used
652: the KAIT $R$-band data during the overlap period and assume that the
653: unfiltered AEOS data have no color term to the $R$ band.\footnote{We
654: attempted to quantify the color term of the unfiltered AEOS data to
655: the standard $R$ system using the local standard stars in the field of
656: GRB 071003, but found no apparent correlation between the scatter of
657: the (unfiltered $-$ $R$) differences versus the colors of the stars.}
658: The final AEOS photometry is listed in Table \ref{tab:aeos}. The
659: reported error bars are only those output by the NN2 method, and do
660: not include a possible large systematic error due to calibration. If
661: the throughput of the AEOS telescope in the unfiltered mode is not
662: drastically different from that of KAIT, we estimate the systematic
663: error to be $\sim$0.07 mag when the GRB OA was bright ($t < 20$ minutes),
664: and $\sim$0.15 mag when the GRB became faint ($t > 40$ minutes). The
665: systematic errors can be much higher if the unfiltered throughput is
666: very different for the two telescopes.
667:
668: \subsection{Keck I/Gemini-S Data Reduction}
669: % S 3.4
670:
671: Due to the large aperture of the Keck I and Gemini-S telescopes, the
672: bright star close to the GRB 071003 OA produces numerous diffraction
673: spikes, as well as two large blooming spikes along the readout
674: direction. Because the orientation, width, and intensity of the
675: spikes change with the seeing conditions, the exposure duration, and
676: the time of the observations, it is difficult to cleanly remove them
677: using the template image subtraction or the NN2 method. However, due
678: to the high resolution of these images, the spikes are well sampled
679: and show distinct axial symmetry. We developed a saturation spike
680: subtraction method, in which we divide the image of the bright star in
681: half, flip the right side, and subtract it from the left side. Due to
682: the symmetry in the spikes, this subtraction process leaves a
683: reasonably clean region around the GRB OA. PSF-fitting photometry was
684: then performed on the GRB OA in the spike-subtracted images, and on a
685: series of local standard stars. The Lick calibration is used to
686: calibrate the Keck I and Gemini-S instrumental magnitudes to the
687: standard system.
688:
689: The final Keck I and Gemini-S photometry is reported in Table
690: \ref{tab:keck}. The error bars of the magnitudes are the
691: uncertainties from the PSF-fitting photometry and those in the
692: calibration process added in quadrature. One special data point is
693: the Keck I HIRES guider image at $t = 9523.7$~s because it bridges the
694: early KAIT/AEOS data to the late-time Keck I and Gemini-S
695: observations. The GRB OA was well detected in the guider image, but
696: because the image has a small FOV and is unfiltered, photometric
697: calibration becomes particularly difficult.
698:
699: We have used three methods to calibrate the measured instrumental
700: magnitude of the OA after the guider images were processed with the
701: saturation spike subtraction method: differential photometry between
702: the AEOS unfiltered data and the guider images, photometric
703: calibration to about half a dozen stars in the HIRES guider images
704: using the KAIT unfiltered images, and photometric calibration to these
705: stars using the Keck I $R$-band images. The measured $R$-band
706: magnitudes from these three methods show a scatter of $\sim$0.25 mag,
707: and their average value and uncertainty are listed in Table
708: \ref{tab:keck}.
709:
710: \subsection{P60 Data Reduction}
711: % S 3.5
712:
713: The P60 data reduction is presented in this section because it employs
714: several methods (illustrated in Figure \ref{fig:imagesub}) discussed
715: earlier in the paper. We obtained template images for the field after
716: the OA of GRB 071003 has faded. However, the saturation spikes of the
717: bright star close to the GRB ruined the template images in the $R$ and
718: $i^\prime$ bands, so we were only able to run image subtraction for
719: the data in the $g$ and $z^\prime$ bands. We also employed the
720: saturation spike subtraction methods as described in \S 3.4. Although
721: P60 does not have the resolution of the Keck I and Gemini-S telescopes,
722: subtraction of half of the saturation spikes helped to clean up the
723: background of the OA considerably.
724:
725: We also applied a third method to reduce the P60 data. Due to the
726: richness of stars in the GRB 071003 field and the large field of view
727: of the P60 camera ($12\arcmin.9\times12\arcmin.9$), we were able to
728: pick a star that is close in brightness (within 0.1 mag in all
729: filters) and thus has similar saturation spikes to the bright star
730: close to GRB 071003. The chosen star is located at $\alpha$ =
731: $20^h07^m14^s.84$, $\delta$= $+10^\circ53\arcmin59\arcsec.8$ (equinox
732: J2000.0), which is $136\arcsec.7$ west and $172\arcsec.0$ south of the
733: GRB 071003 OA. By slightly scaling the PSF of this bright star and
734: subtracting it from the star close to the GRB, we were able to largely
735: remove the complicated background around the GRB OA.
736:
737: PSF-fitting photometry is applied to the images after different ways
738: of image subtraction, and the Lick calibration is used to calibrate
739: the instrumental magnitudes into the standard system. The final
740: photometry from the P60 data is listed in Table \ref{tab:p60}, which
741: is the average of the spike and bright star subtraction methods. The
742: results from the template image subtraction method are not considered
743: because the method can only be applied to a subset of filters, but
744: they are consistent with the other two methods within measurement
745: uncertainties.
746:
747: \subsection{Keck AO Data Reduction}
748: % S 3.6
749:
750: \begin{figure}
751: \centerline{\includegraphics[width=3in,angle=0]{f6.ps}}
752: \caption[relation] {NGS AO image
753: of the GRB~071003 field taken with Keck II on 2007 October 19, 16~days after
754: the burst. The FOV is approximately $12\arcsec \times 10\arcsec$. The
755: afterglow is well detected with $K' = 21.58 \pm 0.03$ mag. No
756: host-galaxy emission is detected. }
757: \label{fig:keckAO}
758: \end{figure}
759:
760:
761: On 2007 October 19 (starting at UT 05:14) we observed the GRB 071003 OA
762: with the NIRC2 \citep{VanDam+2004} narrow-field camera (0$\arcsec$.01
763: pixel$^{-1}$) on Keck II using natural guide star adaptive optics (NGS
764: AO). While the extremely bright nearby star greatly complicated the
765: optical analysis, it was ideal to be used as the natural guide star
766: during NGS AO imaging. We took 15 science exposures, each of 60~s and
767: 2 coadds, resulting in a total integration time of 30 minutes. The images
768: were reduced using standard techniques, including dark subtracting,
769: flat fielding, and filtering for deviant pixels. Each frame was
770: dewarped using the recommended method for NIRC2, and the resulting
771: images were registered to a common origin and combined.
772:
773: The GRB OA is well detected 2 weeks after the burst, as shown in the
774: final combined image in Figure \ref{fig:keckAO}. To measure the
775: brightness of the OA, we created a model of the PSF using short-exposure,
776: unsaturated images of a nearby Two Micron All Sky Survey (2MASS) star ($K_s = 12.011
777: \pm 0.024$ mag, $d = 7.8\arcsec$), taken immediately prior to the
778: science exposures. We then subtracted this model PSF from the OA.
779: With the same 2MASS star as the photometric
780: calibrator, we measure the OA to have $K^\prime$ = 21.65 $\pm$ 0.10 Vega mag.
781: (Galactic reddening of $A_{K^\prime} \approx 0.05$ mag is negligible
782: along this sightline and has not been applied.)
783:
784:
785: \subsection{Keck LRIS Spectroscopy Reduction}
786: % S 3.7
787:
788: We obtained low-resolution optical spectroscopy of the optical
789: afterglow of GRB 071003 on 2007 October 4.335 using the LRIS on the
790: Keck I telescope. A pair of 600~s dithered exposures was taken
791: under clear conditions at airmass 1.2 with 0.6\arcsec\ seeing. We
792: used both the blue and red arms of LRIS, with the light split by the
793: D680 dichroic. The 300/5000 grism on the blue side gave a spectral
794: resolution of 8.4~\AA\ over the range 3300--6500~\AA . We used the
795: 600/10000 grating to achieve 4.1 \AA\ resolution over the range
796: 6500--8630~\AA . The spectrophotometric standard star Feige 110
797: \citep{Stone1977} was observed the following night in the same setup.
798: Intermittent clouds were present the night of the standard-star
799: observation, so the absolute flux scale is unreliable.
800:
801: The long, 1.0\arcsec-wide slit was oriented at a position angle of
802: 10$^\circ$ for the afterglow observations, which was not the
803: parallactic angle \citep{filippenko82}. However, the Cassegrain
804: Atmospheric Dispersion Compensator module \citep{Phillips+2006} was
805: mounted, so the derived spectral shape should be reliable. The
806: exception is in the spectral range of 6000--6500~\AA, where
807: second-order blue light contamination is prominent in the spectrum of
808: the standard star. An attempt was made to correct for the
809: contamination, but the spectral slope in this section is more
810: uncertain than in the rest of the spectrum. We also fitted an
811: extinction-corrected power law to the flux-calibrated spectrum
812: (excluding line and second-order contaminated regions) in an attempt
813: to estimate the spectral slope, but the estimated slope of $f_{\nu}
814: \propto \nu^{-0.87}$ differs significantly from the spectral slope
815: estimated from multi-band late-time photometry (\S 4.7). This may be
816: due to continuum contamination from the nearby star in the spectrum
817: (which is difficult to properly remove), so we do not further consider
818: this spectroscopic spectral index.
819:
820: \begin{figure}
821: \centerline{\includegraphics[width=3.5in,angle=0]{f7.eps}}
822: \caption[relation] {Spectrum of the GRB 071003 afterglow covering the
823: full observed spectral range. The spectrum has been flux-calibrated
824: and corrected for Galactic reddening of $E(B-V)=0.148$ mag. The inset
825: shows an expanded view of the region surrounding the Fe and Mg
826: absorption system at the burst redshift. A power-law continuum was
827: fitted to the regions of the spectrum shown in green, chosen to avoid
828: strong absorption lines and the wavelength range contaminated by
829: second-order blue light. The thick solid blue line shows the
830: resultant fit ($f_{\lambda} \propto \lambda^{-1.13}$, or $f_{\nu}
831: \propto \nu^{-0.87}$), but it differs in slope from our more reliable
832: fit to the broadband photometry; thus, it is used only to normalize
833: the spectrum.}
834: \label{fig:kecksp}
835: \end{figure}
836:
837:
838: The largely featureless spectrum (Figure~\ref{fig:kecksp}) has a
839: S/N $>5$ ${\rm pixel}^{-1}$ down to $\sim$3500~\AA. There is no apparent
840: absorption by the intergalactic medium at these wavelengths,
841: yielding an upper limit to the redshift of the burst of $z_{GRB} <
842: (3500/1216) - 1 = 1.88$. Numerous metal-line absorption lines (but no
843: emission lines) are visible in the spectrum. We have fitted the
844: equivalent widths of all $\gtrsim 5 \sigma$ features in the normalized
845: spectrum using a Gaussian profile and report the rest-frame values in
846: Table~\ref{tab:lines}.
847:
848: We previously presented \citep{GCN6850} analysis of this spectrum,
849: identifying \ion{Mg}{2} absorption systems at $z=0.372$ and
850: $z=1.100$. A VLT spectrum acquired the same night \citep{GCN6851}
851: identified a third absorption system at $z=0.937$, which is confirmed
852: by our observations. These are the only strong absorption systems in
853: the data, and previously we considered it likely that the $z=1.100$
854: system originated from the host galaxy (Figure \ref{fig:intervene}).
855: Surprisingly, however, a more thorough investigation revealed a
856: fourth, weak absorption system at a higher redshift of $z=1.604$
857: (Figure \ref{fig:grbvel}). Contrary to our expectation, the gas at
858: this redshift has the weakest \ion{Mg}{2} absorption of the four
859: systems.
860:
861: This is remarkable: absorption lines associated with GRB environments
862: are generally very strong with rest-frame equivalent widths exceeding
863: several angstroms \citep{sff03,pcw+08}. Figure~\ref{fig:intervene}
864: also indicates, however, the presence of fine-structure \ion{Fe}{2}
865: transitions at this redshift. With the exception of active galactic
866: nucleus environments, these transitions have only been identified in
867: gas surrounding the GRB phenomenon \citep{pcb06}. These transitions
868: are excited by the GRB afterglow itself through indirect ultraviolet
869: pumping \citep{pcb06,vls+07} of gas in the interstellar medium (ISM)
870: of the host galaxy. Altogether, the coincidence of (1) the absence of
871: any higher-redshift absorption systems in our spectrum, (2) the
872: positive detection of fine-structure \ion{Fe}{2} transitions, and
873: (3) the absence of intergalactic medium absorption at $\lambda >
874: 3500$~\AA\ establishes $z=1.604$ as the redshift of GRB~071003.
875:
876: \begin{figure}
877: \centerline{\includegraphics[width=2.7in,angle=90]{f8.ps}}
878: \caption[relation] {Portions of the normalized Keck LRIS spectrum of
879: the GRB~071003 afterglow. We mark the positions of several metal
880: absorption-line features from four distinct extragalactic systems
881: including a series of \ion{Fe}{2} and \ion{Fe}{2}* transitions
882: associated with the host galaxy of GRB~071003 ($z_{GRB} = 1.604$).
883: Note that the \ion{Ca}{2} doublet marked as Galactic may be due to the
884: very bright Galactic star offset by 6.5$''$ from GRB~071003 as opposed
885: to the Galactic ISM.}
886: \label{fig:intervene}
887: \end{figure}
888:
889: \begin{figure}
890: \centerline{\includegraphics[width=3.5in,angle=0]{f9.ps}}
891: \caption[relation] {Velocity plot of strong, resonance-line
892: transitions for gas associated with GRB~071003 ($z_{GRB}=1.60435$).
893: These lines are very weak (note the ordinate scale) with rest-frame
894: equivalent widths of 100--200~m\AA\ (Table~\ref{tab:lines}).
895: Indeed, the \ion{C}{4} absorption is the weakest yet reported for a
896: GRB afterglow \citep{pcw+08}.}
897: \label{fig:grbvel}
898: \end{figure}
899:
900:
901: It might seem unusual to have detected fine-structure \ion{Fe}{2}
902: transitions in such a late-time spectrum ($t \approx 24.3$~hr).
903: Because the lines are excited by the GRB afterglow, they will decay as
904: the afterglow fades on hour-long timescales
905: \citep{dcp+06,vls+07,DElia+2008}. The presence of fine-structure
906: transitions in our spectrum, however, is consistent with the late-time
907: rebrightening of GRB~071003 provided that the gas lies within a few kiloparsecs of
908: the GRB. In Figure~\ref{fig:grbvel} we present a velocity plot of
909: strong resonance-line transitions for $z = z_{GRB}$. We report the
910: positive detections of \ion{C}{4}~$\lambda$1548,
911: \ion{Fe}{2}~$\lambda\lambda$2382, 2586, 2600, and
912: \ion{Mg}{2}~$\lambda$2803, and we note probable but statistically
913: insignificant absorption at \ion{Al}{2}~$\lambda$1670 and
914: \ion{Mg}{2}~$\lambda$2796. The rest-frame equivalent widths are
915: among the lowest ever recorded for the ISM surrounding long-duration
916: GRBs. The equivalent width of \ion{Mg}{2}, for example, is fully an
917: order of magnitude below the general population \citep{Cenko+2008},
918: with the sole exception of GRB~070125, and
919: the equivalent width for the \ion{C}{4} gas ($W_{1548} = 0.22 \pm
920: 0.06$~\AA), represents the lowest measurement to date \citep{pcw+08}.
921:
922:
923:
924: \section{Results and Modeling}
925: % S 4
926:
927: \subsection{Light Curve: General Observations}
928: % S 4.1
929:
930: The multi-color photometric evolution of the GRB 071003 OA is shown in
931: Figure \ref{fig:lc}, fitted by our preferred model (described later).
932: Visual inspection of the light curves reveals what appear to be three
933: distinct components: an overall power-law decline that has already set
934: in by the very first measurement at 42~s, a small ``bump'' feature at
935: $\sim$120--600~s, and then a dramatic, but unfortunately not well
936: sampled, rebrightening starting around 3000~s that dominates the
937: remainder of the evolution.
938:
939: \begin{figure*}
940: \centerline{\includegraphics[width=6in,angle=0]{f10.ps}}
941: \caption[relation] {Multi-color, early through late-time light curves
942: of the OA of GRB~071003. The magnitudes are offset
943: according to their early-time colors, showing the color evolution
944: between early and late times. Overplotted colored curves indicate the
945: best-fit three-component, color-evolution model described in the text;
946: the dashed lines represent the individual components that compose this
947: model (a uniform power-law decay, a chromatic early-time bump, and a
948: monochromatic late-time rebrightening). The X-ray and gamma-ray
949: afterglows are also overplotted for comparison. The gamma-ray light
950: curve is scaled arbitrarily; if scaled based on the likely
951: gamma-to-X-ray spectral index it would fall on or near the
952: extrapolation of the X-ray light curve back to early times.}
953: \label{fig:lc}
954: \end{figure*}
955:
956:
957: The bump feature appears to be additive only: fitting a single power
958: law to measure the decay index ($t^{-\alpha}$) for the clear-band data
959: both before this period and after it, the power-law indices ($\alpha =
960: 1.47$ and $\alpha = 1.49$, respectively) are fully consistent with
961: each other and with the overall decay index over both periods
962: ($\alpha = 1.48$).
963:
964: The rebrightening is more difficult to characterize. We have no
965: observations between the Keck I HIRES guider point at $t \approx 2.6$
966: hr and our observations the second night; moreover, the points
967: reported in the GCNs are highly discrepant. An optical $R$-band limit
968: is reported at $t \approx 4$~hr by \citet{GCN6846}, which seems to
969: contradict the rebrightening trend suggested by the AEOS data and
970: guider point. It is unlikely that the OA would show such a dramatic drop
971: ($> 3$ mag) in a short time interval at such late times, so we suspect that
972: the OA might be heavily contaminated by the bright nearby star and was
973: not resolved in the Lulin 1-m telescope images of \citet{GCN6846}. On
974: the other hand, the $U$-band detection at $t \approx 7.5$ hr reported
975: by \cite{GCN6840} supports a rebrightening but is several magnitudes
976: above the extrapolated light curve at this time, seemingly far too
977: bright to be consistent with our observations. Calibration and the
978: contamination from the bright star are the likely causes of the
979: discrepancy.
980:
981:
982: \subsection{Optical to Gamma-Ray and X-Ray Comparison}
983: % S 4.2
984:
985: The BAT and XRT light curves we derive for GRB~071003 are also shown
986: in Figure \ref{fig:lc}. Unfortunately, because \textit{Swift} was
987: still in the process of returning to normal operations after its gyro
988: failure \citep{GCN6760}, automatic slewing to GRB~071003 was disabled
989: at the time when the GRB was detected. As a result, there were no
990: prompt XRT observations for GRB 071003, leaving a long gap in the
991: gamma-ray/X-ray light curve at $t = 200$--20000~s. In
992: particular, there are no X-ray observations until approximately the
993: peak of the rebrightening in the optical band. Nevertheless, direct
994: comparison of the data available reveals three relevant facts.
995:
996: First, there is no obvious optical prompt counterpart to the last
997: spike of the gamma-ray light curve. However, this spike is nearly
998: contemporaneous with the much more slowly rising optical bump feature;
999: we return to this possible connection in our later modeling (\S
1000: 4.5).
1001:
1002: Second, at late times the X-ray light curve declines as a power law
1003: with decay index consistent with that observed in the optical. A
1004: simple power law fits the data well, with a best-fit decay index of
1005: $\alpha_X = 1.68 \pm 0.04$. In addition, the late-time OA behavior
1006: (after $t \approx 5 \times 10^4$~s) is consistent with a single
1007: power-law decay with an index of $\alpha_O = 1.72 \pm 0.07$, fully
1008: consistent with this value. As we note later, an extrapolation
1009: of the X-ray spectral index is also consistent with the optical
1010: observations, suggesting that at late times there is no need for an
1011: additional X-ray contribution (such as inverse Compton) or large
1012: amounts of host-galaxy extinction.
1013:
1014: Finally, while the gamma rays are scaled arbitrarily in Figure
1015: \ref{fig:lc}, we note that if we extrapolate the gamma-ray spectrum
1016: into the X-rays to compare the BAT and XRT light curves, the evolution
1017: between the end of the prompt emission and the start of the XRT
1018: observations is nearly consistent with a simple extension of the
1019: late-time XRT power law back to earlier times, without a need for a
1020: rebrightening or break. However, \textit{Swift} has shown previously
1021: \citep{Nousek+2006} that early-time X-ray light curves can conceal a
1022: wide variety of complex features, so we will not speculate further as
1023: to whether or not this was actually the case.
1024:
1025: \subsection{Detailed Optical Modeling}
1026: % S 4.3
1027:
1028: The procedure used to model the optical light curve is generally the
1029: same as that employed by \citet{Perley+2008}, but further generalized.
1030: For our fit model, we employ an unbroken power-law decay (component 0)
1031: plus two \citet{Beuermann+1999} functions (broken power-law pairs,
1032: components 1 and 2), but allow for different values of the functional
1033: parameters for each filter and component. The functional form is
1034:
1035: \begin{equation}
1036: \begin{array}{rcl}
1037: F_\nu & = & F_{0,\nu}\,(t- dt_0)^{-\alpha_0} \\
1038: && + F_{1,\nu}\,(0.5 \,(\frac{t - dt_1}{t_{p1}})^{-s_1 \alpha_{1,b}} + 0.5 \,(\frac{t - dt_1}{t_{p1}})^{-s_1 \alpha_{1,a}})^{-\frac{1}{s_1}} \\
1039: && + F_{2,\nu}\,(0.5 \,(\frac{t - dt_2}{t_{p2}})^{-s_2 \alpha_{2,b}} + 0.5 \,(\frac{t - dt_2}{t_{p2}})^{-s_2 \alpha_{2,a}})^{-\frac{1}{s_2}},
1040: \end{array}
1041: \end{equation}
1042:
1043: \noindent
1044: where for component 0, $\alpha_0$ is the power-law decay
1045: index, and $dt_0$ is an adjustment to the \textit{Swift}/BAT trigger
1046: time. For cmponent 1, $\alpha_{1,b}$ and $\alpha_{1,a}$ are the
1047: power-law decay indices for the rising and declining components,
1048: respectively, $dt_1$ is an adjustment to the \textit{Swift}/BAT
1049: trigger time, $t_{p1}$ is the time of the peak flux, and $s_1$ is the
1050: sharpness parameter. Component 2 has a similar function as component
1051: 1.
1052:
1053: Fitting this function with no constraints generates unrealistic
1054: results because of non-uniform sampling in different filters.
1055: However, we can make the following physically motivated assumptions to
1056: tie specific parameters and produce more physically meaningful
1057: results.
1058:
1059: \begin{enumerate}
1060: \item{We assume that the temporal decay index at any given time is
1061: independent of the filter, as is implicit in the notation ($\alpha$
1062: does not depend on $\nu$). This means that the color of a component
1063: cannot change except while the light curve of that component is
1064: breaking.}
1065:
1066: \item{Most importantly, we assume that differences between the spectra
1067: of the various model components can be described by changes in the
1068: power-law index of the intrinsic spectrum, modified by an arbitrary,
1069: but fixed, extinction law. Mathematically, this constraint is
1070: expressed as $F_{i,\nu} = \nu^{\Delta\beta_{ij}} F_{j,\nu}$.
1071: Physically, this assumption requires that external effects such as
1072: extinction, which might cause the spectrum of any component to deviate
1073: from a power law, affect all components equally and are not
1074: time-dependent. The extinction law itself (as well as the
1075: \emph{absolute} underlying index of any specific component) is fully
1076: general and can be fitted according to various models later.}
1077:
1078: \item{In addition, we assume that the rising segments of each
1079: component are also power laws, but not necessarily the same power laws
1080: as the falling segment, to allow for chromatic breaks. This imposes
1081: the following condition: $(\frac{t_{p},x}{t_{p},y}) =
1082: (\frac{\nu_y}{\nu_x})^{\Delta\beta_{ba}/\Delta\alpha_{ba}}$. Here $b$
1083: and $a$ refer to ``before'' and ``after'' the break of a specific
1084: component (0, 1, or 2), where the component index is omitted for
1085: clarity, and $x$ and $y$ refer to two different filters.}
1086: \end{enumerate}
1087:
1088: Fitting is performed, under these assumptions, using the IDL package
1089: mpfit \footnote{\url
1090: http://cow.physics.wisc.edu/$\sim$craigm/idl/idl.html .}.
1091:
1092: The assumptions involved in these constraints are, of course,
1093: oversimplifications for the full array of models that might be
1094: considered. In particular, this model allows only one break per
1095: component, but with an evolving synchrotron light curve plus a jet we
1096: may expect as many as three. However, it has the advantage of being
1097: simple and generates a single physically motivated parameter
1098: quantifying color change over each component.
1099:
1100:
1101: We perform a variety of fits under varying combinations of
1102: assumptions. Some of the possibilities we considered include
1103: the following:
1104:
1105: \begin{enumerate}
1106: \item{Forcing the bump (component 1) to have the same color as the
1107: uniform decay (component 0), or allowing it to be a different color
1108: overall.}
1109:
1110: \item{Forcing the bump itself to be achromatic over its evolution, or
1111: allowing it to contain a chromatic break.}
1112:
1113: \item{Forcing the late rebrightening (component 2) to have the same
1114: color as the uniform decay, or allowing it to have a different color.}
1115:
1116: \item{Fixing $dt_0$ for the early steep decay to be zero (the BAT
1117: trigger time), or allowing it to be free to vary.}
1118:
1119: \item{Fixing $dt_1$ for the bump component to be zero, to be equal to
1120: the beginning of the prompt-emission pulse that is nearly
1121: contemporaneous with it, or allowing it to be free to vary.}
1122:
1123: \item{Fixing $dt_2$ for the late rebrightening to be zero, or allowing
1124: it to be free to vary.}
1125: \end{enumerate}
1126:
1127: The results under various combinations of these assumptions are
1128: presented in Tables \ref{tab:colorchange} and \ref{tab:t0}. We
1129: discuss the implications of these results in the remainder of the
1130: paper.
1131:
1132: \subsection{Color Change}
1133: % S 4.4
1134:
1135: Detection of a GRB afterglow in filtered observations during prompt
1136: emission, as was the case here, is rare. The situation is even more
1137: intriguing since our multi-color prompt OA observations show an
1138: apparent bump feature (component 1) that is nearly contemporaneous
1139: with a rebrightening pulse in the gamma-ray light curve. Therefore,
1140: it is of great interest to attempt to measure the color of Component
1141: 1. By the same token, we have good spectral coverage of the afterglow
1142: both during the primary normal decay and during the fading of the
1143: dramatic late rebrightening, and any color difference may shed light
1144: on the origin of these features.
1145:
1146: We tested for color differences in three places: between component 0
1147: (rapid decay) and component 1 (bump), between component 0 and
1148: component 2 (rebrightening), and over the break of component 1 itself
1149: (since the rising spectral index may differ from the falling spectral
1150: index). In all cases we find evidence for color variation, although in
1151: each case only at the $\sim2\sigma$ level. The fading component of
1152: the bump is redder than the fading component of the uniform decay by
1153: $\Delta\beta$ = 0.75 $\pm$ 0.34, the bump feature is chromatic with a
1154: shift from the rising to falling component of $\Delta\beta$ = 1.11
1155: $\pm$ 0.47, and the rebrightening (for which we only have color
1156: information during the fading component) is also redder, by
1157: $\Delta\beta$ = 0.84 $\pm$ 0.31.
1158:
1159:
1160:
1161: One must be somewhat cautious in interpreting these results --- since
1162: different filters sample the data differently, systematic errors that
1163: affect only one portion of the light curve can masquerade as color
1164: change. Data reduction for GRB 071003 was also challenging due to the
1165: presence of the nearby bright star, as detailed in \S 3. In addition,
1166: we note that the degree of spectral index shifts noted is dependent
1167: on the model. In spite of these considerations, however, we feel that
1168: our conclusion of color change is reasonably secure in each case.
1169:
1170:
1171: \subsection{Energy Injection Times}
1172: % S 4.5
1173:
1174: It is often unclear what time is most appropriate to use as $t_0$ when
1175: fitting a power law to a GRB afterglow. Thanks to the extremely
1176: early-time clear-band data, it is possible to fit $t_0$ and constrain
1177: this within a few seconds in the case of GRB 071003. This fit,
1178: notably, gives a $t_0$ of exactly the trigger time ($dt_0 =
1179: -0.01 \pm 3.01$~s). The gamma-ray light curve (Figure
1180: \ref{fig:grblc}) fluence is strongly dominated by the initial pulse,
1181: which rises sharply and peaks within a few seconds, so this is not
1182: necessarily surprising.
1183:
1184:
1185: Some authors \citep{Blake+2005, Vestrand+2005, Vestrand+2006,
1186: Yost+2007} have presented evidence of an optical component rising
1187: coincident with the prompt emission, although significantly longer
1188: lasting. We can analyze whether the bump component observed in
1189: GRB~071003 may be such a
1190: feature by determining whether or not it can be fitted with a pulse that
1191: rises abruptly, contemporaneous with the prompt emission. While our
1192: power-law model is somewhat simplified and the sampling of the rise is
1193: extremely poor, we find that it generally does not: the best-fit $t_0$
1194: is intermediate between the trigger time and the time of the prompt
1195: emission spike ($\sim$125~s) at $dt_1 = 60 \pm 20$~s. This is a
1196: model-independent result, although it rests mostly on one data-point:
1197: the initial $V$-band measurement, representing an integration from
1198: 97 to 117~s after the BAT trigger ($\sim$18~s before the rise of the
1199: prompt emission spike), lies 0.14 mag above a simple power-law
1200: extrapolation from regions of the data excluding the bump, compared to
1201: a photometric error of only 0.03 mag. While it is possible to
1202: envision scenarios where a relatively slow optical rise might follow a
1203: gamma-ray pulse (any broadband feature with hard-to-soft evolution, or
1204: perhaps a late internal shock that later collides with and energizes
1205: the external shock), no model to our knowledge can explain why an
1206: optical flare would precede a gamma-ray pulse, so we take this as
1207: evidence that the two features are physically unconnected.
1208:
1209: While our sampling around the rise and peak of the late-time rebrightening is
1210: poor (and dominated by the difficult-to-calibrate AEOS and HIRES guider images),
1211: we can also attempt to fit the $t_0$ for the rebrightening
1212: component. This is significantly different from $t = 0$, with a
1213: best-fit initial time of $dt_2 = 1245 \pm 311$~s. (This is well short
1214: of its peak time of approximately 20~ks, so the effect on the light
1215: curve is minor.) No prompt-like fluctuations or other features are
1216: observed in the light curve in this region.
1217:
1218:
1219: \subsection{Radio Modeling}
1220: % S 4.6
1221:
1222: GRB 071003 is rare among \textit{Swift} bursts for having a bright
1223: radio afterglow. We were able to successfully detect the afterglow at
1224: two frequencies and several epochs spanning $\sim$2--20~days after the
1225: burst, including observations nearly contemporaneous with our optical
1226: data. The data are plotted in Figure \ref{fig:vlalc}.
1227:
1228: This GRB is not far off the Galactic plane, and the radio observations
1229: are affected by scintillation. Following \citet{Walker1998,
1230: Walker2001}, the afterglow is in the strong scattering regime for both
1231: X and C bands. An approximate modulation index (which estimates the
1232: fractional rms variation) is 0.4 in the C band and 0.6 in the X band,
1233: over a refractive timescale of $\sim$0.5~days in the X band and 2~days in
1234: the C band. This is longer than any integration (so the error is not
1235: reduced by integration time) but shorter than the interval between
1236: exposures (so errors are uncorrelated).
1237:
1238: Radio data were fitted using both an unbroken power-law model and a singly
1239: broken power-law model. We attempted the fit both before including
1240: uncertainties due to scintillation and with an additional 40\% flux
1241: error added to all C-band points and 60\% error added to all X-band
1242: points.
1243:
1244: \begin{figure}
1245: \centerline{\includegraphics[width=3.5in,angle=0]{f11.ps}}
1246: \caption[relation] {\small VLA radio light curve fitted to an unbroken
1247: power law. The uncertainties in the measurements have been increased
1248: compared to their statistical values to take into account the effect
1249: of interstellar scintillation. Some contemporaneous late-time optical
1250: points (scaled arbitrarily) are shown for comparison.}
1251: \label{fig:vlalc}
1252: \end{figure}
1253:
1254: Without the additional flux errors, the unbroken model is a poor fit,
1255: with $\chi^2/\nu = 15.32/6$. A single, monochromatic break
1256: improves the fit dramatically ($\chi^2/\nu = 2.45/4$). This
1257: improvement is significant at 97.4\% confidence. A monochromatic
1258: radio break of this nature is very difficult to explain physically.
1259: However, with scintillation flux errors folded into the light curve,
1260: we find that a simple power law is a more than adequate fit to the
1261: data ($\chi^2/\nu = 1.4/6$), which may suggest that we have
1262: overestimated the degree of modulation somewhat. (This is to be
1263: expected: the modulation index calculated is an upper limit as it
1264: strictly applies only for a point source. The afterglow has a
1265: physical size, which quenches the scintillation modulation somewhat.)
1266: Therefore, as a final modification, we scaled down this additional
1267: error until the final $\chi^2/\nu \approx 1$. Properties of the
1268: temporal fits are given in Table \ref{tab:radiomodel}.
1269:
1270: The uncertainty due to scintillation is in any event too large to
1271: allow any firm conclusions about the light curve. However, since only
1272: refractive scintillation is expected to be significant, the refractive
1273: timescale is much longer than the several-hour timescale of individual
1274: observations, and the C-band observations were in all cases taken
1275: immediately after the X-band observations, we do consider the
1276: measurement of the radio spectral index ($\beta_R = -1.15 \pm 0.42$)
1277: to be trustworthy regardless of any scintillation uncertainty.
1278:
1279: \subsection{Spectral Energy Distribution and Extragalactic Extinction}
1280: % S 4.7
1281:
1282: If our modeling assumptions are accurate (or nearly so), we can use
1283: our model to calculate the SED at any time using a combination of all
1284: the data available, rather than restricting the measurement to a small
1285: subset of the photometry and filters, even if the data were acquired
1286: at very different times in the evolution of the GRB and the color is not
1287: constant.
1288:
1289: We calculate the SED at two epochs. First, we calculate the SED at $t
1290: = 2.67$~days after the burst, the time of our four-color Gemini-South
1291: observations. In calculating this SED, we perform a slightly modified
1292: light-curve fit: we do not perform any filter transformations (e.g., to
1293: convert $r$ to $R$), but we fix all non-SED parameters to that derived
1294: from the light-curve analysis. In addition, we add in quadrature a
1295: calibration uncertainty equal to 5\% in all filters, with a few
1296: exceptions. For $z$, we use a 15\% uncertainty. For $u$, we use a
1297: 30\% uncertainty, for reasons described earlier. Finally, for $K'$,
1298: we use a large extra uncertainty of 50\% due to the possibility of a
1299: temporal break sometime between our last optical observations and the
1300: AO observations. (However, if such a break is absent, then the $K'$
1301: observation is much more precise than is given on the plots.)
1302: Unfiltered observations are not used. We also calculate an early-time
1303: SED during the ``normal'' power-law decay at $t = 1000$~s, using a fit
1304: excluding late-time measurements and measurements during the (possibly
1305: chromatic) bump. Addition of uncertainties is as for the late-time
1306: SED.
1307:
1308: The resulting SEDs are plotted in Figures \ref{fig:sed1000} and
1309: \ref{fig:sed2day}. After removing the effects of Galactic extinction
1310: (but not yet considering non-Galactic extinction), both SEDs are a
1311: reasonable fit to a power law, providing a general confirmation of our
1312: assumptions as well as indicating that the host or intervening
1313: galaxies do not impose a great deal of frequency-dependent extinction.
1314: In support of our analysis from the light-curve modeling, the spectral
1315: indices appear to differ from early to late times: $\beta_{1000s}$ =
1316: 0.62 $\pm$ 0.33, while $\beta_{2.67d}$ = 1.25 $\pm$ 0.09. (These
1317: values are direct fits to the data and do not include the effects of
1318: the small amount of extragalactic extinction we do believe to be
1319: present, which we discuss shortly.)
1320:
1321:
1322: Unfortunately there were no early-time observations outside the
1323: optical band, since \textit{Swift} was unable to slew rapidly.
1324: However, this GRB was observed nearly simultaneously in X-rays,
1325: optical, and radio during the declining phase of the late
1326: rebrightening. Therefore, it is possible to calculate a coeval
1327: late-time spectrum at all wavelengths simultaneously. The values at
1328: 2.67~days (the same as the first optical-only SED, above, which is
1329: also contemporaneous with XRT observations and within about half a day
1330: of the first VLA observation) are given in Table \ref{tab:flux} and
1331: plotted in Figure \ref{fig:bbsed}.
1332:
1333: Even without considering host-galaxy extinction, the optical and X-ray
1334: observations are nearly consistent with a common spectral index:
1335: $\beta_{O}$ = 1.25 $\pm$ 0.09, $\beta_{X}$ = 1.14 $\pm$ 0.12, and
1336: $\beta_{OX}$ = 0.90 $\pm$ 0.03. This consistency, plus the fact that
1337: the optical and X-ray temporal decays are identical ($\alpha_{O}$ =
1338: 1.72 $\pm$ 0.31, $\alpha_{X}$ = 1.68 $\pm$ 0.05), argues that
1339: both X-ray and optical are in the same synchrotron regime and the
1340: spectrum across this range is a simple power law. We assume this
1341: throughout the remainder of the analysis.
1342:
1343: \begin{figure}
1344: \centerline{\includegraphics[width=3.5in,angle=0]{f12.eps}}
1345: \caption[relation] {Optical SED of the
1346: GRB 071003 OA at 1000~s after the burst, fitted using the extinction
1347: constraints derived using the late-time SED. The intrinsic
1348: (pre-extinction) model spectrum is also shown.}
1349: \label{fig:sed1000}
1350: \end{figure}
1351:
1352:
1353: \begin{figure}
1354: \centerline{\includegraphics[width=3.5in,angle=0]{f13.eps}}
1355: \caption[relation] {Same as Figure \ref{fig:sed1000} but for $t =
1356: 2.67$~d after the burst. The data (plus an X-ray normalization,
1357: not shown) have been fitted with an SMC-like extinction law, with the
1358: best-fit curve overplotted. The intrinsic (pre-extinction) model
1359: spectrum is also shown.}
1360: \label{fig:sed2day}
1361: \end{figure}
1362:
1363: The deviations in the observed spectral index suggest the presence of
1364: a small amount of extragalactic extinction. Because of the presence
1365: of numerous absorbers and the unusually weak nature of the
1366: highest-redshift absorption system, however, the appropriate
1367: assumptions for modeling the extinction contribution are not clear.
1368: Although \ion{Mg}{2} is not an exact tracer of the presence of dust,
1369: the extremely weak line absorption at the likely host-galaxy redshift
1370: of $z=1.604$ suggests that the dust column at that redshift is nearly
1371: negligible. Among the remaining absorbers, the \ion{Mg}{2} system at
1372: $z=0.372$ is by far the strongest (by a factor of $\sim$3 in
1373: equivalent width compared to the next strongest system at $z=1.10$),
1374: and is likely to be the dominant contributor to any observed dust
1375: absorption. However, this is partially offset by the fact that dust
1376: at higher redshift is much more opaque (since the observed optical
1377: frequencies are in the rest-frame UV at $z > 1$), so for the moment we
1378: remain agnostic as to the actual redshift of the absorbing dust.
1379:
1380: We fit the optical spectrum simultaneously with the normalized X-ray
1381: flux of $F_{1{\rm keV}}$ = 0.036 $\pm$ 0.004 $\mu$Jy at 2.67~days.
1382: This value has already been corrected for photoelectric absorption (\S
1383: 2.1), and X-ray absorption is not considered in the fit, allowing the
1384: gas-to-dust ratio to be independent of the amount of extinction, $A_V$.
1385:
1386: Four different extinction models were tested. In addition to a
1387: control fit with no extinction, we fit for Milky Way-like, Small
1388: Magellanic Cloud (SMC)-like, and Large Magellanic Cloud (LMC)-like
1389: extinction using the parameterization of the
1390: \citet{FitzpatrickMassa1990} (``FM'') model, and a model for
1391: extinction in starburst galaxies parameterized by
1392: \citet{Calzetti+2000}. In all cases the standard average value of the
1393: ratio of total-to-selective extinction $R_V$ in the reference galaxy
1394: in question was used. (Fits with varying $R_V$ were attempted, but
1395: lacking infrared or ultraviolet measurements we were unable to
1396: constrain this parameter.) We performed separate fits assuming dust
1397: at $z = 0.372$, 1.100, and 1.604.
1398:
1399: Results are given in Table \ref{tab:extfits}. We find significant
1400: evidence ($f$-test: 96\% confidence) for a small amount ($A_V$ =
1401: 0.1--0.3 mag, depending on the model) of extinction along the light of
1402: sight. We cannot strongly constrain its nature; all four extinction
1403: laws, at each of the three possible redshifts, give reasonable fits to
1404: the observations. The intrinsic (pre-extinction) spectral slope
1405: $\beta$ is strongly constrained to be 0.94 $\pm$ 0.03, averaged across
1406: the different models. This is consistent (although marginally, at about
1407: the 90\% confidence level) with the absorption-corrected X-ray
1408: measurement of $\beta$ = 1.14 $\pm$ 0.12.
1409:
1410: As expected, the spectrum turns over dramatically somewhere redward of
1411: the optical and is declining with decreasing frequency by the radio
1412: band. The radio results are discussed further in \S 5.4.
1413:
1414: \begin{figure}
1415: \centerline{\includegraphics[width=3.5in,angle=0]{f14.eps}}
1416: \caption[relation] {\small Broadband SED at
1417: $t=2.67$~days from radio through X-ray observations. The shaded region
1418: shows an unbroken extrapolation of the X-ray fit (90\% confidence
1419: region), which is consistent with the optical measurements. The
1420: optical points are corrected for Galactic but not extragalactic
1421: extinction; a best-fit model for the effects of host-galaxy and
1422: intervening-galaxy extinction is shown (thin cyan line). The
1423: locations of the cooling break and peak frequency shown are
1424: arbitrarily chosen; the actual frequencies are not constrained by the
1425: available data except that both are located between the radio and
1426: optical bands.}
1427: \label{fig:bbsed}
1428: \end{figure}
1429:
1430:
1431: \subsection{Photometric Limits on a Host Galaxy and Intervening Absorbers}
1432: % S 4.8
1433:
1434: Neither our LRIS imaging nor our late-time NGS AO imaging
1435: show any evidence of extension or host-galaxy emission consistent with
1436: the afterglow position. We searched for emission from a host
1437: coincident with the OA position by smoothing and binning the
1438: PSF-subtracted AO image. No host emission was detected to a
1439: conservative upper limit of $K^\prime$ $\approx$ 23 Vega mag.
1440:
1441: In our first-night LRIS image (when the seeing was best and
1442: contamination from the bright nearby star relatively minimized), a
1443: faint, extended source is visible slightly southwest of the OA. The
1444: same source is also visible in the AO image, clearly resolved into a
1445: faint galaxy with $K^\prime \approx 19$ mag at an offset of
1446: 2.07$\arcsec$ southwest of the OA.
1447:
1448: We know from the spectral analysis that there are at least four
1449: systems that intersect the sightline between the $z=1.604$ GRB and
1450: Earth, including the host itself. Of these, the strongest candidate
1451: for association with the observed galaxy is clearly the $z=0.372$
1452: system, which both is closest and exhibits the strongest absorption
1453: signature. (Unfortunately, we have no spectra of the galaxy to
1454: confirm this.) This source appears to be a small irregular galaxy,
1455: which at this redshift would be offset by $\sim$ 10 kpc (a reasonable
1456: distance to explain the observed absorption) and approximately 0.5 kpc
1457: in half-light radius.
1458:
1459: No other extended sources are detected within 3\arcsec\ of the
1460: afterglow, so our upper limit rules out detection of both a host
1461: galaxy and any absorbing systems within this distance. The
1462: corresponding limit on a galaxy luminosity is only mild, compared to
1463: the known GRB host distribution. At the presumptive GRB redshift of
1464: $z = 1.6$, any host galaxy is limited to a $K$-band absolute magnitude
1465: of $M(K') = -22.2$ Vega mag.
1466: This value falls roughly in the middle of the typical range of
1467: previously studied GRB hosts, which appear to have $K$-band
1468: luminosities on the order of 0.1 $L_*$, and are bluer and fainter than
1469: typical SCUBA galaxies \citep{LeFloch+2003}.
1470:
1471: \subsection{Spectroscopic Constraints on the Host Galaxy and Intervening Absorbers}
1472: % S 4.9
1473:
1474: The very weak absorption at the host redshift in our spectrum suggests
1475: a lower than average \ion{H}{1} column density along the sightline
1476: and/or a metal-poor gas. Because of our low spectral resolution,
1477: however, the absorption is unresolved and the line profiles may be
1478: saturated \citep{pro06}. We may conservatively report a lower limit
1479: to the column densities by assuming the weak limit. In this manner,
1480: we estimate $N_{Mg^+} > 10^{12.6}~ {\rm cm}^{-2}$ based on the
1481: equivalent width of \ion{Mg}{2}~$\lambda$2803. For a solar
1482: metallicity gas, this implies $\log \mnhi > 10^{17}~ {\rm cm}^{-2}$.
1483: This is a conservative estimate because the gas metallicity is
1484: presumably subsolar. Nevertheless, it is unlikely that the gas has an
1485: \ion{H}{1} column density matching the values typical of most GRBs.
1486:
1487: In addition to the gas associated with GRB~071003, the afterglow
1488: spectrum reveals three foreground \ion{Mg}{2} absorbers. Two of these
1489: have moderate rest-frame equivalent widths ($W_{2796} \approx
1490: 0.7$~\AA), but the lowest redshift system exhibits a very large value
1491: ($z=0.3722$, $W_{2796} = 2.5$~\AA). The incidence of such strong
1492: \ion{Mg}{2} absorption at $z< 0.5$ has not yet been established along
1493: quasar sightlines. These absorbers are very rare at $z \approx 0.5$
1494: however, and the incidence is declining with redshift
1495: \citep{ntr05,ppb06}. The number of absorbers with $W_{2796} > 1~$\AA\
1496: per unit redshift is $\ell(z) = 0.13$ at $z = 0.5$, and the incidence
1497: of absorbers with $W_{2796} > 2~$\AA\ is an order of magnitude lower.
1498: This implies that one would need to observe of order 100 quasar
1499: sightlines to detect a single absorber with $W_{2796} > 2~$\AA\ at $z<
1500: 0.5$. Although these are {\it a posteriori} statistics, this analysis
1501: reminds one of the apparent enhancement of strong \ion{Mg}{2}
1502: absorbers along GRB sightlines \citep{ppc+06}. Given its low
1503: redshift, this system will be an excellent case to perform follow-up
1504: observations and examine the properties of the galaxies hosting such
1505: systems (Pollack et al. 2008, submitted)
1506: The bright nearby star, however, poses a formidable obstacle
1507: for non-AO ground-based observations.
1508:
1509: \subsection{Energetics}
1510: % S 4.10
1511:
1512: The measured gamma-ray fluence of 5.32 ($-$0.67, +0.30) $\times 10^{-5}$
1513: erg cm$^{-2}$ (Konus, 20~keV--4~MeV: \citealt{GCN6849}) can be converted
1514: to an isotropic-equivalent total energy release in the host frame:
1515: $E_{\rm iso}$ $=$ 3.4 ($-$0.6, +0.2) $\times 10^{53}$ erg --- well in
1516: the upper range of \textit{Swift} events.
1517:
1518: No clear jet break is observed over the course of our observations, in
1519: either the optical bands or the X-ray, out to at least 6 $\times$
1520: $10^5$~s. There is a possible monochromatic break in the radio
1521: bands at around 8~days ($7 \times 10^5$~s), but it appears likely
1522: to be a scintillation artifact (see \S 4.6).
1523:
1524: Using this limit, and following \cite{Sari+1999}, for a uniform
1525: circumburst medium we can calculate the minimum jet opening angle and
1526: minimum collimation-corrected energy. Using standard values for the
1527: radiative efficiency ($\eta = 0.5$) and circumburst density ($n = 3.0$
1528: cm$^{-3}$) (the end result is nearly insensitive to these parameters),
1529: we have
1530:
1531: \begin{equation}
1532: \theta_{\rm jet} = 6.5^{\circ} (\frac{t_{\rm jet}}{\rm d})^{3/8}
1533: (\frac{n}{3 {\rm ~cm^{-3}}})^{1/8} (\frac{1+z}{2})^{-3/8} (\frac{E_{\rm
1534: iso}/\eta}{10^{53} {\rm ~erg}})^{-1/8}.
1535: \end{equation}
1536:
1537: \noindent
1538: However, as we discuss later, the late-time afterglow behavior in
1539: this case favors a wind model. Thus, following \citet{LiChevalier2003}
1540: we have
1541:
1542: \begin{equation}
1543: \theta_{\rm jet} = 5.4^{\circ} (\frac{t_{\rm jet}}{\rm d})^{1/4}
1544: (A_*)^{1/4} (\frac{1+z}{2})^{-1/4} (\frac{E_{\rm iso}/\eta}{10^{53}
1545: {\rm ~erg}})^{-1/4}
1546: \end{equation}
1547:
1548: The upper limit on $t_{\rm jet}$ of 7~days gives a limit on the
1549: opening of at least 3.1 $(A_*/0.1)^{1/4}$ deg. (As discussed
1550: later in \S 5.4, we estimate $A_* \approx 0.1$ from the broadband
1551: spectrum.) Therefore the collimation-corrected energy is at least
1552: $E_{\gamma}$ $\gtrsim$ 2 $\times 10^{50}$ $(A_*/0.1)^{1/2}$ erg.
1553:
1554: It is also possible that the jet break is hidden by the
1555: complicated evolution of the burst, including the rebrightening, which
1556: would imply more modest energetics for this burst. However, as the
1557: late-time slope is still relatively shallow ($\alpha = 1.72$;
1558: generally we expect $\alpha \geq 2$ after a jet break) we consider
1559: this relatively unlikely.
1560:
1561:
1562: \section {Discussion}
1563: % S 5
1564:
1565: \subsection{Initial Power-Law Decline}
1566: % S 5.1
1567:
1568: We first turn our attention to the rapidly declining power law. The
1569: temporal behavior of this feature is quite simple, with a decay constant
1570: $\alpha$ = 1.466 $\pm$ 0.006 and no evident substructure before the
1571: ``bump'' or after it. There is no evidence of a rising component or
1572: any early break. The observed spectral index $\beta$ = 0.62 $\pm$
1573: 0.33, although if the extinction measured at late times is also present
1574: at early times (as we expect), the intrinsic index is actually shallower;
1575: correcting this using our preferred extinction model, we derive $\beta$
1576: = 0.29 $\pm$ 0.49.
1577:
1578: Especially when the decay is observed to flatten later, very
1579: early-time decay of this nature is often interpreted as a reverse
1580: shock. This seems possible --- the spectral and temporal indices are
1581: within the range of predictions for reverse-shock models
1582: (specifically, the thick-shell case of \citealt{Kobayashi2000}).
1583: However, a forward-shock origin is also consistent. Examining the
1584: standard closure relations between $\alpha$ and $\beta$ (as in, for
1585: example, \citealt{Price+2003}), all environment models (ISM, wind, and
1586: jet) are consistent with the constraints derived from the data,
1587: largely because the early-time constraint on $\beta$ is poor. (We
1588: discuss the forward vs.~reverse-shock models for this emission
1589: again in \S 5.3, in connection with the late-time rebrightening.)
1590:
1591: \subsection{The Bump: Internal Shock Origin Without a Prompt-Emission Connection}
1592: % S 5.2
1593:
1594: The bump feature is of considerable interest, since it is nearly
1595: simultaneous with a prompt-emission pulse. However, as discussed
1596: earlier, the temporal analysis seems to disfavor the interpretation as
1597: a prompt reverberation: the bump seems to be already rising even
1598: before the prompt spike.
1599:
1600: Another possible explanation for the origin of this feature is a large
1601: density variation in the surrounding medium (a large clump or other
1602: discrete physical feature in the path of the expanding shock). The
1603: observed pulse width $\Delta t / t \approx 1$ is consistent with a
1604: density variation, and the general appearance of the light curve over
1605: this region is reminiscent of simulations of a GRB forward shock
1606: intersecting ISM density enhancements (e.g., Figure 3 of
1607: \citealt{Nakar+2003}). However, our observation of possible color
1608: change across the bump would (if real) disfavor this hypothesis, at least in
1609: the simplest models: density variation will not change the intrinsic
1610: spectrum, unless either the microphysical parameters or cooling
1611: frequency suddenly and significantly change. We consider this
1612: unlikely, although some authors \citepeg{Yost+2003,Granot+2006} have
1613: discussed the role of variable microphysics in previous GRB
1614: afterglows.
1615:
1616: Alternatively, the observation that the fast-declining component seems
1617: completely unaffected by the afterglow (the temporal indices before
1618: and after are effectively identical) leads us to interpret the bump
1619: as originating from a distinct emission episode --- given the rapid
1620: rise and fall and the hint of blue-to-red evolution we suggest that it
1621: arises from internal-shock emission. Hard-to-soft evolution and an
1622: underlying power-law decay not affected by the flare have also been
1623: seen in X-ray flares \citep{ButlerKocevski2007a, Chincarini+2007}. We
1624: also note that earlier studies of GRB prompt emission have shown pulses
1625: observed at lower energy to be broader than those at higher energy
1626: \citep{Fenimore+1995}; this trend may continue into the optical band.
1627: The broader, smoother profile of this pulse relative to the much
1628: faster-evolving X-ray flares may in this case illustrate important
1629: attributes of the emission --- either from viewing effects or resulting
1630: from the physics of the emission itself.
1631:
1632: \subsection{The Late Rebrightening}
1633: % S 5.3
1634:
1635: The rebrightening phase of this burst is quite dramatic. While our
1636: observations do not sample the peak of the emission, a fit with a
1637: reasonable assumption of the sharpness parameter suggests that the
1638: flux increased by approximately 1 mag, and the amount of
1639: integrated optical flux released during the rebrightening is
1640: comparable to or more than that emitted by the early afterglow. A
1641: rise in optical flux of more than a magnitude at intermediate times
1642: (well after the end of prompt emission, but before any supernova
1643: component) has to our knowledge been seen in only a handful of
1644: previous cases: GRBs 970508 \citep{CastroTirado+1998}, 041219A
1645: \citep{Blake+2005}, 060729 \citep{Grupe+2007}, 070420 \citep{GCN6338},
1646: and 070311 \citep{Guidorzi+2007}.
1647:
1648: The rebrightening is also notable because it appears to differ subtly
1649: from the early decay, even though the evolution of both curves is
1650: generally quite simple. The decay index and spectral index both
1651: steepen, by $\Delta\alpha$ = 0.25 $\pm$ 0.14 and $\Delta\beta$ = 0.80
1652: $\pm$ 0.30, respectively. Assuming a synchrotron spectrum, there are
1653: only two possible origins for this --- the optical band is in
1654: different synchrotron regimes at different times (specifically, $\nu <
1655: \nu_c$ before cooling, and $\nu > \nu_c$ after cooling, consistent
1656: with the changes observed), or because of a shift in the electron
1657: index $p$ by approximately $\Delta p$ = +0.4.
1658:
1659: We consider several physical origins for the rebrightening feature:
1660: the appearance of the forward shock when the burst ejecta first
1661: decelerate against the ISM, the late-time peak of a pre-existing
1662: forward shock due to evolution of the critical frequencies, impact of
1663: the forward shock through a density variation, and rebrightening
1664: caused by a refreshed shock.
1665:
1666: \emph{Appearance of forward shock} --- When the GRB ejecta first
1667: begin to sweep up an amount of matter from the ISM
1668: comparable to the energy in the ejecta, they begin to decelerate, and
1669: reverse and forward shocks are propagated back into the ejecta and
1670: forward into the ISM, respectively; depending on the Lorentz factor,
1671: both shocks can then rise very quickly. We consider this scenario
1672: extremely unlikely to be relevant, since by necessity the forward and
1673: reverse shocks must rise simultaneously, and there is no explanation
1674: for the bright early-time component in the burst --- save for a prompt
1675: model connected with internal shocks, but as we have already shown,
1676: there is no evidence linking the early optical behavior with the
1677: high-energy emission.
1678:
1679: \emph{Spectral peak of existing forward shock} --- A more reasonable
1680: model postulates that the reverse and forward shocks both formed
1681: extremely early, but because they evolve differently (the reverse
1682: shock, whose synchrotron parameters are boosted down by factors of
1683: $\gamma^2$, begins to fade immediately, while the forward shock will
1684: rise at lower frequencies), the reverse shock fades rapidly, while the
1685: forward shock can rise and peak when the synchrotron frequency $\nu_m$
1686: passes through the optical band. This model has, for example, been
1687: invoked to explain early-time bumps in the light curves of GRB~021004
1688: \citepeg{Kobayashi+2003}, GRB~050525A \citep{ShaoDai2005}, and
1689: GRB~080319B \citep{Bloom+2008}, which level off significantly (but do
1690: not rebrighten) at around $10^4$ s. However, this model is
1691: problematic here: although we have only sparse observations of
1692: the rebrightening, the observed rising temporal index of $\alpha$ =
1693: $-$1.12 $\pm$ 0.16 is far too fast to be consistent with a rising
1694: phase of a forward adiabatic shock, which predicts $F \propto
1695: t^{(2-s)/(4-s)}$ (= $t^{1/2}$ for a constant-density ISM and constant
1696: for a wind). Therefore, the synchrotron peak of the forward shock
1697: alone cannot explain this feature.
1698:
1699: \emph{Density variation} --- A third possibility, not invoking the
1700: transition between reverse and forward shocks, might be a dramatic
1701: density variation: for example, the impact of the shock wave into a
1702: previously ejected circumstellar shell, or emergence of the shock from
1703: a low-density cavity into a dense external medium. Density fluctuations
1704: have been successfully invoked to explain low-level variations in
1705: several previous studies \citepeg{Lazzati+2002} and the timescale of
1706: the rebrightening ($\Delta t/t$ $\approx$ 1) is consistent with a
1707: density-fluctuation origin \citep{Nakar+2003}. However, in this case
1708: we would expect neither a change in the spectral index (as is probably
1709: observed) nor such a slow decline after the peak, with a temporal
1710: index that differs significantly but only slightly from the value of
1711: the initial decay. Furthermore, detailed numerical studies by
1712: numerous authors \citep{Huang+2006,Nakar+2003,Nakar+2007} have failed
1713: to reproduce anything but the smallest rebrightening signatures in
1714: previous GRBs using density variations.
1715:
1716: \emph{Multi-component jet} --- The complicated light curve of GRB~030329
1717: has been interpreted \citep{Berger+2003} as the result of two
1718: separate forward shocks, arising from two different jet components: a
1719: narrow, highly relativistic jet whose emission peaks extremely early,
1720: plus a wide, more mildly relativistic jet that dominates the late-time
1721: and radio evolution. Could this model conceivably explain the
1722: observations of GRB~071003? While a complete analysis is beyond the
1723: scope of this paper, we note that the observations do seem consistent:
1724: the similarities of late-time decay of both rapid and late-time
1725: components are naturally explained, the timescale of our rebrightening
1726: is similar to that observed in GRB~030329, and (notably) the most
1727: significant criticism of the two-jet interpretation of GRB~030329
1728: (that the rebrightening rose too rapidly and peaked too sharply ---
1729: \citealt{Huang+2006}) does not apply here: the rebrightening in this
1730: case is much smoother than that observed for GRB 030329.
1731:
1732: \emph{Refreshed shock} --- Finally, we consider the possibility that
1733: this feature is due to a discrete energy reinjection energizing the
1734: forward shock, such as via a slow-moving shell that catches up to the
1735: forward shock at late times after it decelerates. This seems
1736: consistent with all observations, although largely by virtue of not
1737: making strong predictions; by invoking a customized pattern of energy
1738: reinjection at the right times, a very broad space of light curve
1739: behavior can be modeled \citep{Huang+2006}. We do note that a
1740: large, sudden rebrightening of this nature may also produce a (second)
1741: reverse shock, which would be observable in radio and decline rapidly
1742: with time. The radio flux does in fact decay somewhat (in contrast to the
1743: expectation from a forward-shock model, where the radio flux is constant
1744: or rising), and the measured $\alpha = 0.33 \pm 0.10$ is not far from the
1745: predicted decay constant for a reverse shock of $\alpha \approx 1/2$ in
1746: the $\nu < \nu_m$ frequency regime \citep{Kobayashi2000}. However, the
1747: radio decay could conceivably be due to other effects (e.g., late jet break),
1748: and without an independent measurement of the synchrotron peak frequency $\nu_m$
1749: and late-time Lorentz factor $\Gamma$ we are unable to further constrain the
1750: presence or absence of such a feature with the limited observations available.
1751:
1752: We therefore find that only the multi-component jet and refreshed
1753: shock models are consistent with all available data. Unfortunately,
1754: we do not have sufficient observations during the rising phase of the
1755: rebrightening to distinguish the two models; in particular, we can set
1756: no constraints on the color evolution and lack a detailed light curve
1757: of the rise to peak of the rebrightening. We do note that the X-ray
1758: observations are already decaying well before the (probable) optical
1759: peak by an extrapolation of our observations (Figure \ref{fig:lc}),
1760: which may suggest hard-to-soft evolution in this feature as well.
1761: However, as noted earlier, the X-ray decay extrapolates back to
1762: the BAT light curve without explicit need for a rebrightening, so
1763: without earlier X-ray measurements this association is speculative.
1764:
1765:
1766: \subsection{Environmental Constraints}
1767: % S 5.4
1768:
1769: In the simplest models, the late-time light curve of any GRB
1770: is fixed by a number of basic parameters: microphysical
1771: parameters $\epsilon_B$ (the fraction of energy in magnetic fields),
1772: $\epsilon_e$ (the fraction of energy in electrons), and $p$ (the electron
1773: energy index); macroscopic parameters $E_{K}$ (the blastwave
1774: energy) and $\theta_j$ (the jet opening angle); and a parameter
1775: quantifying the density of the surrounding medium, $n$ (for a uniform
1776: density) or $A_*$ (for an $r^{-2}$ density profile). Our broadband
1777: observations (spanning from radio to X-rays) should, in principle,
1778: allow us to firmly constrain most of these parameters for GRB~071003 ---
1779: or, more accurately, to its late rebrightening phase, as this
1780: component is dominant at late times.
1781:
1782: The indices $\alpha$ and $\beta$ are both well constrained at late
1783: times in the optical through X-ray bands, thanks to the wide range of
1784: temporal and spatial sampling: $\alpha_{O+X}$ = 1.71 $\pm$ 0.14,
1785: $\beta_{OX}$ = 0.93 $\pm$ 0.03. Two environment models satisfy these
1786: constraints within 90\% confidence: a wind-driven medium ($\rho
1787: \propto r^{-2}$) in which $p \approx 2.9$, and a model in which the
1788: jet break has already occurred with $p \approx 1.9$ (but consistent
1789: with $p = 2$). Notably, ISM models are a poor fit: the late-time
1790: decay rate is too fast for the shallow spectral index. The radio
1791: observations appear to support this conclusion: the rising light curve
1792: predicted by the ISM model is clearly ruled out, and while the slow
1793: radio decay ($\alpha_R = 0.33 \pm 0.1$) is inconsistent in detail with
1794: the wind prediction of constant evolution as well, it is conceivable
1795: that variations from an exact $s=-2$ profile, an additional source of
1796: radio emission at early times (e.g., a reverse shock), or a soft jet
1797: break at $t \approx 5$~days may explain this difficulty.
1798:
1799: The apparent spectral index of $\beta \approx -1.1$ observed in the
1800: radio is notable. A synchrotron spectrum is expected to have a
1801: self-absorbed $\beta = -2$ spectrum below the self-absorption
1802: frequency $\nu_a$ and a spectrum of $\beta = -0.5$ above it. The fact
1803: that the observed spectral index is intermediate between these values
1804: and consistent with neither (to $\sim$90\% confidence) tells us
1805: that, if the spectrum is really synchrotron, the absorption break is
1806: likely to be very close to these frequencies, although exact constraints
1807: are difficult with only two frequencies since the break is likely to be
1808: quite soft. The radio evolution appears nearly achromatic, which
1809: would argue against this interpretation, but considering the
1810: relatively narrow time and frequency window of the observations and
1811: unknown break sharpness, we feel that this is not a major concern.
1812:
1813: Because the ISM model is notably discrepant with the measured values
1814: of $\alpha$ and $\beta$, we unfortunately cannot use the afterglow as
1815: a probe of the ambient density. If the wind model, which is more
1816: consistent with the observations in this case, is correct, we can
1817: calculate the parameter $A_*$ using (for example) equation 2 in
1818: \cite{Chevalier+1999}:
1819:
1820: \begin{equation}
1821: F_{\nu_m} = {\rm 20~ mJy} (\frac{d_L}{\rm 5403~Mpc}) (1+z)^{1/2}
1822: (\frac{\epsilon_B}{0.1})^{-1/2} E_{52}^{1/2} A_{*} t_{\rm d}^{1/2}.
1823: \end{equation}
1824:
1825: \noindent
1826: While we have no direct measurement of $F_{\nu_m}$, it is constrained
1827: by the radio and optical observations (see Figure \ref{fig:bbsed}) to
1828: be $\sim$1 mJy (within a factor of $\sim$3). We therefore
1829: measure $A_{*} = 0.07 (\epsilon_B/0.1)^{1/2}$, an
1830: interestingly low value regardless of the value of $\epsilon_B$.
1831: While $\epsilon_B$ is not strongly constrained, the absence of a
1832: cooling break between the X-ray and optical bands during the first
1833: 5 days (the cooling frequency $\nu_c$ increases in a wind model)
1834: requires $\epsilon_B \gtrsim 0.3$.
1835:
1836: It is possible that the wind model is inappropriate and the rapid
1837: optical decay is due to a jet that broke before our multicolor
1838: late-time observations. (One possible criticism of the wind model is
1839: that in this case, the color transition between early and late times
1840: is hard to explain; because the cooling frequency rises with time, if
1841: $\nu > \nu_c$ late it must have been early as well under standard
1842: synchrotron evolution. However, because the rebrightening appears to
1843: be either a separate phenomenon or a large energy impulse that could
1844: conceivably have ``reset'' the synchrotron parameters [including
1845: $\nu_c$] to new values, this may not be a major concern.) No jet break
1846: is observed in the light curve, but it is possible that a jet
1847: signature was concealed by the rebrightening. This case would
1848: certainly rule out the wide-angle jet interpretation of the secondary
1849: peak and would significantly reduce the energetics.
1850:
1851:
1852: \subsection{Spectral Implications on the Environment and Host Galaxy}
1853: % S 5.5
1854:
1855: The late-time spectroscopy and imaging tell a coherent story: unlike
1856: the vast majority of GRBs \citep{Wainwright+2007,pcw+08},
1857: GRB~071003 did not occur in a gas-rich\footnote{Since our measurement
1858: is based on magnesium, we are directly measuring the metal column,
1859: not the gas column. An alternate possibility, therefore, is that the
1860: host is ``normal'' but extraordinarily metal-poor, less than
1861: $10^{-2}$ of the average solar abudance. However, we consider a highly
1862: subluminous host a more likely possibility. Both effects may be
1863: in play: low-luminosity galaxies, and those with low equivalent widths,
1864: tend to be relatively metal-poor \citep{Prochaska+2008}.}
1865: galaxy. The environment is
1866: more consistent with a progenitor located in an outer galactic halo,
1867: or in an extremely small (even compared to ``normal'' long-duration GRB hosts) and
1868: gas-poor galaxy. While the possibility of line saturation prevents us
1869: from setting definitive upper limits, the column density through any
1870: host is consistent with being 3 orders of magnitude below typical
1871: GRB-derived values, and the contrast to the overall GRB population -
1872: which is dominated by subluminous galaxies to begin with \citepeg{Fruchter+2006,Fynbo+2008},
1873: is dramatic.
1874:
1875: While it is well established that long-duration GRBs generally
1876: originate from massive stars, we should be careful to ensure that our
1877: prior experience does not blind us to the existence of rarer
1878: subclasses of events. We note that one other GRB on
1879: record, GRB~070125, had very similar properties: extremely low
1880: \ion{Mg}{2} absorption and no coincident host \citep{Cenko+2008}, as
1881: well as a very bright afterglow and extreme energetics ($E_{\gamma}$ =
1882: $3 \times 10^{52}$ erg; \citealt{Chandra+2007}), and even a (mild) late-time
1883: rebrightening \citep{Updike+2008}. Both are also among
1884: the few \textit{Swift} bursts detected at radio wavelengths.
1885:
1886: However, GRB~070125 and GRB~071003 show evidence from their
1887: broadband light curves of origins typical of ordinary long GRBs. In
1888: the case of GRB~070125, a constant but very high circumstellar density
1889: suggested that it occurred in what was locally a dense environment,
1890: not an empty galactic halo, despite the near absence of a large-scale
1891: gas signature in the spectrum. In our case, for GRB~071003, we find
1892: evidence of a wind-like stratified environment, a characteristic of a
1893: massive star. Together, these events appear to suggest an origin for
1894: these ``halo'' bursts similar to those of all other GRBs.
1895:
1896: If GRB~071003 did occur in a star-forming region, then there are two
1897: possibilities consistent with the extremely small metal absorption in
1898: the spectrum. First, the burst may simply have formed in an extremely
1899: subluminous galaxy --- necessarily, the number or distribution of such
1900: objects at very high redshift is not observationally constrained, but
1901: most simulations predict an abundance of small, highly sub-Galactic
1902: halos in the universe that could very well harbor limited star
1903: formation. Alternatively, GRB~071003 may have occurred in a tidally
1904: stripped tail from another, larger galaxy. In this case, further
1905: follow-up observations should reveal a disturbed, star-forming host in
1906: the close vicinity of the burst.
1907:
1908: Either scenario seems plausible to explain the constraints derived on
1909: the burst environment. In either case, if GRBs are shown to be
1910: reasonable tracers of star formation at high redshift, then future
1911: large-sample GRB spectroscopy missions may be able to place important
1912: constraints on the star-formation history of the universe not possible
1913: by any other means. While the sample size of such low-column-density GRBs
1914: is now small (two events, with
1915: GRB~061021 [\citealt{Thoene06_GCN5747}] possibly constituting a third
1916: example), these results are already suggestive that this
1917: fraction may be significant (on the order of a few percent), and
1918: systematic rapid afterglow spectroscopy should continue to increase
1919: the number considerably over the years and decades to come. It would
1920: be an interesting discovery if the distribution of \ion{Mg}{2}
1921: equivalent widths turns out to be bimodal.
1922:
1923: On a related note, the existence of GRB~071003 and GRB~070125 may have
1924: important implications regarding the escape fraction of ionizing photons
1925: and the reionization history of the universe. Although the relatively
1926: low redshift of these systems keeps the Lyman-$\alpha$ and Lyman-break
1927: absorption features out of our spectral range and prevents us from
1928: measuring the H~I column density directly \citep{Chen+2007}, these GRBs
1929: provide evidence that massive stars can form well outside of gas-dense
1930: hosts, where there is little to shield the intergalactic medium from
1931: their ionizing UV radiation. If the fraction of these events is more
1932: than a few percent at $z > 7$, then such ``halo'' stars may in fact be
1933: primarily responsible for the reionization of the universe.
1934: Observationally, spectroscopy of such events at these high redshifts
1935: may allow accurate measurement of the neutral gas fraction
1936: $\bar{x}_{\rm H}$ \citepeg{McQuinn+2007} without the interference of
1937: saturated line profiles originating from the host galaxy.
1938:
1939: \section {Conclusions}
1940: % S 6
1941:
1942: Although the temporal evolution of the optical afterglow of GRB 071003
1943: is complicated, our early through late-time photometric follow-up data
1944: clearly resolve the optical light curve into separate components.
1945: Observations from KAIT during the prompt phase of the GRB revealed a
1946: slowly rising, slowly falling bump or flare component, superimposed
1947: on a simple fading power law that has no observable correlation
1948: with the prompt emission, suggesting that while early internal-shock
1949: flares can be observed in the optical, they are not necessarily the
1950: same as those producing the high-energy signatures. Our late-time
1951: observations revealed one of the most dramatic late rebrightenings
1952: ever recorded in a GRB light curve, and suggest that this feature is
1953: not due to a reverse-forward shock transition or density variation,
1954: requiring either angular jet structure or very discrete late-time
1955: re-energizing of the optical afterglow. This may have important
1956: implications for the interpretation of other, less dramatic bumps and
1957: rebrightenings at similar timescales that appear to be common features
1958: in GRB afterglows.
1959:
1960: The spectroscopic study of GRB 071003 offers a cautionary tale about
1961: the standard use of \ion{Mg}{2} to infer a redshift: while it is
1962: common practice to use the highest-redshift \ion{Mg}{2} system
1963: observed (especially in the cases when the absorption is quite strong)
1964: under the assumption that the GRB host system should show significant
1965: metal absorption, here we have a clear case where this assumption is
1966: fundamentally flawed. Were the S/N of the spectrum worse, or the
1967: host-galaxy absorption even weaker by a factor of only 2--3,
1968: it is likely that we would have missed the higher-redshift system
1969: entirely and proceeded with the assumption that this burst was at a
1970: redshift of 1.100 instead of 1.604. In light of this fact, previous
1971: and future GRB redshift claims based solely on identification of \ion{Mg}{2}
1972: absorption should be regarded with increased skepticism.
1973:
1974: The intervening absorption systems are nevertheless also remarkable.
1975: With three completely independent \ion{Mg}{2} systems along the line
1976: of sight, GRB~071003 is among the most dramatic examples yet of the
1977: bizarre overabundance of these systems in GRB afterglows relative to
1978: those of quasars. Further study of this sightline, especially using
1979: AO systems, may help shed light on this mysterious result.
1980:
1981:
1982:
1983: \acknowledgements
1984:
1985: KAIT and its ongoing operation were made possible by donations from
1986: Sun Microsystems, Inc., the Hewlett-Packard Company, AutoScope
1987: Corporation, Lick Observatory, the National Science Foundation, the
1988: University of California, the Sylvia \& Jim Katzman Foundation, and
1989: the TABASGO Foundation. J.S.B.'s group is supported in part by the
1990: Hellman Faculty Fund, Las Cumbres Observatory Global Telescope
1991: Network, and NASA/\textit{Swift} Guest Investigator grant NNG05GF55G.
1992: A.V.F.'s group is supported by NSF grant AST--0607485 and the TABASGO
1993: Foundation, as well as by NASA/\textit{Swift} Guest Investigator
1994: grants NNG05GF35G and NNG06GI86G. N.R.B. is partially supported by a
1995: SciDAC grant from the Department of Energy. J.X.P. is partially
1996: supported by NASA/\textit{Swift} Guest Investigator grant NNG05GF55G
1997: and NSF CAREER grant AST--0548180. H. Swan has been supported by NSF
1998: grant AST-0335588 and by the Michigan Space Grant Consortium. F. Yuan
1999: has been supported under NASA/Swift Guest Investigator Grant NNX-07AF02G.
2000: We acknowledge helpful discussions
2001: with E. Ramirez-Ruiz and thank D. Whalen and A. Heger for their
2002: calculations of the photon flux from massive stars. We offer
2003: particular thanks to D. A. Kann for useful discussions and feedback.
2004:
2005: This research is based in part on observations obtained at the Gemini
2006: Observatory, which is operated by the Association of Universities for
2007: Research in Astronomy, Inc., under a cooperative agreement with the
2008: NSF on behalf of the Gemini partnership. Some of the data presented
2009: herein were obtained at the W.\ M.\ Keck Observatory, which is
2010: operated as a scientific partnership among the California Institute of
2011: Technology, the University of California, and NASA; the Observatory
2012: was made possible by the generous financial support of the W.\ M.\
2013: Keck Foundation. We wish to extend special thanks to those of Hawaiian
2014: ancestry on whose sacred mountain we are privileged to be guests.
2015: We are grateful to the staffs at the Gemini, Keck, and Lick
2016: Observatories for their assistance.
2017:
2018:
2019: \bibliographystyle{apj}
2020: \begin{thebibliography}{104}
2021: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
2022:
2023: \bibitem[{{Akerlof} {et~al.}(1999){Akerlof}, {Balsano}, {Barthelmy}, {Bloch},
2024: {Butterworth}, {Casperson}, {Cline}, {Fletcher}, {Frontera}, {Gisler},
2025: {Heise}, {Hills}, {Kehoe}, {Lee}, {Marshall}, {McKay}, {Miller}, {Piro},
2026: {Priedhorsky}, {Szymanski}, \& {Wren}}]{Akerlof+1999}
2027: {Akerlof}, C., et al. 1999, \nat, 398, 400
2028:
2029: \bibitem[{{Alard} \& {Lupton}(1998)}]{Alard+1998}
2030: {Alard}, C., \& {Lupton}, R.~H. 1998, \apj, 503, 325
2031:
2032: \bibitem[{{Barris} {et~al.}(2005){Barris}, {Tonry}, {Novicki}, \&
2033: {Wood-Vasey}}]{Barris+2005}
2034: {Barris}, B.~J., {Tonry}, J.~L., {Novicki}, M.~C., \& {Wood-Vasey}, W.~M. 2005,
2035: \aj, 130, 2272
2036:
2037: \bibitem[{{Berger} {et~al.}(2003){Berger}, {Kulkarni}, {Pooley}, {Frail},
2038: {McIntyre}, {Wark}, {Sari}, {Soderberg}, {Fox}, {Yost}, \&
2039: {Price}}]{Berger+2003}
2040: {Berger}, E., et al. 2003, \nat, 426, 154
2041:
2042: \bibitem[{{Bertin} \& {Arnouts}(1996)}]{Bertin+1996}
2043: {Bertin}, E., \& {Arnouts}, S. 1996, \aaps, 117, 393
2044:
2045: \bibitem[{{Beuermann} {et~al.}(1999){Beuermann}, {Hessman}, {Reinsch},
2046: {Nicklas}, {Vreeswijk}, {Galama}, {Rol}, {van Paradijs}, {Kouveliotou},
2047: {Frontera}, {Masetti}, {Palazzi}, \& {Pian}}]{Beuermann+1999}
2048: {Beuermann}, K., et al. 1999, \aap, 352,
2049: L26
2050:
2051: \bibitem[{{Blake} {et~al.}(2005){Blake}, {Bloom}, {Starr}, {Falco},
2052: {Skrutskie}, {Fenimore}, {Duch{\^e}ne}, {Szentgyorgyi}, {Hornstein},
2053: {Prochaska}, {McCabe}, {Ghez}, {Konopacky}, {Stapelfeldt}, {Hurley},
2054: {Campbell}, {Kassis}, {Chaffee}, {Gehrels}, {Barthelmy}, {Cummings},
2055: {Hullinger}, {Krimm}, {Markwardt}, {Palmer}, {Parsons}, {McLean}, \&
2056: {Tueller}}]{Blake+2005}
2057: {Blake}, C.~H., et al. 2005, \nat, 435, 181
2058:
2059: \bibitem[{{Bloom} {et~al.}(2002){Bloom}, {Kulkarni}, \&
2060: {Djorgovski}}]{Bloom+2002}
2061: {Bloom}, J.~S., {Kulkarni}, S.~R., \& {Djorgovski}, S.~G. 2002, \aj, 123, 1111
2062:
2063: \bibitem[{{Bloom} {et~al.}(2008){Bloom}, {Perley}, {Li}, {Butler}, {Miller},
2064: {Kocevski}, {Kann}, {Foley}, {Chen}, {Filippenko}, {Starr}, {Macomber},
2065: {Prochaska}, {Chornock}, {Poznanski}, \& {Klose}}]{Bloom+2008}
2066: {Bloom}, J.~S., et al. 2008, \apj in press. (ArXiv e-prints 0803.3215)
2067:
2068: \bibitem[{{Butler} \& {Kocevski}(2007{\natexlab{a}})}]{ButlerKocevski2007a}
2069: {Butler}, N.~R. \& {Kocevski}, D. 2007{\natexlab{a}}, \apj, 663, 407
2070:
2071: \bibitem[{{Butler} \& {Kocevski}(2007{\natexlab{b}})}]{ButlerKocevski2007b}
2072: ---. 2007{\natexlab{b}}, \apj, 668, 400
2073:
2074: \bibitem[{{Butler} {et~al.}(2006){Butler}, {Li}, {Perley}, {Huang}, {Urata},
2075: {Prochaska}, {Bloom}, {Filippenko}, {Foley}, {Kocevski}, {Chen}, {Qiu},
2076: {Kuo}, {Huang}, {Ip}, {Tamagawa}, {Onda}, {Tashiro}, {Makishima},
2077: {Nishihara}, \& {Sarugaku}}]{Butler+2006}
2078: {Butler}, N.~R., et al. 2006, \apj, 652, 1390
2079:
2080: \bibitem[{{Calzetti} {et~al.}(2000){Calzetti}, {Armus}, {Bohlin}, {Kinney},
2081: {Koornneef}, \& {Storchi-Bergmann}}]{Calzetti+2000}
2082: {Calzetti}, D., {Armus}, L., {Bohlin}, R.~C., {Kinney}, A.~L., {Koornneef}, J.,
2083: \& {Storchi-Bergmann}, T. 2000, \apj, 533, 682
2084:
2085: \bibitem[{{Castro-Tirado} {et~al.}(1998){Castro-Tirado}, {Gorosabel},
2086: {Benitez}, {Wolf}, {Fockenbrock}, {Martinez-Gonzalez}, {Kristen}, {Broeils},
2087: {Pedersen}, {Greiner}, {Costa}, {Feroci}, {Piro}, {Frontera}, {Nicastro},
2088: {Palazzi}, {Bartolini}, {Guarnieri}, {Masetti}, {Piccioni}, {Mignoli},
2089: {Wold}, {Lacy}, {Birkle}, {Broadhurst}, {Brandt}, \&
2090: {Lund}}]{CastroTirado+1998}
2091: {Castro-Tirado}, A.~J., et al. 1998, Science,
2092: 279, 1011
2093:
2094: \bibitem[{{Cenko} \& {Fox}(2007)}]{GCN6839}
2095: {Cenko}, S.~B., \& {Fox}, D.~B. 2007, {GCN Circular} 6839
2096:
2097: \bibitem[{{Cenko} {et~al.}(2006){Cenko}, {Fox}, {Moon}, {Harrison}, {Kulkarni},
2098: {Henning}, {Guzman}, {Bonati}, {Smith}, {Thicksten}, {Doyle}, {Petrie},
2099: {Gal-Yam}, {Soderberg}, {Anagnostou}, \& {Laity}}]{cfm+06}
2100: {Cenko}, S.~B., et al. 2006, \pasp, 118, 1396
2101:
2102: \bibitem[{{Cenko} {et~al.}(2008){Cenko}, {Fox}, {Penprase}, {Cucchiara},
2103: {Price}, {Berger}, {Kulkarni}, {Harrison}, {Gal-Yam}, {Ofek}, {Rau},
2104: {Chandra}, {Frail}, {Kasliwal}, {Schmidt}, {Soderberg}, {Cameron}, \&
2105: {Roth}}]{Cenko+2008}
2106: {Cenko}, S.~B., et al. 2008, \apj, 677, 441
2107:
2108: \bibitem[{{Chandra} {et~al.}(2008){Chandra}, {Cenko}, {Frail}, {Chevalier},
2109: {Macquart}, {Kulkarni}, {Bock}, {Bertoldi}, {Kasliwal}, {Fox}, {Price},
2110: {Berger}, {Soderberg}, {Harrison}, {Gal-Yam}, {Ofek}, {Rau}, {Schmidt},
2111: {Cameron}, {Cowie}, {Cowie}, {Dopita}, {Peterson}, \&
2112: {Penprase}}]{Chandra+2007}
2113: {Chandra}, P., et al. 2008, ArXiv e-prints, 0802.2748
2114:
2115: \bibitem[{{Chen} {et~al.}(2007){Chen}, {Prochaska}, \& {Gnedin}}]{Chen+2007}
2116: {Chen}, H.-W., {Prochaska}, J.~X., \& {Gnedin}, N.~Y. 2007, \apjl, 667, L125
2117:
2118: \bibitem[{{Chevalier} \& {Li}(1999)}]{Chevalier+1999}
2119: {Chevalier}, R.~A. \& {Li}, Z.-Y. 1999, \apjl, 520, L29
2120:
2121: \bibitem[{{Chincarini} {et~al.}(2007){Chincarini}, {Moretti}, {Romano},
2122: {Falcone}, {Morris}, {Racusin}, {Campana}, {Guidorzi}, {Tagliaferri},
2123: {Burrows}, {Pagani}, {Stroh}, {Grupe}, {Capalbi}, {Cusumano}, {Gehrels},
2124: {Giommi}, {La Parola}, {Mangano}, {Mineo}, {Nousek}, {O'Brien}, {Page},
2125: {Perri}, {Troja}, {Willingale}, \& {Zhang}}]{Chincarini+2007}
2126: {Chincarini}, G., et al. 2007, ArXiv Astrophysics e-prints, 0702371
2127:
2128: \bibitem[{{D'Elia} {et~al.}(2007){D'Elia}, {Fiore}, {Meurs}, {Chincarini},
2129: {Melandri}, {Norci}, {Pellizza}, {Perna}, {Piranomonte}, {Sbordone},
2130: {Stella}, {Tagliaferri}, {Vergani}, {Ward}, {Angelini}, {Antonelli},
2131: {Burrows}, {Campana}, {Capalbi}, {Cimatti}, {Costa}, {Cusumano}, {Della
2132: Valle}, {Filliatre}, {Fontana}, {Frontera}, {Fugazza}, {Gehrels}, {Giannini},
2133: {Giommi}, {Goldoni}, {Guetta}, {Israel}, {Lazzati}, {Malesani}, {Marconi},
2134: {Mason}, {Mereghetti}, {Mirabel}, {Molinari}, {Moretti}, {Nousek}, {Perri},
2135: {Piro}, {Stratta}, {Testa}, \& {Vietri}}]{DElia+2007}
2136: {D'Elia}, V., et al. 2007, \aap, 467, 629
2137:
2138: \bibitem[{{D'Elia} {et~al.}(2008){D'Elia}, {Fiore}, {Perna}, {Krongold},
2139: {Covino}, {Fugazza}, {Lazzati}, {Nicastro}, {Antonelli}, {Campana},
2140: {Chincarini}, {D'Avanzo}, {Della Valle}, {Goldoni}, {Guetta}, {Guidorzi},
2141: {Meurs}, {Molinari}, {Norci}, {Piranomonte}, {Stella}, {Stratta},
2142: {Tagliaferri}, \& {Ward}}]{DElia+2008}
2143: {D'Elia}, V., et al. 2008, ArXiv e-prints, 0804.2141
2144:
2145: \bibitem[{{Dessauges-Zavadsky} {et~al.}(2006){Dessauges-Zavadsky}, {Chen},
2146: {Prochaska}, {Bloom}, \& {Barth}}]{dcp+06}
2147: {Dessauges-Zavadsky}, M., {Chen}, H.-W., {Prochaska}, J.~X., {Bloom}, J.~S., \&
2148: {Barth}, A.~J. 2006, \apjl, 648, L89
2149:
2150: \bibitem[{{Dickey} \& {Lockman}(1990)}]{Dickey1990}
2151: {Dickey}, J.~M. \& {Lockman}, F.~J. 1990, \araa, 28, 215
2152:
2153: \bibitem[{{Fenimore} {et~al.}(1995){Fenimore}, {in 't Zand}, {Norris},
2154: {Bonnell}, \& {Nemiroff}}]{Fenimore+1995}
2155: {Fenimore}, E.~E., {in 't Zand}, J.~J.~M., {Norris}, J.~P., {Bonnell}, J.~T.,
2156: \& {Nemiroff}, R.~J. 1995, \apjl, 448, L101
2157:
2158: \bibitem[{{Filippenko}(1982)}]{filippenko82}
2159: {Filippenko}, A.~V. 1982, \pasp, 94, 715
2160:
2161: \bibitem[{{Filippenko}(2005)}]{Filippenko2005}
2162: {Filippenko}, A.~V. 2005, in Astronomical Society of the Pacific Conference
2163: Series, Vol. 332, The Fate of the Most Massive Stars, ed. R.~{Humphreys} \&
2164: K.~{Stanek}, 33
2165:
2166: \bibitem[{{Filippenko} {et~al.}(2001){Filippenko}, {Li}, {Treffers}, \&
2167: {Modjaz}}]{Filippenko+2001}
2168: {Filippenko}, A.~V., {Li}, W.~D., {Treffers}, R.~R., \& {Modjaz}, M. 2001, in
2169: Astronomical Society of the Pacific Conference Series, Vol. 246, IAU Colloq.
2170: 183: Small Telescope Astronomy on Global Scales, ed. B.~{Paczy{\'n}ski},
2171: W.-P. {Chen}, \& C.~{Lemme}, 121
2172:
2173: \bibitem[{{Fitzpatrick} \& {Massa}(1990)}]{FitzpatrickMassa1990}
2174: {Fitzpatrick}, E.~L. \& {Massa}, D. 1990, \apjs, 72, 163
2175:
2176: \bibitem[{{Flewelling-Swan} {et~al.}(2006){Flewelling-Swan}, {Akerlof},
2177: {Smith}, {McKay}, {Rykoff}, {Yost}, {Skinner}, {Alday}, \&
2178: {Moore}}]{Swan+2006}
2179: {Flewelling-Swan}, H., et al., \& {Moore}, K.
2180: 2006, in Bulletin of the American Astronomical Society, Vol.~38, Bulletin of
2181: the American Astronomical Society, 134
2182:
2183: \bibitem[{{Fruchter} {et~al.}(2006){Fruchter}, {Levan}, {Strolger},
2184: {Vreeswijk}, {Thorsett}, {Bersier}, {Burud}, {Castro Cer{\'o}n},
2185: {Castro-Tirado}, {Conselice}, {Dahlen}, {Ferguson}, {Fynbo}, {Garnavich},
2186: {Gibbons}, {Gorosabel}, {Gull}, {Hjorth}, {Holland}, {Kouveliotou}, {Levay},
2187: {Livio}, {Metzger}, {Nugent}, {Petro}, {Pian}, {Rhoads}, {Riess}, {Sahu},
2188: {Smette}, {Tanvir}, {Wijers}, \& {Woosley}}]{Fruchter+2006}
2189: {Fruchter}, A.~S., et al., \& {Woosley}, S.~E. 2006, \nat, 441, 463
2190:
2191: \bibitem[{{Fugazza} {et~al.}(2007){Fugazza}, {Fiore}, {D'Elia}, {D'Avanzo},
2192: {Piranomonte}, {Antonelli}, {Chincarini}, {Covino}, {Tagliaferri}, {Valle},
2193: \& {Soto}}]{GCN6851}
2194: {Fugazza}, D., {Fiore}, F., {D'Elia}, V., {D'Avanzo}, P., {Piranomonte}, S.,
2195: {Antonelli}, L.~A., {Chincarini}, G., {Covino}, S., {Tagliaferri}, G.,
2196: {Valle}, M.~D., \& {Soto}, A.~F. 2007, {GCN Circular} 6851
2197:
2198: \bibitem[{{Fynbo} {et~al.}(2008){Fynbo}, {Prochaska}, {Sommer-Larsen},
2199: {Dessauges-Zavadsky}, \& {Moller}}]{Fynbo+2008}
2200: {Fynbo}, J.~P.~U., et al. 2008, ArXiv e-prints, 0801.3273
2201:
2202: \bibitem[{{Garnavich} {et~al.}(2000){Garnavich}, {Loeb}, \&
2203: {Stanek}}]{Garnavich+2000}
2204: {Garnavich}, P.~M., {Loeb}, A., \& {Stanek}, K.~Z. 2000, \apjl, 544, L11
2205:
2206: \bibitem[{{Gehrels}(2007)}]{GCN6760}
2207: {Gehrels}, N. 2007, {GCN Circular} 6760
2208:
2209: \bibitem[{{Gehrels} {et~al.}(2004){Gehrels}, {Chincarini}, {Giommi}, {Mason},
2210: {Nousek}, {Wells}, {White}, {Barthelmy}, {Burrows}, {Cominsky}, {Hurley},
2211: {Marshall}, {M{\'e}sz{\'a}ros}, {Roming}, {Angelini}, {Barbier}, {Belloni},
2212: {Campana}, {Caraveo}, {Chester}, {Citterio}, {Cline}, {Cropper}, {Cummings},
2213: {Dean}, {Feigelson}, {Fenimore}, {Frail}, {Fruchter}, {Garmire}, {Gendreau},
2214: {Ghisellini}, {Greiner}, {Hill}, {Hunsberger}, {Krimm}, {Kulkarni}, {Kumar},
2215: {Lebrun}, {Lloyd-Ronning}, {Markwardt}, {Mattson}, {Mushotzky}, {Norris},
2216: {Osborne}, {Paczy{\'n}ski}, {Palmer}, {Park}, {Parsons}, {Paul}, {Rees},
2217: {Reynolds}, {Rhoads}, {Sasseen}, {Schaefer}, {Short}, {Smale}, {Smith},
2218: {Stella}, {Tagliaferri}, {Takahashi}, {Tashiro}, {Townsley}, {Tueller},
2219: {Turner}, {Vietri}, {Voges}, {Ward}, {Willingale}, {Zerbi}, \&
2220: {Zhang}}]{Gehrels+2004}
2221: {Gehrels}, N., et al. 2004, \apj, 611, 1005
2222:
2223: \bibitem[{{Golenetskii} {et~al.}(2007){Golenetskii}, {Aptekar}, {Mazets},
2224: {Pal'shin}, {Frederiks}, \& {Cline}}]{GCN6849}
2225: {Golenetskii}, S., {Aptekar}, R., {Mazets}, E., {Pal'shin}, V., {Frederiks},
2226: D., \& {Cline}, T. 2007, {GCN Circular} 6849
2227:
2228: \bibitem[{{Granot} \& {Kumar}(2006)}]{Granot+2006}
2229: {Granot}, J., \& {Kumar}, P. 2006, \mnras, 366, L13
2230:
2231: \bibitem[{{Grupe} {et~al.}(2007){Grupe}, {Gronwall}, {Wang}, {Roming},
2232: {Cummings}, {Zhang}, {M{\'e}sz{\'a}ros}, {Trigo}, {O'Brien}, {Page},
2233: {Beardmore}, {Godet}, {vanden Berk}, {Brown}, {Koch}, {Morris}, {Stroh},
2234: {Burrows}, {Nousek}, {McMath Chester}, {Immler}, {Mangano}, {Romano},
2235: {Chincarini}, {Osborne}, {Sakamoto}, \& {Gehrels}}]{Grupe+2007}
2236: {Grupe}, D., et al. 2007, \apj,
2237: 662, 443
2238:
2239: \bibitem[{{Guidorzi} {et~al.}(2007){Guidorzi}, {Vergani}, {Sazonov}, {Covino},
2240: {Malesani}, {Molkov}, {Palazzi}, {Romano}, {Campana}, {Chincarini},
2241: {Fugazza}, {Moretti}, {Tagliaferri}, {Llorente}, {Gorosabel}, {Antonelli},
2242: {Capalbi}, {Cusumano}, {D'Avanzo}, {Mangano}, {Masetti}, {Meurs}, {Mineo},
2243: {Molinari}, {Morris}, {Nicastro}, {Page}, {Perri}, {Sbarufatti}, {Stratta},
2244: {Sunyaev}, {Troja}, \& {Zerbi}}]{Guidorzi+2007}
2245: {Guidorzi}, C., et al. 2007, \aap,
2246: 474, 793
2247:
2248: \bibitem[{{Holland} {et~al.}(2003){Holland}, {Weidinger}, {Fynbo}, {Gorosabel},
2249: {Hjorth}, {Pedersen}, {M{\'e}ndez Alvarez}, {Augusteijn}, {Castro Cer{\'o}n},
2250: {Castro-Tirado}, {Dahle}, {Egholm}, {Jakobsson}, {Jensen}, {Levan},
2251: {M{\o}ller}, {Pedersen}, {Pursimo}, {Ruiz-Lapuente}, \&
2252: {Thomsen}}]{Holland+2003}
2253: {Holland}, S.~T., et al. 2003, \aj, 125, 2291
2254:
2255: \bibitem[{{Huang} {et~al.}(2006){Huang}, {Cheng}, \& {Gao}}]{Huang+2006}
2256: {Huang}, Y.~F., {Cheng}, K.~S., \& {Gao}, T.~T. 2006, \apj, 637, 873
2257:
2258: \bibitem[{{Jel{\'i}nek} {et~al.}(2007){Jel{\'i}nek}, {Tristram}, {Postigo},
2259: {Gorosabel}, {Hale}, {Gilmore}, {Kilmartin}, \& {Castro-Tirado}}]{GCN6338}
2260: {Jel{\'i}nek}, M., {Tristram}, P., {Postigo}, A.~d.~U., {Gorosabel}, J.,
2261: {Hale}, F., {Gilmore}, A., {Kilmartin}, P., \& {Castro-Tirado}, A.~J. 2007,
2262: {GCN Circular} 6338
2263:
2264: \bibitem[{{Jester} {et~al.}(2005){Jester}, {Schneider}, {Richards}, {Green},
2265: {Schmidt}, {Hall}, {Strauss}, {Vanden Berk}, {Stoughton}, {Gunn},
2266: {Brinkmann}, {Kent}, {Smith}, {Tucker}, \& {Yanny}}]{Jester+2005}
2267: {Jester}, S., et al. 2005, \aj, 130, 873
2268:
2269: \bibitem[{{Kobayashi}(2000)}]{Kobayashi2000}
2270: {Kobayashi}, S. 2000, \apj, 545, 807
2271:
2272: \bibitem[{{Kobayashi} \& {Zhang}(2003)}]{Kobayashi+2003}
2273: {Kobayashi}, S., \& {Zhang}, B. 2003, \apjl, 582, L75
2274:
2275: \bibitem[{{Kocevski} {et~al.}(2007){Kocevski}, {Butler}, \&
2276: {Bloom}}]{Kocevski+2007}
2277: {Kocevski}, D., {Butler}, N., \& {Bloom}, J.~S. 2007, \apj, 667, 1024
2278:
2279: \bibitem[{{Laursen} \& {Stanek}(2003)}]{Laursen+2003}
2280: {Laursen}, L.~T. \& {Stanek}, K.~Z. 2003, \apjl, 597, L107
2281:
2282: \bibitem[{{Lazzati} {et~al.}(2002){Lazzati}, {Rossi}, {Covino}, {Ghisellini},
2283: \& {Malesani}}]{Lazzati+2002}
2284: {Lazzati}, D., {Rossi}, E., {Covino}, S., {Ghisellini}, G., \& {Malesani}, D.
2285: 2002, \aap, 396, L5
2286:
2287: \bibitem[{{Le Floc'h} {et~al.}(2003){Le Floc'h}, {Duc}, {Mirabel}, {Sanders},
2288: {Bosch}, {Diaz}, {Donzelli}, {Rodrigues}, {Courvoisier}, {Greiner},
2289: {Mereghetti}, {Melnick}, {Maza}, \& {Minniti}}]{LeFloch+2003}
2290: {Le Floc'h}, E., et al.
2291: 2003, \aap, 400, 499
2292:
2293: \bibitem[{{Li}(2006)}]{GCN4727}
2294: {Li}, W. 2006, {GCN Circular} 4727
2295:
2296: \bibitem[{{Li}(2007)}]{GCN6838}
2297: ---. 2007, {GCN Circular} 6838
2298:
2299: \bibitem[{{Li} {et~al.}(2007){Li}, {Chornock}, {Filippenko}, {Perley}, \&
2300: {Bloom}}]{GCN6844}
2301: {Li}, W., {Chornock}, R., {Filippenko}, A.~V., {Perley}, D.~A., \& {Bloom},
2302: J.~S. 2007, {GCN Circular} 6844
2303:
2304: \bibitem[{{Li} {et~al.}(2003{\natexlab{a}}){Li}, {Filippenko}, {Chornock}, \&
2305: {Jha}}]{Li+2003b}
2306: {Li}, W., {Filippenko}, A.~V., {Chornock}, R., \& {Jha}, S. 2003{\natexlab{a}},
2307: \apjl, 586, L9
2308:
2309: \bibitem[{{Li} {et~al.}(2003{\natexlab{b}}){Li}, {Filippenko}, {Chornock}, \&
2310: {Jha}}]{Li+2003a}
2311: ---. 2003{\natexlab{b}}, \pasp, 115, 844
2312:
2313: \bibitem[{{Li} {et~al.}(2000){Li}, {Filippenko}, {Treffers}, {Friedman},
2314: {Halderson}, {Johnson}, {King}, {Modjaz}, {Papenkova}, {Sato}, \&
2315: {Shefler}}]{Li+2000}
2316: {Li}, W.~D., et al. 2000, in American Institute of Physics
2317: Conference Series, Vol. 522, American Institute of Physics Conference Series,
2318: ed. S.~S. {Holt} \& W.~W. {Zhang}, 103--106
2319:
2320: \bibitem[{{Li} \& {Chevalier}(2003)}]{LiChevalier2003}
2321: {Li}, Z.-Y. \& {Chevalier}, R.~A. 2003, in Lecture Notes in Physics, Berlin
2322: Springer Verlag, Vol. 598, Supernovae and Gamma-Ray Bursters, ed.
2323: K.~{Weiler}, 419--444
2324:
2325: \bibitem[{{McQuinn} {et~al.}(2007){McQuinn}, {Lidz}, {Zaldarriaga},
2326: {Hernquist}, \& {Dutta}}]{McQuinn+2007}
2327: {McQuinn}, M., {Lidz}, A., {Zaldarriaga}, M., {Hernquist}, L., \& {Dutta}, S.
2328: 2007, ArXiv e-prints, 0710.1018
2329:
2330: \bibitem[{{Misra} {et~al.}(2007){Misra}, {Pandey}, {Roy}, \&
2331: {Castro-Tirado}}]{GCN6840}
2332: {Misra}, K., {Pandey}, S.~B., {Roy}, R., \& {Castro-Tirado}, A.~J. 2007, {GCN
2333: Circular} 6840
2334:
2335: \bibitem[{{Monet} {et~al.}(2003){Monet}, {Levine}, {Canzian}, {Ables}, {Bird},
2336: {Dahn}, {Guetter}, {Harris}, {Henden}, {Leggett}, {Levison}, {Luginbuhl},
2337: {Martini}, {Monet}, {Munn}, {Pier}, {Rhodes}, {Riepe}, {Sell}, {Stone},
2338: {Vrba}, {Walker}, {Westerhout}, {Brucato}, {Reid}, {Schoening}, {Hartley},
2339: {Read}, \& {Tritton}}]{Monet+2003}
2340: {Monet}, D.~G., et al., {Schoening}, W., {Hartley}, M., {Read},
2341: M.~A., \& {Tritton}, S.~B. 2003, \aj, 125, 984
2342:
2343: \bibitem[{{Nakar} \& {Granot}(2007)}]{Nakar+2007}
2344: {Nakar}, E., \& {Granot}, J. 2007, \mnras, 380, 1744
2345:
2346: \bibitem[{{Nakar} \& {Piran}(2003)}]{Nakar+2003}
2347: {Nakar}, E., \& {Piran}, T. 2003, \apj, 598, 400
2348:
2349: \bibitem[{{Nestor} {et~al.}(2005){Nestor}, {Turnshek}, \& {Rao}}]{ntr05}
2350: {Nestor}, D.~B., {Turnshek}, D.~A., \& {Rao}, S.~M. 2005, \apj, 628, 637
2351:
2352: \bibitem[{{Nousek} {et~al.}(2006){Nousek}, {Kouveliotou}, {Grupe}, {Page},
2353: {Granot}, {Ramirez-Ruiz}, {Patel}, {Burrows}, {Mangano}, {Barthelmy},
2354: {Beardmore}, {Campana}, {Capalbi}, {Chincarini}, {Cusumano}, {Falcone},
2355: {Gehrels}, {Giommi}, {Goad}, {Godet}, {Hurkett}, {Kennea}, {Moretti},
2356: {O'Brien}, {Osborne}, {Romano}, {Tagliaferri}, \& {Wells}}]{Nousek+2006}
2357: {Nousek}, J.~A., et al. 2006, \apj, 642, 389
2358:
2359: \bibitem[{{Nysewander} {et~al.}(2007){Nysewander}, {Reichart}, {Crain},
2360: {Foster}, {Haislip}, {Ivarsen}, {Lacluyze}, \& {Trotter}}]{Nysewander+2007}
2361: {Nysewander}, M., {Reichart}, D.~E., {Crain}, J.~A., {Foster}, A., {Haislip},
2362: J., {Ivarsen}, K., {Lacluyze}, A., \& {Trotter}, A. 2007, ArXiv e-prints, 0708.3444
2363:
2364: \bibitem[{{Oke} {et~al.}(1995){Oke}, {Cohen}, {Carr}, {Cromer}, {Dingizian},
2365: {Harris}, {Labrecque}, {Lucinio}, {Schaal}, {Epps}, \& {Miller}}]{Oke+1995}
2366: {Oke}, J.~B., {Cohen}, J.~G., {Carr}, M., {Cromer}, J., {Dingizian}, A.,
2367: {Harris}, F.~H., {Labrecque}, S., {Lucinio}, R., {Schaal}, W., {Epps}, H., \&
2368: {Miller}, J. 1995, \pasp, 107, 375
2369:
2370: \bibitem[{{Perley} {et~al.}(2008){Perley}, {Bloom}, {Butler}, {Pollack},
2371: {Holtzman}, {Blake}, {Kocevski}, {Vestrand}, {Li}, {Foley}, {Bellm}, {Chen},
2372: {Prochaska}, {Starr}, {Filippenko}, {Falco}, {Szentgyorgyi}, {Wren},
2373: {Wozniak}, {White}, \& {Pergande}}]{Perley+2008}
2374: {Perley}, D.~A., et al. 2008, \apj, 672, 449
2375:
2376: \bibitem[{{Perley} {et~al.}(2007{\natexlab{a}}){Perley}, {Chornock}, {Bloom},
2377: {Fassnacht}, \& {Auger}}]{GCN6850}
2378: {Perley}, D.~A., {Chornock}, R., {Bloom}, J.~S., {Fassnacht}, C., \& {Auger},
2379: M.~W. 2007{\natexlab{a}}, {GCN Circular} 6850
2380:
2381: \bibitem[{{Perley} {et~al.}(2007{\natexlab{b}}){Perley}, {Prochaska}, {Bloom},
2382: \& {Smith}}]{GCN6843}
2383: {Perley}, D.~A., {Prochaska}, J.~X., {Bloom}, J.~S., \& {Smith}, G.
2384: 2007{\natexlab{b}}, {GCN Circular} 6843
2385:
2386: \bibitem[{{Phillips} {et~al.}(2006){Phillips}, {Miller}, {Cowley}, \&
2387: {Wallace}}]{Phillips+2006}
2388: {Phillips}, A.~C., {Miller}, J., {Cowley}, D., \& {Wallace}, V. 2006, in
2389: Presented at the Society of Photo-Optical Instrumentation Engineers (SPIE)
2390: Conference, Vol. 6269, Ground-based and Airborne Instrumentation for
2391: Astronomy. Edited by McLean, Ian S.; Iye, Masanori. Proceedings of the SPIE,
2392: Volume 6269, pp. 62691O (2006).
2393:
2394: \bibitem[{{Pollack} {et~al.}(2007){Pollack}, {Dam}, {Mignant}, {Johansson},
2395: {Perley}, \& {Bloom}}]{GCN7010}
2396: {Pollack}, L.~K., {Dam}, M.~A.~v., {Mignant}, D.~L., {Johansson}, E.~M.,
2397: {Perley}, D.~A., \& {Bloom}, J.~S. 2007, {GCN Circular} 7010
2398:
2399: \bibitem[{{Price} {et~al.}(2002){Price}, {Berger}, {Reichart}, {Kulkarni},
2400: {Yost}, {Subrahmanyan}, {Wark}, {Wieringa}, {Frail}, {Bailey}, {Boyle},
2401: {Corbett}, {Gunn}, {Ryder}, {Seymour}, {Koviak}, {McCarthy}, {Phillips},
2402: {Axelrod}, {Bloom}, {Djorgovski}, {Fox}, {Galama}, {Harrison}, {Hurley},
2403: {Sari}, {Schmidt}, {Brown}, {Cline}, {Frontera}, {Guidorzi}, \&
2404: {Montanari}}]{Price+2003}
2405: {Price}, P.~A., et al.
2406: 2002, \apjl, 572, L51
2407:
2408: \bibitem[{{Prochaska}(2006)}]{pro06}
2409: {Prochaska}, J.~X. 2006, \apj, 650, 272
2410:
2411: \bibitem[{{Prochaska} {et~al.}(2006){Prochaska}, {Chen}, \& {Bloom}}]{pcb06}
2412: {Prochaska}, J.~X., {Chen}, H.-W., \& {Bloom}, J.~S. 2006, \apj, 648, 95
2413:
2414: \bibitem[{{Prochaska} {et~al.}(2007){Prochaska}, {Chen}, {Dessauges-Zavadsky},
2415: \& {Bloom}}]{Prochaska+2007}
2416: {Prochaska}, J.~X., {Chen}, H.-W., {Dessauges-Zavadsky}, M., \& {Bloom}, J.~S.
2417: 2007, \apj, 666, 267
2418:
2419: \bibitem[{{Prochaska} {et~al.}(2008{\natexlab{a}}){Prochaska}, {Chen}, {Wolfe},
2420: {Dessauges-Zavadsky}, \& {Bloom}}]{pcw+08}
2421: {Prochaska}, J.~X., {Chen}, H.-W., {Wolfe}, A.~M., {Dessauges-Zavadsky}, M., \&
2422: {Bloom}, J.~S. 2008{\natexlab{a}}, \apj, 672, 59
2423:
2424: \bibitem[{{Prochaska} {et~al.}(2008{\natexlab{b}}){Prochaska}, {Chen}, {Wolfe},
2425: {Dessauges-Zavadsky}, \& {Bloom}}]{Prochaska+2008}
2426: ---. 2008{\natexlab{b}}, \apj, 672, 59
2427:
2428: \bibitem[{{Prochter} {et~al.}(2006{\natexlab{a}}){Prochter}, {Prochaska}, \&
2429: {Burles}}]{ppb06}
2430: {Prochter}, G.~E., {Prochaska}, J.~X., \& {Burles}, S.~M. 2006{\natexlab{a}},
2431: \apj, 639, 766
2432:
2433: \bibitem[{{Prochter} {et~al.}(2006{\natexlab{b}}){Prochter}, {Prochaska},
2434: {Chen}, {Bloom}, {Dessauges-Zavadsky}, {Foley}, {Lopez}, {Pettini}, {Dupree},
2435: \& {Guhathakurta}}]{ppc+06}
2436: {Prochter}, G.~E., et al. 2006{\natexlab{b}}, \apjl, 648, L93
2437:
2438: \bibitem[{{Ricker} {et~al.}(2003){Ricker}, {Atteia}, {Crew}, {Doty},
2439: {Fenimore}, {Galassi}, {Graziani}, {Hurley}, {Jernigan}, {Kawai}, {Lamb},
2440: {Matsuoka}, {Pizzichini}, {Shirasaki}, {Tamagawa}, {Vanderspek}, {Vedrenne},
2441: {Villasenor}, {Woosley}, \& {Yoshida}}]{Ricker+2003}
2442: {Ricker}, G.~R., et al. 2003, in American Institute of Physics
2443: Conference Series, Vol. 662, Gamma-Ray Burst and Afterglow Astronomy 2001: A
2444: Workshop Celebrating the First Year of the HETE Mission, ed. G.~R. {Ricker}
2445: \& R.~K. {Vanderspek}, 3--16
2446:
2447: \bibitem[{{Rodgers} {et~al.}(2006){Rodgers}, {Canterna}, {Smith}, {Pierce}, \&
2448: {Tucker}}]{Rodgers+2006}
2449: {Rodgers}, C.~T., {Canterna}, R., {Smith}, J.~A., {Pierce}, M.~J., \& {Tucker},
2450: D.~L. 2006, \aj, 132, 989
2451:
2452: \bibitem[{{Rykoff} {et~al.}(2006){Rykoff}, {Mangano}, {Yost}, {Sari},
2453: {Aharonian}, {Akerlof}, {Ashley}, {Barthelmy}, {Burrows}, {Gehrels},
2454: {G{\"o}{\v g}{\"u}{\c s}}, {G{\"u}ver}, {Horns}, {K{\i}z{\i}lo{\v g}lu},
2455: {Krimm}, {McKay}, {{\"O}zel}, {Phillips}, {Quimby}, {Rowell}, {Rujopakarn},
2456: {Schaefer}, {Smith}, {Swan}, {Vestrand}, {Wheeler}, {Wren}, \&
2457: {Yuan}}]{Rykoff+2006}
2458: {Rykoff}, E.~S., et al. 2006, \apjl, 638, L5
2459:
2460: \bibitem[{{Sari} {et~al.}(1999){Sari}, {Piran}, \& {Halpern}}]{Sari+1999}
2461: {Sari}, R., {Piran}, T., \& {Halpern}, J.~P. 1999, \apjl, 519, L17
2462:
2463: \bibitem[{{Savaglio} {et~al.}(2003){Savaglio}, {Fall}, \& {Fiore}}]{sff03}
2464: {Savaglio}, S., {Fall}, S.~M., \& {Fiore}, F. 2003, \apj, 585, 638
2465:
2466: \bibitem[{{Schady} {et~al.}(2007){Schady}, {Evans}, {Gehrels}, {Guidorzi},
2467: {Holland}, {Kennea}, {Markwardt}, {O'Brien}, {Page}, {Palmer}, {Sato}, \&
2468: {Starling}}]{GCN6837}
2469: {Schady}, P., {Evans}, P.~A., {Gehrels}, N., {Guidorzi}, C., {Holland}, S.~T.,
2470: {Kennea}, J.~A., {Markwardt}, C.~B., {O'Brien}, P.~T., {Page}, K.~L.,
2471: {Palmer}, D.~M., {Sato}, G., \& {Starling}, R.~L.~C. 2007, {GCN Circular}
2472: 6837
2473:
2474: \bibitem[{{Schmidt} {et~al.}(1998){Schmidt}, {Suntzeff}, {Phillips},
2475: {Schommer}, {Clocchiatti}, {Kirshner}, {Garnavich}, {Challis}, {Leibundgut},
2476: {Spyromilio}, {Riess}, {Filippenko}, {Hamuy}, {Smith}, {Hogan}, {Stubbs},
2477: {Diercks}, {Reiss}, {Gilliland}, {Tonry}, {Maza}, {Dressler}, {Walsh}, \&
2478: {Ciardullo}}]{schmidt98}
2479: {Schmidt}, B.~P., et al. 1998, \apj, 507, 46
2480:
2481: \bibitem[{{Shao} \& {Dai}(2005)}]{ShaoDai2005}
2482: {Shao}, L., \& {Dai}, Z.~G. 2005, \apj, 633, 1027
2483:
2484: \bibitem[{{Shih} {et~al.}(2007){Shih}, {Lee}, {Urata}, \& {Huang}}]{GCN6846}
2485: {Shih}, C.~Y., {Lee}, Y.~H., {Urata}, Y., \& {Huang}, K.~Y. 2007, {GCN
2486: Circular} 6846
2487:
2488: \bibitem[{{Stone}(1977)}]{Stone1977}
2489: {Stone}, R.~P.~S. 1977, \apj, 218, 767
2490:
2491: \bibitem[{{Swan} {et~al.}(2007){Swan}, {Smith}, {Rujopakarn}, {Yuan}, {Yost},
2492: {Akerlof}, {Skinner}, \& {Russell}}]{GCN6841}
2493: {Swan}, H., {Smith}, I., {Rujopakarn}, W., {Yuan}, F., {Yost}, S., {Akerlof},
2494: C., {Skinner}, M., \& {Russell}, R. 2007, {GCN Circular} 6841
2495:
2496: \bibitem[{{Thoene} {et~al.}(2006){Thoene}, {Fynbo}, \&
2497: {Jakobsson}}]{Thoene06_GCN5747}
2498: {Thoene}, C.~C., {Fynbo}, J.~P.~U., \& {Jakobsson}, P. 2006, {GCN Circular}
2499: 5747
2500:
2501: \bibitem[{{Updike} {et~al.}(2008){Updike}, {Haislip}, {Nysewander}, {Fruchter},
2502: {Kann}, {Klose}, {Milne}, {Williams}, {Zheng}, {Hergenrother}, {Prochaska},
2503: {Halpern}, {Mirabal}, {Thorstensen}, {van der Horst}, {Starling}, {Racusin},
2504: {Burrows}, {Kuin}, {Roming}, {Bellm}, {Hurley}, {Li}, {Filippenko}, {Blake},
2505: {Starr}, {Falco}, {Brown}, {Dai}, {Deng}, {Xin}, {Qiu}, {Wei}, {Urata},
2506: {Nanni}, {Maiorano}, {Palazzi}, {Greco}, {Bartolini}, {Guarnieri},
2507: {Piccioni}, {Pizzichini}, {Terra}, {Misra}, {Bhatt}, {Anupama}, {Fan},
2508: {Jiang}, {Wijers}, {Reichart}, {Eid}, {Bryngelson}, {Puls}, {Goldthwaite}, \&
2509: {Hartmann}}]{Updike+2008}
2510: {Updike}, A.~C., et al. 2008, ArXiv e-prints, 0805.1094
2511:
2512: \bibitem[{{Van Dam} {et~al.}(2004){Van Dam}, {Le Mignant}, \&
2513: {Macintosh}}]{VanDam+2004}
2514: {Van Dam}, M.~A., {Le Mignant}, D., \& {Macintosh}, B.~A. 2004, Applied Optics,
2515: 43, 5458
2516:
2517: \bibitem[{{Vestrand} {et~al.}(2005){Vestrand}, {Wozniak}, {Wren}, {Fenimore},
2518: {Sakamoto}, {White}, {Casperson}, {Davis}, {Evans}, {Galassi}, {McGowan},
2519: {Schier}, {Asa}, {Barthelmy}, {Cummings}, {Gehrels}, {Hullinger}, {Krimm},
2520: {Markwardt}, {McLean}, {Palmer}, {Parsons}, \& {Tueller}}]{Vestrand+2005}
2521: {Vestrand}, W.~T., et al. 2005, \nat, 435, 178
2522:
2523: \bibitem[{{Vestrand} {et~al.}(2006){Vestrand}, {Wren}, {Wozniak}, {Aptekar},
2524: {Golentskii}, {Pal'Shin}, {Sakamoto}, {White}, {Evans}, {Casperson}, \&
2525: {Fenimore}}]{Vestrand+2006}
2526: {Vestrand}, W.~T., et al., \& {Fenimore}, E. 2006, \nat, 442, 172
2527:
2528: \bibitem[{{Vreeswijk} {et~al.}(2007){Vreeswijk}, {Ledoux}, {Smette}, {Ellison},
2529: {Jaunsen}, {Andersen}, {Fruchter}, {Fynbo}, {Hjorth}, {Kaufer}, {M{\o}ller},
2530: {Petitjean}, {Savaglio}, \& {Wijers}}]{vls+07}
2531: {Vreeswijk}, P.~M., et al. 2007, \aap, 468, 83
2532:
2533: \bibitem[{{Wainwright} {et~al.}(2007){Wainwright}, {Berger}, \&
2534: {Penprase}}]{Wainwright+2007}
2535: {Wainwright}, C., {Berger}, E., \& {Penprase}, B.~E. 2007, \apj, 657, 367
2536:
2537: \bibitem[{{Walker}(1998)}]{Walker1998}
2538: {Walker}, M.~A. 1998, \mnras, 294, 307
2539:
2540: \bibitem[{{Walker}(2001)}]{Walker2001}
2541: ---. 2001, \mnras, 321, 176
2542:
2543: \bibitem[{{Wo{\'z}niak} {et~al.}(2006){Wo{\'z}niak}, {Vestrand}, {Wren},
2544: {White}, {Evans}, \& {Casperson}}]{Wozniak+2006}
2545: {Wo{\'z}niak}, P.~R., {Vestrand}, W.~T., {Wren}, J.~A., {White}, R.~R.,
2546: {Evans}, S.~M., \& {Casperson}, D. 2006, \apjl, 642, L99
2547:
2548: \bibitem[{{Yost} {et~al.}(2003){Yost}, {Harrison}, {Sari}, \&
2549: {Frail}}]{Yost+2003}
2550: {Yost}, S.~A., {Harrison}, F.~A., {Sari}, R., \& {Frail}, D.~A. 2003, \apj,
2551: 597, 459
2552:
2553: \bibitem[{{Yost} {et~al.}(2007){Yost}, {Swan}, {Rykoff}, {Aharonian},
2554: {Akerlof}, {Alday}, {Ashley}, {Barthelmy}, {Burrows}, {Depoy}, {Dufour},
2555: {Eastman}, {Forgey}, {Gehrels}, {G{\"o}{\u g}{\"u}{\c s}}, {G{\"u}ver},
2556: {Halpern}, {Hardin}, {Horns}, {Kiziloglu}, {Krimm}, {Lepine}, {Liang},
2557: {Marshall}, {McKay}, {Mineo}, {Mirabal}, {{\"O}zel}, {Phillips}, {Prieto},
2558: {Quimby}, {Romano}, {Rowell}, {Rujopakarn}, {Schaefer}, {Silverman},
2559: {Siverd}, {Skinner}, {Smith}, {Smith}, {Tonnesen}, {Troja}, {Vestrand},
2560: {Wheeler}, {Wren}, {Yuan}, \& {Zhang}}]{Yost+2007}
2561: {Yost}, S.~A., et al. 2007, \apj, 657, 925
2562:
2563: \bibitem[{{Zhang} {et~al.}(2006){Zhang}, {Fan}, {Dyks}, {Kobayashi},
2564: {M{\'e}sz{\'a}ros}, {Burrows}, {Nousek}, \& {Gehrels}}]{Zhang+2006}
2565: {Zhang}, B., et al. 2006, \apj, 642, 354
2566:
2567: \end{thebibliography}
2568:
2569:
2570:
2571: \clearpage
2572:
2573:
2574: \renewcommand{\arraystretch}{0.60}
2575:
2576: \begin{deluxetable}{rrrrrrrrrrr}
2577: \tabletypesize{\footnotesize}
2578: \tablewidth{0pc}
2579: \tablecaption{Photometry of comparison stars in the field of GRB 071003}
2580: \tablehead{
2581: \colhead{ID}&\colhead{$\alpha_{J2000}$\tablenotemark{a}}&\colhead{$\delta_{J2000}$\tablenotemark{a}}&\colhead{$B$}& \colhead{N$_B$} &
2582: \colhead{$V$}&\colhead{N$_V$} & \colhead{$R$}& \colhead{N$_R$} &
2583: \colhead{$I$} &\colhead{N$_I$}
2584: }
2585: \startdata
2586: 1&301.9066&10.9743&16.760(014)& 4&15.749(012)& 4&15.235(008)& 4&14.720(007)& 4\\
2587: 2&301.9009&10.9357&13.760(009)& 9&12.789(007)& 8&12.251(008)& 9&11.715(009)& 8\\
2588: 3&301.8926&10.9249&14.865(008)& 9&13.243(006)& 8&12.337(008)& 9&11.451(008)& 7\\
2589: 4&301.8903&10.9968&15.686(006)& 6&14.748(009)& 8&14.255(009)& 9&13.818(008)& 8\\
2590: 5&301.8811&10.9343&15.026(009)& 8&13.514(008)& 9&12.679(007)&10&11.882(006)& 7\\
2591: 6&301.8799&10.9803&17.088(009)& 7&16.050(009)& 8&15.499(008)& 8&14.939(010)& 6\\
2592: 7&301.8727&10.9562&17.184(014)& 4&16.372(009)& 8&15.898(008)& 8&15.479(010)& 7\\
2593: 8&301.8705&11.0005&16.124(010)& 5&14.787(012)& 6&14.044(008)& 7&13.353(008)& 6\\
2594: 9&301.8693&10.9253&16.473(009)& 8&15.631(008)&10&15.149(006)&10&14.714(008)& 9\\
2595: 10&301.8688&10.9989&15.139(010)& 5&14.523(009)& 6&14.153(008)& 7&13.804(008)& 6\\
2596: 11&301.8671&10.9417&15.310(009)& 8&14.537(008)&10&14.095(007)&10&13.692(008)&10\\
2597: 12&301.8570&11.0033&15.828(012)& 5&15.004(012)& 4&14.569(009)& 6&14.158(010)& 4\\
2598: 13&301.8467&10.9749&13.470(008)& 7&12.899(008)& 9&12.554(008)&10&12.220(007)& 8\\
2599: 14&301.8396&10.9967&16.468(009)& 7&15.752(008)& 9&15.331(007)&10&14.936(009)& 9\\
2600: 15&301.8386&10.9736&16.778(009)& 8&15.891(009)& 6&15.367(008)& 9&14.842(008)& 7\\
2601: 16&301.8383&10.9800&15.305(010)& 7&14.034(008)& 9&13.357(006)&10&12.773(008)& 9\\
2602: 17&301.8354&11.0043&17.455(017)& 3&16.560(004)& 3&15.970(010)& 5&15.469(008)& 5\\
2603: 18&301.8342&10.9820&16.109(008)& 9&14.992(009)& 7&14.385(007)&10&13.832(010)& 8\\
2604: 19&301.8334&10.9963&16.472(012)& 6&15.430(010)& 7&14.786(008)&10&14.254(008)& 9\\
2605: 20&301.8311&11.0048&16.227(012)& 4&15.529(009)& 3&15.056(007)& 6&14.631(006)& 5\\
2606: 21&301.8279&10.9884&17.045(008)& 7&16.107(008)& 9&15.546(008)&10&15.007(009)& 8\\
2607: 22&301.8247&10.9781&14.276(008)& 9&13.825(008)& 9&13.510(008)&10&13.223(008)& 9\\
2608: 23&301.8221&10.9256&16.606(009)& 9&16.002(012)& 6&15.623(007)& 9&15.204(010)& 7\\
2609: \enddata
2610: \tablecomments{Uncertainties (standard deviation of the mean) are
2611: indicated in parentheses. This table has been truncated; additional standard stars
2612: are available in the online material.}
2613: \tablenotetext{a}{In degrees.}
2614: \label{tab:calib}
2615: \end{deluxetable}
2616:
2617: \begin{deluxetable}{lrccl}
2618: \tablecaption{KAIT photometry of GRB 071003}
2619: \tablehead{
2620: \colhead{$t_{\rm start}$\tablenotemark{a}}&
2621: \colhead{Exp. time}&\colhead{Mag}&
2622: \colhead{Error} & \colhead{Filter} \\
2623: \colhead{(s)}&
2624: \colhead{(s)}&\colhead{}&
2625: \colhead{} & \colhead{}
2626: }
2627: \startdata
2628: 42.0 & 5 & 12.791 & 0.019 & $R$\tablenotemark{b} \\
2629: 49.0 & 5 & 12.999 & 0.024 & $R$ \\
2630: 55.0 & 5 & 13.193 & 0.021 & $R$ \\
2631: 61.0 & 5 & 13.321 & 0.024 & $R$ \\
2632: 67.0 & 5 & 13.500 & 0.019 & $R$ \\
2633: 97.0 & 20 & 14.465 & 0.027 & $V$\\
2634: 128.0 & 20 & 13.919 & 0.032 & $I$\\
2635: 157.0 & 20 & 14.382 & 0.031 & $R$ \\
2636: 188.0 & 20 & 14.916 & 0.034 & $V$\\
2637: 219.0 & 20 & 14.121 & 0.022 & $I$\\
2638: 249.0 & 20 & 14.750 & 0.023 & $R$ \\
2639: 279.0 & 20 & 15.409 & 0.035 & $V$\\
2640: 310.0 & 20 & 14.578 & 0.030 & $I$\\
2641: 340.0 & 20 & 15.401 & 0.024 & $R$ \\
2642: 370.0 & 20 & 16.034 & 0.107 & $V$\\
2643: 401.0 & 20 & 15.478 & 0.063 & $I$\\
2644: 431.0 & 20 & 16.239 & 0.082 & $R$ \\
2645: 462.0 & 20 & 16.853 & 0.200 & $V$\\
2646: 492.0 & 20 & 15.977 & 0.069 & $I$\\
2647: 522.0 & 20 & 16.749 & 0.091 & $R$ \\
2648: 565.0 & 20 & 16.255 & 0.077 & $I$\\
2649: 595.0 & 20 & 16.849 & 0.097 & $R$ \\
2650: 624.0 & 20 & 16.364 & 0.106 & $I$\\
2651: 654.0 & 20 & 17.041 & 0.089 & $R$ \\
2652: 749.09\tablenotemark{c} & 3$\times$20 & 16.830 & 0.113 & $I$\\
2653: 787.68\tablenotemark{c} & 3$\times$20 & 17.362 & 0.121 & $R$ \\
2654: 1007.74\tablenotemark{c} & 6$\times$20 & 17.314 & 0.148 & $I$\\
2655: 1009.95\tablenotemark{c} & 5$\times$20 & 17.711 & 0.147 & $R$ \\
2656: 1422.34\tablenotemark{c} & 8$\times$20 & 18.103 & 0.154 & $R$ \\
2657: 1464.58\tablenotemark{c} & 7$\times$20 & 17.473 & 0.135 & $I$\\
2658: \enddata
2659: \tablenotetext{a}{The start time of the exposure, in seconds after the BAT trigger.}
2660: \tablenotetext{b}{The R-band photometry is derived from unfiltered observations.}
2661: \tablenotetext{c}{The time (s) at the middle point of several combined images.}
2662: \label{tab:kait}
2663: \end{deluxetable}
2664:
2665: \clearpage
2666:
2667: \begin{deluxetable}{lcc}
2668: \tablecaption{AEOS $R$-band photometry from unfiltered observations \tablenotemark{a}}
2669: \tablehead{
2670: \colhead{$t_{\rm mid}$\tablenotemark{b}}
2671: &\colhead{$R$}&
2672: \colhead{$\sigma_R$} \\
2673: \\
2674: \colhead{(s)}&
2675: \colhead{}& \colhead{}
2676: }
2677: \startdata
2678: 568.6&16.708&0.016\\
2679: 681.3&17.046&0.018\\
2680: 794.0&17.337&0.020\\
2681: 906.8&17.573&0.020\\
2682: 1019.6&17.766&0.020\\
2683: 1132.3&17.940&0.020\\
2684: 1245.1&18.101&0.020\\
2685: 1357.9&18.229&0.023\\
2686: 1470.7&18.339&0.025\\
2687: 1583.5&18.454&0.026\\
2688: 1696.3&18.545&0.028\\
2689: 1809.1&18.640&0.034\\
2690: 1922.0&18.724&0.033\\
2691: 2034.8&18.814&0.040\\
2692: 2147.6&18.865&0.037\\
2693: 2260.3&18.924&0.042\\
2694: 2373.1&18.941&0.048\\
2695: 2485.9&18.941&0.064\\
2696: 2598.7&19.042&0.044\\
2697: 2711.4&19.087&0.049\\
2698: 2824.2&19.104&0.050\\
2699: 2937.0&19.109&0.054\\
2700: 3049.7&19.142&0.054\\
2701: 3162.5&19.150&0.052\\
2702: 3275.3&19.160&0.056\\
2703: 3388.2&19.169&0.055\\
2704: 3501.0&19.158&0.050\\
2705: 3613.8&19.139&0.056\\
2706: 3726.5&19.185&0.055\\
2707: 3839.3&19.206&0.056\\
2708: 4062.4&19.188&0.056\\
2709: 4175.3&19.196&0.055\\
2710: 4288.1&19.194&0.057\\
2711: 4401.0&19.208&0.051\\
2712: 4513.8&19.167&0.054\\
2713: 4626.6&19.150&0.053\\
2714: 4739.4&19.163&0.065\\
2715: 4852.2&19.160&0.057\\
2716: 5002.6&19.145&0.050\\
2717: \enddata
2718: \tablenotetext{a}{The original data set was grouped and combined into 39 images.}
2719: \tablenotetext{b}{The time at the middle point of several combined images. }
2720: \label{tab:aeos}
2721: \end{deluxetable}
2722:
2723: \begin{deluxetable}{lrccll}
2724: \tablecaption{Keck/Gemini-S photometry of GRB 071003}
2725: \tablehead{
2726: \colhead{$t_{\rm mid}$\tablenotemark{a}}&
2727: \colhead{Exp. time}&\colhead{Mag}&
2728: \colhead{Error} & \colhead{Filter} &\colhead{Telescope} \\
2729: \colhead{(s)}&
2730: \colhead{(s)}& \colhead{}&
2731: \colhead{} & \colhead{} & \colhead{}
2732: }
2733: \startdata
2734: 9523.7 & 2$\times$20 & 18.59\tablenotemark{b} & 0.25 & $C_R$ & Keck I \\
2735: 76891.8 & 300 & 20.32 & 0.07 & $g$ & Keck I \\
2736: 77044.0 & 300 & 19.43 & 0.06 & $R$ & Keck I \\
2737: 231174.7 & 450 & 22.33 & 0.20 & $g$ & Gemini-S \\
2738: 231802.3 & 450 & 21.57 & 0.32 & $i$ & Gemini-S \\
2739: 232430.8 & 450 & 21.97 & 0.22 & $r$ & Gemini-S \\
2740: 233056.5 & 450 & 21.35 & 0.40 & $z$ & Gemini-S \\
2741: 515855.0 & 1485 & 22.61 & 0.30 & $R$ & Keck I \\
2742: 516250.5 & 975 & 23.56 & 0.30 & $g$ & Keck I \\
2743: 517510.4 & 720 & 23.56 & 0.45 & $u$ & Keck I \\
2744: 604978.0 & 660 & 23.06 & 0.50 & $R$ & Keck I \\
2745: 605144.4 & 780 & 24.05 & 0.40 & $g$ & Keck I \\
2746: 682211.1 & 330 & 23.56 & 0.40 & $V$ & Keck I \\
2747: 682211.9 & 660 & 24.42 & 0.50 & $u$ & Keck I \\
2748: 682940.0 & 840 & 23.40 & 0.60 & $R$ & Keck I \\
2749: 1373589. & 1800 & 21.58 & 0.03 & $K'$& Keck II \\
2750: \enddata
2751: \tablenotetext{a}{The time (s) at the middle point of the observations.}
2752: \tablenotetext{b}{Measured from unfiltered images from the Keck I HIRES guider. }
2753: \label{tab:keck}
2754: \end{deluxetable}
2755:
2756: \begin{deluxetable}{lrccl}
2757: \tablecaption{P60 photometry of GRB 071003}
2758: \tablehead{
2759: \colhead{$t_{\rm start}$\tablenotemark{a}}&
2760: \colhead{Exp. time}&\colhead{Mag}&
2761: \colhead{Error} & \colhead{Filter} \\
2762: \colhead{(s)}&
2763: \colhead{(s)}&\colhead{}&
2764: \colhead{} & \colhead{}
2765: }
2766: \startdata
2767: 176.0 & 60 & 14.57 & 0.06 & $R$\\
2768: 261.0 & 60 & 15.08 & 0.05 & $i^\prime$\\
2769: 347.0 & 60 & 15.37 & 0.07 & $z^\prime$\\
2770: 432.0 & 60 & 16.03 & 0.07 & $R$\\
2771: 518.0 & 60 & 16.41 & 0.07 & $i^\prime$\\
2772: 603.0 & 60 & 16.43 & 0.08 & $z^\prime$\\
2773: 689.0 & 60 & 16.88 & 0.09 & $R$\\
2774: 775.0 & 60 & 17.12 & 0.10 & $i^\prime$\\
2775: 860.0 & 60 & 17.02 & 0.11 & $z^\prime$\\
2776: 1309.0 & 120 & 17.92 & 0.20 & $z^\prime$\\
2777: 1454.0 & 120 & 18.71 & 0.10 & $g$\\
2778: 1891.0 & 120 & 18.67 & 0.28 & $z^\prime$\\
2779: 2037.0 & 120 & 19.33 & 0.15 & $g$\\
2780: 2618.0 & 120 & 19.44 & 0.20 & $g$\\
2781: \enddata
2782: \tablenotetext{a}{The start time of the exposure, in seconds after the BAT trigger.}
2783: \label{tab:p60}
2784: \end{deluxetable}
2785:
2786: %\clearpage
2787:
2788:
2789: \begin{deluxetable}{lrrrrr}
2790: \tablecaption{Radio observations of GRB 071003}
2791: \tablewidth{0pt}
2792: \tablehead{
2793: \colhead{UT Date} & \colhead{$t_{\rm mid}$} & \colhead{Frequency} & \colhead{Flux density} &
2794: \colhead{Error} \\
2795: \colhead{Observation} & \colhead{hr} & \colhead{GHz} & \colhead{$\mu$Jy} &
2796: \colhead{$\mu$Jy}
2797: }
2798: \startdata
2799: 2007 Oct. 05, 1.85 & 42.168 & 8.46 & 393 & 55\\
2800: 2007 Oct. 07, 3.38 & 91.698 & 8.46 & 430 & 50\\
2801: 2007 Oct. 07, 3.92 & 92.238 & 4.86 & 220 & 54\\
2802: 2007 Oct. 12, 1.03 & 209.248 & 8.46 & 431 & 51\\
2803: 2007 Oct. 14, 14.84& 271.158 & 8.46 & 332 & 67\\
2804: 2007 Oct. 24, 23.58& 519.898 & 8.46 & 260 & 42\\
2805: 2007 Oct. 25, 0.04 & 520.358 & 4.86 & 119 & 46\\
2806: 2007 Nov. 05, 0.01 & 785.328 & 8.46 & 109 & 45\\
2807: 2007 Nov. 07, 0.18 & 833.336 & 4.86 & 93 & 52\\
2808: \enddata
2809: \label{tab:vla}
2810: \end{deluxetable}
2811:
2812:
2813:
2814: %\clearpage
2815:
2816:
2817: \begin{deluxetable}{ccccccccc}
2818: \tablewidth{0pc}
2819: \tablecaption{Absorption Lines in the Afterglow Spectrum of GRB~071003\label{tab:ew}}
2820: \tabletypesize{\footnotesize}
2821: \tablehead{\colhead{$\lambda$} & \colhead{$z$} & \colhead{Transition} & \colhead{$W^a$} & \colhead{$\sigma(W)^b$} \\
2822: (\AA) & & & (\AA) & (\AA) }
2823: \startdata
2824: 3549.69 & 0.37223&FeII 2586&$<2.51$&\\
2825: 3568.06 & 0.37223&FeII 2600&$2.33$&0.59\\
2826: 3837.72 & 0.37223&\ion{Mg}{2} 2796&$2.48$&0.20\\
2827: 3847.65 & 0.37223&\ion{Mg}{2} 2803&$2.14$&0.19\\
2828: 3915.45 & 0.37223&MgI 2852&$1.02$&0.17\\
2829: 4032.63 & 1.60435&CIV 1548&$0.22$&0.06\\
2830: 4039.88 & 1.60435&CIV 1550&$<0.28$&\\
2831: 4351.92 & 1.60435&AlII 1670&$<0.14$&\\
2832: 5003.26 & 1.10019&FeII 2382&$0.20$&0.05\\
2833: 5276.54 & 1.60435&ZnII 2026&$<0.08$&\\
2834: 5399.79 & 0.37223&CaII 3934&$0.61$&0.07\\
2835: 5417.99 & 0.93740&\ion{Mg}{2} 2796&$0.61$&0.05\\
2836: 5432.79$^c$ & 1.10019&FeII 2586&$0.46$&0.05\\
2837: 5447.85 & 0.37223&CaII 3969&$0.46$&0.07\\
2838: 5872.31 & 1.10019&\ion{Mg}{2} 2796&$0.80$&0.05\\
2839: 5888.27 & 1.10019&\ion{Mg}{2} 2803&$0.68$&0.06\\
2840: 6105.90 & 1.60435&FeII 2344&$<0.17$&\\
2841: 6206.91 & 1.60435&FeII 2382&$0.26$&0.04\\
2842: 6240.46 & 1.60435&FeII* 2396a&$0.25$&0.04\\
2843: 6265.95 & 1.60435&FeII* 2405&$<0.16$&\\
2844: 6282.68 & 1.60435&FeII* 2411b&$0.18$&0.03\\
2845: 6284.57 & &&$0.72$&0.12\\
2846: 6734.47 & &&$0.97$&0.15\\
2847: 6737.28 & 1.60435&FeII 2586&$0.16$&0.04\\
2848: 6772.60 & 1.60435&FeII 2600&$0.27$&0.05\\
2849: 7301.58 & 1.60435&\ion{Mg}{2} 2803&$0.17$&0.05\\
2850: 7430.06 & 1.60435&MgI 2852&$<0.24$&\\
2851: 8091.56 & &&$0.92$&0.13\\
2852: 8436.10$^d$ & &&$0.86$&0.26\\
2853: 8534.91$^d$ & &&$0.72$&0.17\\
2854: 8599.02$^d$ & &&$1.34$&0.17\\
2855: \enddata
2856: \tablenotetext{a}{Equivalent widths are rest-frame values and assume the redshift given in Column 2.}
2857: \tablenotetext{b}{Limits are $2 \sigma$ statistical values.}
2858: \tablenotetext{c}{Blended with \ion{Mg}{2}~$\lambda$2803 at $z=0.937$.}
2859: \tablenotetext{d}{These features may be residuals from sky subtraction.}
2860: \label{tab:lines}
2861: \end{deluxetable}
2862:
2863:
2864: \begin{deluxetable}{llllll}
2865: \tablewidth{0pc}
2866: \tablecaption{Optical Light-Curve Fits: Color Change}
2867: \tablehead{ \colhead{Model Description} & \colhead{$\Delta\beta_{0-1}$} & \colhead{$\Delta\beta_1(b-a)$} & \colhead{$\Delta\beta_{0-2}$} & \colhead{$\beta_2$} & \colhead{$\chi^2/\nu$}}
2868: \startdata
2869: % description Dbeta_0-1 Dbeta_1(b-a) Dbeta_0-2 beta_2 chisq/dof
2870: Fully monochromatic & 0 & 0 & 0 & 0.72$\pm$0.10 & 125.765 / 81 \\
2871: Uniformly chromatic bump & 0.22$\pm$0.27 & 0 & 0 & 0.68$\pm$0.10 & 125.100 / 80 \\
2872: Variably chromatic bump & 0.66$\pm$0.33 & 1.05$\pm$0.47 & 0 & 0.70$\pm$0.10 & 120.339 / 79 \\
2873: Chromatic rebrightening & 0 & 0 & 0.77$\pm$0.31 & 1.26$\pm$0.11 & 120.040 / 80 \\
2874: Chromatic bump+rebrightening& 0.75$\pm$0.33 & 1.09$\pm$0.47 & 0.84$\pm$0.31 & 1.26$\pm$0.11 & 113.713 / 78 \\
2875: \enddata
2876: \tablecomments{Summary of relevant parameters and $\chi^2$ for models
2877: allowing or disallowing color transitions and chromatic breaks between
2878: the various components. Values without uncertainties are fixed.
2879: Component 0 is the fast-decay component, Component 1 is the bump, and
2880: Component 2 is the late rebrightening. The absolute late-time
2881: spectral index $\beta_2$ is not a model parameter, but is fit
2882: externally after completion of the fit.}
2883: \label{tab:colorchange}
2884: \end{deluxetable}
2885:
2886: \begin{deluxetable}{lllll}
2887: \tablewidth{0pc}
2888: \tablecaption{Optical Light-Curve Fits: $t_0$}
2889: \tablehead{ \colhead{Model Description} & \colhead{$dt_0$} & \colhead{$dt_1$} & \colhead{$dt_2$} & \colhead{$\chi^2/\nu$} \\
2890: \colhead{} & \colhead{(s)} & \colhead{(s)} & \colhead{(s)} & \colhead{}
2891: }
2892: \startdata
2893: % t_0(0) t_0(1) t_0(2) chisq/dof
2894: Reference & 0 & 0 & 0 & 113.713 / 78 \\
2895: Decay & $-0.01 \pm 3.01$ & 0 & 0 & 113.713 / 77 \\
2896: Bump & 0 & 60.5$\pm$20.4 & 0 & 112.700 / 77 \\
2897: Bump (prompt pulse)& 0 & 125.0 & 0 & 115.118 / 78 \\
2898: Rebrightening & 0 & 0 & 1245$\pm$311 & 111.149 / 77 \\
2899: \enddata
2900: \tablecomments{Summary of relevant parameters and $\chi^2$ for models
2901: using a $t_0$ different from the trigger time. In all cases, the
2902: favored color-change model (chromatic bump and rebrightening) was
2903: used. Values without uncertainties are fixed. Component 0 is the
2904: fast-decay component, Component 1 is the bump, and Component 2 is the
2905: late rebrightening.}
2906: \label{tab:t0}
2907: \end{deluxetable}
2908:
2909:
2910: \begin{deluxetable}{llll}
2911: \tablecaption{Radio Modeling of GRB 071003}
2912: \tablewidth{0pt}
2913: \tablehead{
2914: \colhead{Parameter} & \colhead{Value (broken power law)} & \colhead{Value (unbroken)} & \colhead{Value (unbroken w/scintillation\tablenotemark{a})}
2915: }
2916: \startdata
2917: $\alpha_b$ & $-$0.11 $\pm$ 0.21 & 0.27 $\pm$ 0.06 & 0.34 $\pm$ 0.10 \\
2918: $\alpha_a$ & 0.81 $\pm$ 0.25 & -- & --\\
2919: $t_{break}$ & 8.51 $\pm$ 3.78 & -- & --\\
2920: $\beta$ & $-$1.11 $\pm$ 0.34 & $-$1.15 $\pm$ 0.44 & $-$1.15 $\pm$ 0.42 \\
2921: $\chi^2/\nu$& 2.45 / 4 & 15.32 / 6 & 6.07 / 6 \\
2922: \enddata
2923: \tablecomments{Best-fit parameters of a fit to the radio afterglow of
2924: GRB 071003 using a \cite{Beuermann+1999} broken power-law model versus
2925: an unbroken power-law model. The improvement for the broken power-law
2926: fit is significant given the flux uncertainties, but due to
2927: interstellar scintillation may be coincidental. If a small amount of
2928: interstellar scintillation uncertainty is added in quadrature, an
2929: unbroken power-law fit is reasonable.}
2930: \tablenotetext{a}{In this model, we added a 15\% error to all X-band
2931: points and a 22\% error to all C-band points.}
2932: \label{tab:radiomodel}
2933: \end{deluxetable}
2934:
2935:
2936: \begin{deluxetable}{llll}
2937: \tablewidth{0pc}
2938: \tablecaption{Model fluxes at $t=2.67$ days}
2939: \tablehead{ \colhead{Band/Filter} & \colhead{E} & \colhead{Flux} & \colhead{Uncertainty} \\
2940: \colhead{} & \colhead{eV} & \colhead{$\mu$Jy} & \colhead{$\mu$Jy}}
2941: \startdata
2942: X-ray & 1000 & 0.036 & 0.006 \\
2943: $u$ & 3.46 & 3.17 & 1.42 \\
2944: $g$ & 2.55 & 4.47 & 0.30 \\
2945: $V$ & 2.25 & 5.07 & 0.27 \\
2946: $r$ & 1.97 & 5.97 & 1.14 \\
2947: $R$ & 1.88 & 8.01 & 0.47 \\
2948: $i$ & 1.61 & 9.16 & 0.56 \\
2949: $I$ & 1.54 & 10.34 & 0.54 \\
2950: $z$ & 1.34 & 14.74 & 2.29 \\
2951: $K$' & 0.584 & 33.59 & 16.8 \\
2952: X & 3.5e-5 & 414.6 & 91.8 \\
2953: C & 2.0e-5 & 256.1 & 73.9 \\
2954: % X & 3.5e-5 & 411.1 & 24.3 \\
2955: % C & 2.0e-5 & 218.8 & 40.2 \\
2956: \enddata
2957: \tablecomments{Fluxes of the afterglow interpolated to $t=2.67$~d
2958: after the BAT trigger using all available X-ray, optical, and radio
2959: data. Galactic extinction ($E(B-V) = 0.148$ mag) is not accounted for;
2960: however, the X-ray flux is corrected for photoelectric absorption.}
2961: \label{tab:flux}
2962: \end{deluxetable}
2963:
2964:
2965:
2966: \begin{deluxetable}{lcccll}
2967: \tablecaption{Extinction models for optical/X-ray fits of GRB 071003}
2968: \tablehead{
2969: \colhead{model}&
2970: \colhead{$A_V$}&\colhead{$R_V$}&
2971: \colhead{$\beta$} & \colhead{$\chi^2/\nu$}
2972: }
2973: \startdata
2974: none & 0 & - & 0.913 $\pm$ 0.029 & 12.4 / 8 \\
2975: {\bf $z=0.372$} & & & & \\
2976: Milky Way & 0.239 $\pm$ 0.093 & 3.09 & 0.939 $\pm$ 0.028 & 5.80 / 7 \\
2977: SMC & 0.209 $\pm$ 0.082 & 2.74 & 0.934 $\pm$ 0.028 & 5.95 / 7 \\
2978: LMC & 0.256 $\pm$ 0.099 & 3.41 & 0.941 $\pm$ 0.029 & 5.87 / 7 \\
2979: Calzetti & 0.279 $\pm$ 0.108 & 4.05 & 0.945 $\pm$ 0.029 & 5.80 / 7 \\
2980: {\bf $z=1.10$} & & & & \\
2981: Milky Way & 0.133 $\pm$ 0.058 & 3.09 & 0.935 $\pm$ 0.029 & 7.16 / 7 \\
2982: SMC & 0.127 $\pm$ 0.052 & 2.74 & 0.935 $\pm$ 0.028 & 6.38 / 7 \\
2983: LMC & 0.132 $\pm$ 0.057 & 3.41 & 0.934 $\pm$ 0.028 & 7.16 / 7 \\
2984: Calzetti & 0.247 $\pm$ 0.095 & 4.05 & 0.957 $\pm$ 0.032 & 5.78 / 7 \\
2985: {\bf $z=1.60$} & & & & \\
2986: Milky Way & 0.139 $\pm$ 0.048 & 3.09 & 0.943 $\pm$ 0.028 & 3.94 / 7 \\
2987: SMC & 0.096 $\pm$ 0.037 & 2.74 & 0.934 $\pm$ 0.028 & 5.77 / 7 \\
2988: LMC & 0.131 $\pm$ 0.045 & 3.41 & 0.940 $\pm$ 0.028 & 3.98 / 7 \\
2989: Calzetti & 0.240 $\pm$ 0.093 & 4.05 & 0.965 $\pm$ 0.033 & 5.84 / 7 \\
2990: \enddata
2991: \tablecomments{Results of various fits to the contemporaneous optical
2992: and X-ray fluxes for extinction due to either the host galaxy or the
2993: intervening absorbers at $z=0.372$ and $z=1.10$. A small amount of
2994: extinction is required to accurately fit the data, but its nature is
2995: not strongly constrained. We adopt SMC-like extinction at $z=0.372$
2996: in the discussion and plots based on the relative strength of the
2997: intervening absorber at this redshift in the spectrum.}
2998: \label{tab:extfits}
2999: \end{deluxetable}
3000:
3001:
3002:
3003: \end{document}
3004:
3005:
3006:
3007: