1: % Template article for preprint document class `elsart'

2: % with harvard style bibliographic references

3: % SP 2001/01/05

4: 

5: \documentclass{elsart}

6: 

7: % Use the option doublespacing or reviewcopy to obtain double line spacing

8: % \documentclass[doublespacing]{elsart}

9: 

10: % the natbib package allows both number and author-year (Harvard)

11: % style referencing;

12: \usepackage{natbib}

13: 

14: %------------Personal definitions----------------------------------

15: \newcommand{\arcsecs}{\mbox{$^{\prime\prime}$}}

16: \newcommand{\arcmins}{\mbox{$^{\prime}$}}

17: \newcommand{\parcsec}{\mbox{$\stackrel{\prime\prime}{\textstyle .}$}}

18: \newcommand{\parcmin}{\mbox{$\stackrel{\prime}{\textstyle .}$}}

19: \newcommand{\psec}{\mbox{$\stackrel{s}{\textstyle .}$}}

20: \newcommand{\hours}{\mbox{$^{h}$}}

21: \newcommand{\mins}{\mbox{$^{m}$}}

22: \newcommand{\secs}{\mbox{$^{s}$}}

23: %--------------------------------------------------------------------

24: 

25: \usepackage{graphicx}

26: 

27: \journal{New Astronomy}

28: 

29: % The amssymb package provides various useful mathematical symbols

30: \usepackage{amsmath,amssymb}

31: 

32: \begin{document}

33: 

34: \begin{frontmatter}

35: 

36: % Title, authors and addresses

37: 

38: % use the thanksref command within \title, \author or \address for footnotes;

39: % use the corauthref command within \author for corresponding author footnotes;

40: % use the ead command for the email address,

41: % and the form \ead[url] for the home page:

42: \vspace*{-1.1cm}

43: \title{Small-scale variations in the radiating surface of the 

44:  GRB~011211 jet}

45: \author[label1,label2]{P.~Jakobsson\corauthref{cor1}},

46: \ead{pallja@astro.ku.dk}

47: \author[label1]{J.~Hjorth},

48: \author[label3]{E.~Ramirez-Ruiz},

49: \author[label4]{C.~Kouveliotou},

50: \author[label1]{K.~Pedersen},

51: \author[label1,label5]{J.~P.~U.~Fynbo},

52: \author[label6,label22]{J.~Gorosabel},

53: \author[label1]{D.~Watson},

54: \author[label1]{B.~L.~Jensen},

55: \author[label7]{T.~Grav},

56: \author[label7]{M.~W.~Hansen},

57: \author[label1]{R.~Michelsen},

58: \author[label8]{M.~I.~Andersen},

59: \author[label5,label33]{M.~Weidinger},

60: \author[label1]{H.~Pedersen}

61: 

62: \corauth[cor1]{Corresponding author.}

63: 

64: \address[label1]{Niels Bohr Institute, Astronomical Observatory, 

65:   University of Copenhagen, Juliane Maries Vej 30, 

66:   DK-2100 Copenhagen, Denmark}

67: \address[label2]{Science Institute, University of Iceland, Dunhaga 3,

68:   107 Reykjav\'{\i}k, Iceland}

69: \address[label3]{Institute of Astronomy, University of Cambridge, 

70:   Madingley Road, Cambridge, CB3 0HA, England, UK}

71: \address[label4]{NSSTC, SD-50, 320 Sparkman Drive, Huntsville, Alabama

72:   35805, USA}

73: \address[label5]{Department of Physics and Astronomy, University of

74:   Aarhus, Ny Munkegade, 8000 \AA rhus C, Denmark}

75: %\address[label6]{IAA-CSIC, P.O. Box 03004, E-18080 Granada, Spain}

76: \address[label6]{STScI, 3700 San Martin

77:   Drive, Baltimore, MD 21218, USA}

78: \address[label22]{Instituto de Astrof\'{\i}sica de Andaluc\'{\i}a 

79: (IAA-CSIC), P.O. Box 03004, E-18080 Granada, Spain}

80: \address[label7]{Institute of Theoretical Astrophysics, University of

81:   Oslo, PB 1029 Blindern, 05315 Oslo, Norway}

82: \address[label8]{Astrophysikalisches Institut Potsdam, An der

83:   Sternwarte 16, 14482 Potsdam, Germany}

84: \address[label33]{European Southern Observatory, 

85:   Karl-Schwarzschild-Stra\ss e 2, 85748, Garching bei M\"unchen, Germany}

86: \begin{abstract}

87: We report the discovery of the afterglow of the X-ray

88: rich, long-duration gamma-ray burst GRB~011211 and present evidence

89: for oscillatory behaviour in its early optical light curve. 

90: The time-scale of the fluctuations, $\sim$1 hour, is much 

91: smaller than the time of the observations, $\sim$12 hours from 

92: the onset of the gamma-ray burst. The character and strength of 

93: the fluctuations are unprecedented and are inconsistent with causally 

94: connected variations in the emission of a symmetric, relativistic 

95: blast wave, i.e. flux variations which are produced 

96: uniformly throughout the shell surface are ruled out. Therefore, 

97: the wiggles are the result of spherically asymmetric density or energy

98: variations. Additionally, there is evidence for fluctuations in the 

99: X-ray afterglow light curve. If real, the resulting difference in the

100: observed time of the peaks of the short-term

101: variations at X-ray and optical frequencies, would demonstrate that 

102: the energy content across the jet-emitting surface is not uniform.

103: \end{abstract}

104: 

105: \begin{keyword}

106: % keywords here, in the form: keyword \sep keyword

107: Gamma rays: bursts \sep X-rays: general

108: 

109: % PACS codes here, in the form: 

110: \PACS 95.85.Kr \sep 95.85.Nv \sep 98.70.Rz 

111: 

112: \end{keyword}

113: 

114: \end{frontmatter}

115: 

116: % main text

117: 

118: %------------------------Intro---------------------------------------

119: \section{Introduction}

120: GRB 011211 was detected on 2001 December 11.798 UT \citep{gand} 

121: with one of the Wide Field Cameras (WFCs) on board the 

122: Italian-Dutch satellite Beppo\-SAX. The prompt gamma-ray 

123: emission lasted $\sim$270~s, making it one of the 

124: longest bursts observed with the satellite. Following the 

125: distribution of a 2\arcmins\ radius error circle \citep{gand2}, we 

126: identified \citep{grav} the optical counterpart of the gamma-ray burst

127: (GRB). An X-ray afterglow was subsequently detected \citep{santos} 

128: and its redshift measured via absorption lines in the optical spectrum 

129: to be $z=2.140$ \citep{andy,holland}. The analysis of 

130: the X-ray spectrum \citep{james} showed emission

131: lines arising in metal-enriched material with an outflow velocity of

132: $\sim$0.1$c$. The presence of such line features strongly

133: suggests a massive stellar progenitor, but the details remain model

134: dependent \citep{lazzati}.

135: \par

136: Afterglow observations of long-duration GRBs typically show 

137: their flux to decline as a power-law in time. However, recent

138: continuous early-time monitoring of GRB optical light curves has shown  

139: evidence for variations or bumps superposed on power-law decays

140: \citep{bersier,fox,japan}. We note that these are not supernova 

141: bumps as observed in the optical afterglow (OA) light curves of

142: several bursts \citep[see e.g.][]{bloom,garnavich,jens2}. We present 

143: here for the first time a comparative study of multi-wavelength 

144: oscillations observed in the light curve of GRB~011211. 

145: Despite intensive monitoring of many GRB counterparts 

146: \citep{jens,stanek,halpern,burenin,ls,jav}, the only other burst for 

147: which short-term time-scale (less than one hour) 

148: variations have been detected in its optical light curve is 

149: GRB 021004, albeit in a random fashion \citep{bersier}. 

150: The oscillatory behaviour and the rapid, sharp decline and rise of 

151: the pulses of the light curve of GRB~011211 presented here, are unlike 

152: any other light curve reported before. Our observations provide a useful 

153: probe of the GRB explosion, the structure of the 

154: emerging jet, as well as the structure of the surrounding environment. 

155: %---------------------FIGURE----------------------------------

156: \begin{figure}

157:    \centering

158:    \includegraphics[width=0.48\textwidth]{fig1a.ps}

159:    \includegraphics[width=0.50\textwidth]{fig1b.ps}

160: \caption{\small \em Images of the field of the optical transient of 

161: GRB~011211. \emph{Left:} The $R$-band DSS-2 image. 

162: \emph{Right:} The discovery $R$-band 

163: image of the GRB~011211 optical afterglow (OA), taken with 

164: the NOT. The position of the OA is marked with an arrow. 

165: The BeppoSAX localization \citep{gand2} is indicated by a 

166: circle of radius $2\arcmins$. Both images cover a square of 

167: $5\arcmins \times 5\arcmins$.}

168:    \label{uno.fig}

169: \end{figure}

170: %----------------------END FIGURE-----------------------------

171: %---------------------------Observations-----------------------

172: \section{Observations \& data analysis}

173: We observed GRB 011211 $\sim$9.6 hours after 

174: the GRB trigger (starting 2001 December 12.22 UT) with the 

175: Nordic Optical Telescope (NOT) at the Canary Island of La Palma 

176: and with the Danish 1.54-m telescope at La Silla in Chile, in 

177: standard $B$-, $V$-, $R$-, and $I$-bands \citep{grav}. 

178: A detailed log of our observations is listed in 

179: \mbox{Table~\ref{obs.tab}}. A comparison of our first 

180: image with the Digitized Sky Survey 2 (DSS-2) frames, identified 

181: a new source shown in Fig.~\ref{uno.fig}. Subsequent observations with the 

182: NOT revealed that the source was fading, thus establishing that the 

183: nature of the transient was consistent with being the OA of GRB 011211. 

184: \par

185: Reference stars for astrometry were collected using the 

186: large field of view provided by the Danish Faint Object Spectrograph 

187: and Camera (DFOSC) mounted on the Danish 1.54-m 

188: ($13.7 \times 13.7$~arcmin$^2$). Based on $\sim$50 

189: USNO stars per image, we 

190: determined 11 independent positions of the afterglow. The mean 

191: value of the 11 afterglow coordinates is RA(J2000) = 

192: $11\hours15\mins17\psec98$ and 

193: Dec(J2000) = $-21^{\circ}56\arcmins56\parcsec1$ 

194: with an error of $0\parcsec3$.

195: \par

196: %-----------------------TABLE---------------------------------------

197: \begin{table}

198: \caption{\small \em Photometry of the afterglow of GRB~011211. 

199: The magnitude 

200: of the optical afterglow was measured relative to eight stars in 

201: the field. The calibrated magnitudes of these stars are given in 

202: J03. Due to saturation we only used a subset of these stars in some 

203: of our images. The magnitudes were calculated using point-spread 

204: function fitting photometry.}

205:   \centering

206:   \setlength{\arrayrulewidth}{0.8pt}   % Default is 0.4pt

207:   \begin{tabular}{ccccc}

208:   \vspace{-4 mm} \\

209:   \hline

210:   \hline

211: Date (UT)  & Tel. & Magnitude & Seeing   & Exp. time \\

212: (2001 Dec) &      &           & (arcsec) & (s) \\

213: \hline

214: \hspace{-5 mm}

215: \emph{B-band:}    &     &                  &      & \\

216: 12.2816  & NOT    & $ 21.229 \pm 0.047$ & 1.1 &   300 \\

217: 12.3539  & 1.54-m & $ 21.262 \pm 0.089$ & 1.4 &   200 \\

218: \hspace{-5 mm}

219: \emph{V-band:}    &     &                  &      & \\

220: 12.2907  & NOT    & $ 20.761 \pm 0.080$ & 1.0 &   300 \\

221: 12.3586  & 1.54-m & $ 20.987 \pm 0.042$ & 1.2 &   300 \\

222: \hspace{-5 mm}

223: \emph{R-band:}    &     &                  &      & \\

224: 12.2181  & NOT    & $ 20.012 \pm 0.033$ & 1.7 &   300 \\

225: 12.2227  & NOT    & $ 19.877 \pm 0.029$ & 1.4 &   300 \\

226: 12.2272  & NOT    & $ 19.918 \pm 0.022$ & 1.3 &   300 \\

227: 12.2831  & 1.54-m & $ 20.368 \pm 0.040$ & 1.5 &   600 \\

228: 12.2912  & 1.54-m & $ 20.298 \pm 0.033$ & 1.5 &   600 \\

229: 12.2994  & 1.54-m & $ 20.230 \pm 0.038$ & 1.4 &   600 \\

230: 12.3075  & 1.54-m & $ 20.285 \pm 0.042$ & 1.4 &   600 \\

231: 12.3157  & 1.54-m & $ 20.274 \pm 0.051$ & 1.3 &   600 \\

232: 12.3238  & 1.54-m & $ 20.330 \pm 0.046$ & 1.1 &   600 \\

233: 12.3633  & 1.54-m & $ 20.539 \pm 0.063$ & 1.1 &   300 \\

234: 12.3681  & 1.54-m & $ 20.616 \pm 0.100$ & 1.0 &   300 \\

235: \hspace{-5 mm}

236: \emph{I-band:}    &     &                  &      & \\

237: 12.2861  & NOT    & $ 19.965 \pm 0.110$ & 0.9 &   300 \\

238: 12.3350  & 1.54-m & $ 19.951 \pm 0.054$ & 1.3 &   300 \\

239: 12.3397  & 1.54-m & $ 19.981 \pm 0.071$ & 1.8 &   300 \\

240: 12.3491  & 1.54-m & $ 20.064 \pm 0.061$ & 1.4 &   300 \\

241: \hline

242: \end{tabular}

243:   \label{obs.tab}

244: \end{table}

245: %----------------------END TABLE------------------------------------

246: %---------------------FIGURE----------------------------------

247: \begin{figure}

248:    \centering

249:    \includegraphics[width=0.99\textwidth]{fig2.ps}

250: \caption{\small \em Comparison of the residuals in the optical and 

251: X-ray light curves of the afterglow of GRB 011211. \emph{Top:} The 

252: $R$-band data are drawn from Table~\ref{obs.tab} and the 

253: literature \citep{holland}. \emph{Middle:} 

254: The same as in the top panel, except here we also plot the 

255: $I$-, $V$-, and $B$-band residuals 

256: (data taken from Table~\ref{obs.tab}). Offsets relative to 

257: the $R$-band were applied using color 

258: information from the spectral energy distribution (J03). 

259: \emph{Bottom:} The XMM-Newton X-ray light curve residuals 

260: (0.5--10~keV). In order to compare the X-ray results with 

261: the optical, the units are expressed in 

262: terms of $-2.5 \times \log(\textrm{flux})$. In all three 

263: panels, the horizontal error bars are equal to the integration 

264: time for each data point, and $t$ is the time since 

265: the burst.}

266:    \label{dos.fig}

267: \end{figure}

268: %----------------------END FIGURE-----------------------------

269: We obtained very well sampled early $R$-band observations of the OA 

270: $\sim$10--14 hours after the burst. Our data indicated significant 

271: temporal variations superposed on the power-law fit to the 

272: light curve of the GRB \citep[][hereafter J03]{palli}. To 

273: enhance the visibility of these features, we subtracted from our data 

274: the power-law decay contribution ($\alpha_1 = -0.95$) in 

275: the optical flux; the result is shown in Fig.~\ref{dos.fig} 

276: (top panel). The deviation of the data points from a pure 

277: power-law decline 

278: is significant ($>$$5\sigma$) as derived from Spearman's rank 

279: correlation. The residual weighted mean is $0.01 \pm 0.03$ with a 

280: root mean square (rms) of $13\%$ (after correcting for the 

281: photon noise). The amplitude of the largest oscillation 

282: is $\sim$0.25 magnitudes. We detect two maxima with a single 

283: minimum in-between and by fitting Gaussians to the data set 

284: we measure the time-scale of the temporal oscillations to 

285: be $\Delta t_{R} = 0.9 \pm 0.2$ hours. In the middle panel 

286: of Fig.~\ref{dos.fig} we display the combined multicolor light 

287: curve of GRB~011211; a wave-like behaviour similar to that in the 

288: $R$-band light curve is evident in the combined $I$-, $V$-, and $B$-band 

289: residuals.

290: \par

291: The bottom panel of Fig.~\ref{dos.fig} 

292: shows the temporal variations present in the residuals of the 

293: X-ray afterglow of GRB~011211 observed with the X-ray Multi 

294: Mirror-Newton satellite (XMM-Newton) \citep{santos}. The reduction 

295: of the X-ray data set is described in detail in J03. We observe 

296: significant deviations from the power-law decline 

297: ($\alpha_{\mathrm{X}} = -1.62$) in the X-ray 

298: flux ($>$3$\sigma$ as derived from Spearman's rank

299: correlation). Although 

300: the X-ray fluctuations are not as significant as the optical ones, 

301: the value of the (X-ray) Spearman rank correlation coefficient 

302: translates to a probability of less than $3\times10^{-3}$ that this 

303: correlation is a statistical fluctuation. 

304: The weighted mean of the residuals is $0.00 \pm 0.19$, with an 

305: $\sim$11$\%$ rms variation around the mean flux (after correcting 

306: for the photon noise). There is weak evidence that the fluctuations

307: are of a similar oscillatory nature as the optical ones.

308: The fluctuation time-scale is comparable to 

309: that observed in the 

310: optical, with $\Delta t_{\textrm{X}} / \Delta t_R = 0.75 \pm 0.20$.  

311: From the short overlap of the optical 

312: and X-ray data, we find that the oscillations are not in 

313: phase with a time difference of at least $0.70 \pm 0.15$ 

314: hours (as derived from the cross correlation function). 

315: \par

316: %-----------------------DISCUSSION-----------------------------

317: \section{Discussion}

318: The short-term wave-like behaviour of the GRB 011211 optical 

319: light curve reported here is unprecedented.

320: This behaviour was not detected in the earlier analysis

321: of the OA of GRB~011211 reported by \citet{holland},

322: due to insufficient data sampling. Moreover, the variations detected

323: in the optical afterglow of GRB~021004 \citep{bersier} appeared 

324: highly random and erratic, and all other GRBs have displayed 

325: variations with much longer time-scales. We discuss below the 

326: origin of the bumps in the optical light curve of GRB~011211.

327: \par 

328: The simplest afterglow model, where a

329: relativistic jet decelerates as it expands into the ambient matter

330: leading to a radiative output with a characteristic power-law decay,

331: has been remarkably successful so far in accommodating the present

332: data \citep{pkw}. How does this model account for the unusual 

333: optical light

334: curve of GRB 011211? The strong temporal variations in the early

335: afterglow could be interpreted as a result of \emph{i)} refreshed

336: shocks created as the leading edge of the jet decelerates and is

337: caught up by slower-moving jet gas \citep[e.g.][]{rm98,pmr98,rmr01},

338: \emph{ii)} the relativistic jet impacting an external medium of

339: variable density \citep[e.g.][]{wl,enrico,lazzati2,heyl,nakar}, or

340: \emph{iii)} a non-uniform jet structure

341: \citep[e.g.][]{mes98,kumar}. We note that a smooth jet traveling

342: through a clumpy medium would quickly cease to be homogeneous, i.e.

343: it is possible that \emph{ii)} could give rise to \emph{iii)}.  

344: Refreshed shocks covering a large fraction of the emitting surface can 

345: only increase the energy of the blast wave, and therefore cannot account 

346: for the rapid decay seen in the optical data

347: (top panel of Fig.~\ref{dos.fig}). 

348: \par 

349: The sharpness of the features observed at optical frequencies is

350: difficult to reconcile with a density discontinuity

351: covering most of the visible surface of the jet \citep{np}. On the 

352: other hand, the short time-scale of the oscillations provides 

353: interesting upper limits of $\sim$1--10~AU on the size of the 

354: clumps around the source. These limits are

355: lower than the fluctuation amplitudes seen on similar scales in the

356: local interstellar medium \citep{wl}, though they may reflect the

357: length scale of comet-like clumps observed in ring nebulae surrounding

358: massive stars \citep[e.g.][]{gg}.

359: \par 

360: A promising alternative

361: interpretation is energy variations within the expanding jet. It is

362: possible for a large fraction of the emitting surface of the jet to

363: become active and the flare region to remain small, because the energy

364: content of the jet varies strongly as a function of angle. At the time

365: $t$ of the observations the region contributing to the total observed flux

366: would have a transverse size of $2 \Gamma c t$, where $\Gamma$ is the

367: Lorentz factor of the jet. Only regions the size of $\Gamma c \Delta

368: t$ can produce afterglow fluctuations with time structures of $\Delta

369: t \ll t$. In other words, the jet-emitting surface is peppered with many

370: regions (hot spots) on a scale much smaller than the narrow cone

371: visible along the line of motion (see Fig.~\ref{tres.fig}). Variations in 

372: the initial conditions as a function of the opening angle could spread the 

373: causally disconnected regions out along the line of sight to the

374: observer such that the emission would arrive at different times.

375: The corresponding overall emission at a given wavelength is then 

376: averaged over the observed

377: region, and it may increase or decrease locally depending on the

378: number, intensity, and spectrum of the different flare sites across

379: the jet surface. This can happen either because the radiating material

380: is in narrow fingers or, as the jet slows down, we observe afterglow

381: emission not just from material moving directly towards us, but also from a

382: wider range of angles.  Thus, an inhomogeneous energy jet is a viable

383: explanation for the observed rapid variations in the light curve, with the

384: size of a hot spot determining the duration of the bump.

385: \par 

386: %---------------------FIGURE----------------------------------

387: \begin{figure}[!h]

388:    \centering

389:    \includegraphics[width=0.45\textwidth]{fig3.ps}

390: \caption{\small \em GRB tomography. The emission from a radiating 

391: object moving close to the speed of light is beamed into a very 

392: narrow cone. \emph{(a)} A relativistic jet 

393: that radiates uniformly would have an emitting region

394: with a transverse size of $2 \Gamma c t$. The afterglow profile would

395: be the sum of many emitting regions, and thus, should be 

396: smooth. \emph{(b)} Only regions the size of $\Gamma c \Delta t$ can

397: produce afterglow fluctuations with a time structure of $\Delta

398: t$. The same jet pattern observed face-on is shown in 

399: \emph{(c)}. The photon flux emitted per 

400: unit solid angle and per unit time at this flare site 

401: should be larger than that arising from the rest of

402: the blast wave. The corresponding emission would then average over the

403: observed region and it may increase (or decrease) depending on the

404: intrinsic properties of different flare sites across the jet

405: surface. For example, a single region with a $\Gamma$ that is ten

406: times larger than for other regions could produce emission that doubles

407: the overall count rate in a different energy band.}

408:    \label{tres.fig}

409: \end{figure}

410: %----------------------END FIGURE-----------------------------

411: From our optical observations alone we can safely conclude that the wiggles

412: in the light curve of GRB~011211 are the result of \emph{spherically

413: asymmetric density or energy variations}, i.e. variations that cover

414: less than the observed $1/ \Gamma$ region.\footnote[1]{This is because the

415: observed time-scale of the variations ($\sim$1~hour) is much shorter

416: than the overall elapsed time after the burst trigger

417: ($\sim$12~hours), when angular smoothing would have smeared out the

418: light curve variations.} The question remains: Can we decide between 

419: the above two scenarios. In principle, simultaneous multi-wavelength

420: observations can help settle this question (see Salmonson \& Ramirez-Ruiz, in

421: preparation, for a detailed treatment) and, although the X-ray variations

422: are of lower significance than the

423: optical, in what follows we explore the implications of fluctuations

424: in both bands. As shown by J03,

425: the X-ray band is located above the cooling frequency at the time of our

426: observations, where density variations of the ambient medium have a much

427: weaker impact on the emitted flux 

428: \citep[see e.g. equation~7 in][]{nakar}.

429: Given the observed $\sim$11$\%$ variation in the X-ray flux, 

430: we would then expect a much stronger {\it correlated} optical one, 

431: which we do not observe. On the other hand, uncorrelated fluctuations 

432: at different energy bands can be produced by hot

433: spots (small regions with higher energy and/or $\Gamma$). Since 

434: $\Gamma$ strongly affects the synchrotron peak frequency, a

435: situation can easily arise where we observe a hot spot releasing 

436: most of its energy in the X-rays, without a significant optical 

437: contribution above the continuum. Conversely, an optical/infrared 

438: bump can be detected without an X-ray counterpart, at least when the 

439: cooling frequency of the afterglow continuum spectrum is located between

440: the optical and the X rays. Thus, the combined departure from a 

441: power-law decay observed in the optical and X-ray light curves 

442: seems to disfavour a scenario in which the density of the ambient 

443: medium varies and the energy is kept constant across the jet surface.

444: \par 

445: The above reasoning assumes that the X-ray and optical photons are 

446: emitted from the surface of the expanding jet, an assumption 

447: justified by the fact that the variations seen at different energies 

448: have similar time-scales. Interestingly, the analysis of the X-ray 

449: spectrum during this time frame uncovered the presence of emission 

450: line features arising from metal-enriched material \citep{james}. 

451: It is feasible that a fraction of X-ray photons reach the observer 

452: after being reflected by material away from the line of sight

453: \citep{lazzati}. While the presence of such spectral features is not

454: conclusive \citep[but see e.g.][]{darach}, the X-ray wiggles observed

455: here could be attributed to variations in the line emission intensity,

456: which in turn are caused by changes in the illuminating continuum.  The

457: X-ray afterglow could then, at least in part, be due to the continuing

458: power output from the GRB central engine interacting with the envelope

459: of a massive progenitor star \citep{rees}.  

460: \par Collapsar \citep{mwh}

461: or magnetar-like \citep{wheeler} models not only provide a natural

462: scenario for a sudden burst followed by a decaying energy output but

463: could also easily imprint a non-uniform structure to the GRB jet as it

464: bores its way through the stellar mantle \citep{zhang}. Further data

465: on early X-ray spectral and temporal features provided by

466: the Swift satellite will offer additional clues to the nature of the

467: progenitor and the relativistic jet that it triggers.

468: 

469: \section*{Acknowledgements}

470: It is a pleasure to thank Sir M. Rees, S. E. Woosley, G.  Bj\"ornsson,

471: E. H. Gudmundsson, A. Loeb, S. T. Holland and E. Nakar for 

472: helpful comments and suggestions. We are grateful to the anonymous 

473: referee for stimulating remarks which have improved the paper. 

474: Our results are based on

475: observations made with the Nordic Optical Telescope, operated on the

476: island of La Palma jointly by Denmark, Finland, Iceland, Norway, and

477: Sweden, in the Spanish Observatorio del Roque de los Muchachos of the

478: Instituto de Astrofisica de Canarias. We acknowledge the availability

479: of DSS-2 exposures used in this work. P.J. gratefully acknowledges

480: support from The Icelandic Research Fund for Graduate Students, and a

481: Special Grant from the Icelandic Research Council. E.R. thanks

482: CONACyT and the ORS for sponsorship. J.P.U.F. and K.P. acknowledge

483: support from the Carlsberg foundation. This

484: work was supported by the Danish Natural Science Research Council

485: (SNF).  The authors acknowledge benefits from collaboration within the

486: EU FP5 Research Training Network ``Gamma-Ray Bursts: An Enigma and a

487: Tool''.

488: 

489: % Bibliographic references with the natbib package:

490: % Parenthetical: \citep{Bai92} produces (Bailyn 1992).

491: % Textual: \citet{Bai95} produces Bailyn et al. (1995).

492: % An affix and part of a reference:

493: %   \citep[e.g.][Ch. 2]{Bar76}

494: %   produces (e.g. Barnes et al. 1976, Ch. 2).

495: 

496: \begin{thebibliography}{}

497: 

498: % \bibitem[Names(Year)]{label} or \bibitem[Names(Year)Long names]{label}.

499: % (\harvarditem{Name}{Year}{label} is also supported.)

500: % Text of bibliographic item

501: 

502: \bibitem[Bersier et al.(2003)]{bersier} Bersier, D., et al., 2003. 

503: ApJ 584, L43

504: 

505: \bibitem[Bloom et al.(1999)]{bloom} Bloom, J. S., et al., 1999. 

506: Nature 401, 453

507: 

508: \bibitem[Burenin et al.(2003)]{burenin} Burenin, R., et al., 2003. 

509: AstL 29, 573

510: 

511: \bibitem[Fox et al.(2003)]{fox} Fox, D. W., et al., 2003. Nature 422, 284

512: 

513: \bibitem[Fruchter et al.(2001)]{andy} Fruchter, A., Vreeswijk, P.,

514:   Rhoads, J., Burud, I., 2001. GCN Circ. 1200

515: 

516: \bibitem[Gandolfi(2001a)]{gand} Gandolfi, G., 2001a. GCN Circ. 1188

517: 

518: \bibitem[Gandolfi(2001b)]{gand2} Gandolfi, G., 2001b. GCN Circ. 1189

519: 

520: \bibitem[Garcia-Segura et al.(1996)]{gg} Garcia-Segura G.,

521:   Langer N., Mac Low M.-M., 1996. A\&A 316, 133

522: 

523: \bibitem[Garnavich et al.(2003)]{garnavich} Garnavich, P., et al., 

524: 2003. ApJ 582, 924

525: 

526: \bibitem[Gorosabel et al.(2003)]{jav} Gorosabel, J., et al., 2003. 

527: A\&A, submitted (astro-ph/0309748)

528: 

529: \bibitem[Grav et al.(2001)]{grav} Grav, T., et al., 2001. GCN Circ. 1191

530: 

531: \bibitem[Halpern et al.(2003)]{halpern} Halpern, J. P., Mirabal, N.,

532:   Bureau, M., Fathi, K., 2003. GCN Circ. 2021 

533: 

534: \bibitem[Heyl \& Perna(2003)]{heyl} Heyl, J. S., Perna, R., 2003. 

535: ApJ 586, L13 

536: 

537: \bibitem[Hjorth et al.(1999)]{jens} Hjorth, J., et al., 1999. 

538: GCN Circ. 320

539: 

540: \bibitem[Hjorth et al.(2003)]{jens2} Hjorth, J., et al., 2003. 

541: Nature 423, 847

542: 

543: \bibitem[Holland et al.(2002)]{holland} Holland, S., et al., 2002. 

544: AJ 124, 639

545: 

546: \bibitem[Jakobsson et al.(2003)]{palli} Jakobsson, P., et al., 2003. 

547: A\&A 408, 941 (J03)

548: 

549: \bibitem[Kumar \& Piran(2000)]{kumar} Kumar, P., Piran, T., 2000. 

550: ApJ 535, 152

551: 

552: \bibitem[Laursen \& Stanek(2003)]{ls} Laursen, L. T., Stanek, K. Z., 

553: 2003. ApJ 597, L107

554: 

555: \bibitem[Lazzati et al.(2002a)]{lazzati} Lazzati, D., Ramirez-Ruiz, E., 

556: Rees, M.~J., 2002a. ApJ 572, L57

557: 

558: \bibitem[Lazzati et al.(2002b)]{lazzati2} Lazzati, D., Rossi, E., 

559:   Covino, S., Ghisellini, G., Malesani, D., 2002b. A\&A 396, L5  

560: 

561: \bibitem[MacFadyen et al.(2001)]{mwh} MacFadyen,~A.~I., 

562: Woosley,~S.~E., Heger,~A., 2001. ApJ 550, 410

563: 

564: \bibitem[M\'esz\'aros et al.(1998)]{mes98}

565: M\'esz\'aros, P., Rees, M. J., Wijers, R., 1998. ApJ 499, 301

566: 

567: \bibitem[Nakar \& Piran(2003)]{np} Nakar, E., Piran, T., 2003. 

568: ApJ 598, 400

569: 

570: \bibitem[Nakar et al.(2003)]{nakar} Nakar, E., Piran, T., 

571: Granot, J., 2003. NewA 8, 495

572: 

573: \bibitem[Panaitescu et al.(1998)]{pmr98}

574: Panaitescu, A., M\'esz\'aros, P., Rees, M. J., 1998. ApJ 503, 314

575: 

576: \bibitem[Ramirez-Ruiz et al.(2001a)]{enrico} Ramirez-Ruiz, E., 

577:   Dray, L. M., Madau, P., Tout, C. A., 2001a. MNRAS 327, 829 

578: 

579: \bibitem[Ramirez-Ruiz et al.(2001b)]{rmr01} Ramirez-Ruiz, E.,

580: Merloni A., Rees M. J., 2001b. MNRAS 324, 1147

581: 

582: \bibitem[Rees \& M\'{e}sz\'{a}ros(1998)]{rm98} Rees, M. J., 

583: M\'{e}sz\'{a}ros, P., 1998. ApJ 496, L1

584: 

585: \bibitem[Rees \& M\'esz\'aros(2000)]{rees} Rees, M. J., 

586: M\'esz\'aros, P., 2000. ApJ 545, L73

587: 

588: \bibitem[Reeves et al.(2002)]{james} Reeves, J. N., et al., 2002. 

589: Nature 416, 512

590: 

591: \bibitem[Santos-Lleo et al.(2001)]{santos} Santos-Lleo, M., 

592: Loiseau, N., Rodriguez, P., Altieri, B., Schartel, N., 2001. 

593: GCN Circ. 1192

594: 

595: \bibitem[Stanek et al.(1999)]{stanek} Stanek, K. Z., Garnavich, P.,

596:   Kaluzny, J., Pych, W., Thompson, I., 1999. ApJ 522, L39 

597: 

598: \bibitem[Uemura et al.(2003)]{japan} Uemura, M., et al., 2003. 

599: Nature 423, 843

600: 

601: \bibitem[van Paradijs et al.(2000)]{pkw} van Paradijs, J., 

602: Kouveliotou, C., Wijers, R. A. M. J., 2000. ARA\&A 38, 379

603: 

604: \bibitem[Vietri \& Stella(1999)]{vs} Vietri, M., Stella, L., 1999. 

605: ApJ 527, L43

606: 

607: \bibitem[Wang \& Loeb(2000)]{wl} Wang, X., Loeb, A., 2000. 

608: ApJ 535, 788

609: 

610: \bibitem[Watson et al.(2003)]{darach} Watson, D., Reeves, J. N.,

611:   Hjorth, J., Jakobsson, P., Pedersen, K., 2003. ApJ 595, L29

612: 

613: \bibitem[Wheeler et al.(2000)]{wheeler} Wheeler, J. C., Yi, I., 

614: Hoefflich, P., Wang, L., 2000. ApJ 537, 810

615: 

616: \bibitem[Zhang et al.(2003)]{zhang} Zhang,~W.,

617:   Woosley,~S.~E., MacFadyen,~A.~I., 2003. ApJ 586, 356

618: 

619: \end{thebibliography}

620: 

621: \end{document}

622: 

623: