1: 

2: %% Beginning of file 'sample.tex'

3: 

4: %%

5: 

6: %% Modified 03 Jan 01

7: 

8: %%

9: 

10: %% This is a sample manuscript marked up using the

11: 

12: %% AASTeX v5.x LaTeX 2e macros.

13: 

14: 

15: 

16: %% The first piece of markup in an AASTeX v5.x document

17: 

18: %% is the \documentclass command. LaTeX will ignore

19: 

20: %% any data that comes before this command.

21: 

22: 

23: 

24: %% The command below calls the preprint style

25: 

26: %% which will produce a one-column, single-spaced document.

27: 

28: %% Examples of commands for other substyles follow. Use

29: 

30: %% whichever is most appropriate for your purposes.

31: 

32: 

33: 

34: \documentclass[12pt,preprint]{aastex}

35: 

36: 

37: 

38: %% manuscript produces a one-column, double-spaced document:

39: 

40: 

41: 

42: % \documentclass[manuscript]{aastex}

43: 

44: 

45: 

46: %% preprint2 produces a double-column, single-spaced document:

47: 

48: 

49: 

50: % \documentclass[preprint2]{aastex}

51: 

52: 

53: 

54: %% If you want to create your own macros, you can do so

55: 

56: %% using \newcommand. Your macros should appear before

57: 

58: %% the \begin{document} command.

59: 

60: %%

61: 

62: %% If you are submitting to a journal that translates manuscripts

63: 

64: %% into SGML, you need to follow certain guidelines when preparing

65: 

66: %% your macros. See the AASTeX v5.x Author Guide

67: 

68: %% for information.

69: 

70: 

71: 

72: \newcommand{\vdag}{(v)^\dagger}

73: 

74: \newcommand{\myemail}{hakkilaj@cofc.edu}

75: 

76: 

77: 

78: %% You can insert a short comment on the title page using the command below.

79: 

80: 

81: 

82: %\slugcomment{Not to appear in Nonlearned J., 45.}

83: 

84: 

85: 

86: %% If you wish, you may supply running head information, although

87: 

88: %% this information may be modified by the editorial offices.

89: 

90: %% The left head contains a list of authors,

91: 

92: %% usually a maximum of three (otherwise use et al.).  The right

93: 

94: %% head is a modified title of up to roughly 44 characters.  Running heads

95: 

96: %% will not print in the manuscript style.

97: 

98: 

99: 

100: \shorttitle{Two Distinct GRB Components}

101: 

102: \shortauthors{J. Hakkila and T. W. Giblin}

103: 

104: 

105: 

106: %% This is the end of the preamble.  Indicate the beginning of the

107: 

108: %% paper itself with \begin{document}.

109: 

110: 

111: 

112: \begin{document}

113: 

114: 

115: 

116: %% LaTeX will automatically break titles if they run longer than

117: 

118: %% one line. However, you may use \\ to force a line break if

119: 

120: %% you desire.

121: 

122: 

123: 

124: \title{Quiescent Burst Evidence for \\ Two Distinct GRB Emission Components}

125: 

126: 

127: 

128: %% Use \author, \affil, and the \and command to format

129: 

130: %% author and affiliation information.

131: 

132: %% Note that \email has replaced the old \authoremail command

133: 

134: %% from AASTeX v4.0. You can use \email to mark an email address

135: 

136: %% anywhere in the paper, not just in the front matter.

137: 

138: %% As in the title, you can use \\ to force line breaks.

139: 

140: 

141: 

142: \author{Jon Hakkila and Timothy W. Giblin}

143: 

144: \affil{Department of Physics and Astronomy, College of Charleston,

145:     Charleston, SC 29424-0001}

146: 

147: 

148: 

149: 

150: 

151: %% Mark off your abstract in the ``abstract'' environment. In the manuscript

152: 

153: %% style, abstract will output a Received/Accepted line after the

154: 

155: %% title and affiliation information. No date will appear since the author

156: 

157: %% does not have this information. The dates will be filled in by the

158: 

159: %% editorial office after submission.

160: 

161: 

162: 

163: \begin{abstract}

164: 

165: We have identified two quiescent GRBs (bursts having two or more 

166: widely-separated distinct emission episodes) in which the post-quiescent 

167: emission exhibits distinctly different characteristics than the pre-quiescent 

168: emission. In these two cases (BATSE GRBs 960530 and 980125), the second 

169: emission episode has a longer lag, a smoother morphology, and softer 

170: spectral evolution than the first episode. Although the pre-quiescent 

171: emission satisfies the standard internal shock paradigm, we demonstrate 

172: that the post-quiescent emission is more consistent with external shocks. 

173: We infer that some observed soft, faint, long-lag GRBs are external shocks 

174: in which the internal shock signature is not observed. We further note 

175: that the peak luminosity ratio between quiescent episodes is not in 

176: agreement with the ratio predicated by the lag vs. peak luminosity 

177: relationship. We briefly discuss these observations in terms of current 

178: collapsar jet models.

179: 

180: \end{abstract}

181: 

182: 

183: 

184: %% Keywords should appear after the \end{abstract} command. The uncommented

185: 

186: %% example has been keyed in ApJ style. See the instructions to authors

187: 

188: %% for the journal to which you are submitting your paper to determine

189: 

190: %% what keyword punctuation is appropriate.

191: 

192: 

193: 

194: \keywords{gamma rays: bursts---methods: data analysis}

195: 

196: 

197: 

198: %% From the front matter, we move on to the body of the paper.

199: 

200: %% In the first two sections, notice the use of the natbib \citep

201: 

202: %% and \citet commands to identify citations.  The citations are

203: 

204: %% tied to the reference list via symbolic KEYs. The KEY corresponds

205: 

206: %% to the KEY in the \bibitem in the reference list below. We have

207: 

208: %% chosen the first three characters of the first author's name plus

209: 

210: %% the last two numeral of the year of publication as our KEY for

211: 

212: %% each reference.

213: 

214: 

215: 

216: \section{Introduction}

217: 

218: 

219: 

220: There is strong evidence that the majority of gamma-ray burst (GRB) pulses 

221: result from internal shocks in relativistic winds 

222: \citep{sar97a,kss97,dm98,rrf00,np02}. However, a number of bursts exhibit 

223: a soft component indicative of external shocks that could be interpreted 

224: as onset of afterglow \citep{gib99,bur99,gib02}. This emission begins 

225: preferentially towards the end of the burst or even after the GRB has 

226: ended (as suggested by the BeppoSAX WFC and NFI observations of serveral 

227: x-ray afterglows \citep{cos00}), but can also appear at a time early enough 

228: to overlap the short-timescale emission (as observed in GRB 980923 

229: \citep{gib99}). In addition, extended soft $\gamma-$ray emission has been 

230: observed by co-adding the fluxes of many BATSE bursts; this might also be 

231: an indicator of the same external shock phenomenon \citep{con02}.

232: 

233: 

234: 

235: ``Quiescent'' GRBs may provide new clues to the origins and physics of 

236: GRB shocks. Quiescent bursts are GRBs that release their gamma-ray energy 

237: in more than one distinct emission episode \citep{rm01,rmr01}; e.g. they 

238: have at least one extended period during which emission is absent. A 

239: rigorous definition of quiescent time can be problematic due to the 

240: diverse nature or complexity of GRB temporal structures. 

241: 

242: 

243: 

244: \citet{rm01} introduced a quiescent GRB definition for a sample of bright 

245: BATSE bursts that used a sliding temporal window with a width of 

246: $0.05 \times {\rm T90}$ on summed, background-subtracted, BATSE 4-channel 

247: 64-ms data. These authors chose GRBs which had at least one time period 

248: for which the count rate dropped below $2\sigma$ of the summed background 

249: rate. This criterion was applied to only the brightest, longest bursts in 

250: the BATSE 4B Catalog \citep{pac99}.

251: 

252: 

253: 

254: It is possible that this definition is too permissive, since it might allow 

255: for inclusion of some GRBs that do not have distinct emission episodes. For 

256: example, some bright bursts that do not satisfy this criterion (because the 

257: first episode's decay and the second episode's rise overlap) might be 

258: classified as quiescent if they are observed at lower signal-to-noise, as 

259: might some bursts with faint, long, underlying pulses. Because the existing 

260: quiescent definition might be too permissive, we have tentatively further 

261: required quiescent GRBs to be those in which the emission drops to the 

262: background for an extended period of time that is greater than or equal 

263: to the duration of the previous emission episode. This ensures that the 

264: quiescent time is greater than the interpulse duration.

265: 

266: 

267: 

268: %Long decay tails can also affect the quiescent 

269: 

270: %definition through the T90 measurement. This is 

271: 

272: %because T90 is relatively insensitive to faint 

273: 

274: %extended emission, so that bursts with long tails 

275: 

276: %can have significantly shorter T90 values than 

277: 

278: %might be obtained using some other duration measure. 

279: 

280: %There is thus a potential classification bias 

281: 

282: %because the quiescent GRB definition depends on T90. 

283: 

284: %The definition is also less successful at 

285: 

286: %identifying faint quiescent GRBs than bright ones. 

287: 

288: %This is because inter-episodic emission is less 

289: 

290: %likely to be observed in low signal-to-noise data. 

291: 

292: %This can cause faint non-quiescent GRBs to take on 

293: 

294: %quiescent characteristics. Sampling techniques and 

295: 

296: %instrumental characteristics can also play a part 

297: 

298: %in quiescent GRB classification. For example, long 

299: 

300: %integration times increase signal-to-noise ratios, 

301: 

302: %so detection of inter-episodic emission is more 

303: 

304: %likey using a detector with long integration times. 

305: 

306: %Additionally, the definition is dependent upon the 

307: 

308: %choice of energy channels summed as well as upon 

309: 

310: %the spectral characteristics of the emission episodes. 

311: 

312: 

313: 

314: In their study of very bright quiescent GRBs, \citet{rm01} operated under 

315: the assumption that the post-quiescent episodes are due to internal shocks. 

316: We offer the hypothesis that quiescent GRBs potentially provide a laboratory 

317: in which {\it external} shock signatures indicative of the onset of afterglow 

318: may be isolated from the rest of the burst. We have thus far found two GRBS 

319: that support this hypothesis.

320: 

321: 

322: 

323: \section{Analysis}

324: 

325: 

326: 

327: GRB 960530 and GRB 980125 are bursts from the BATSE 5B catalog that deviate 

328: significantly from what is considered to be standard GRB behavior. Neither 

329: burst was observed by Bepposax, and the RXTE PCA was unable to localize GRB 

330: 980125. 

331: 

332: 

333: 

334: The BATSE 4-channel time histories of GRB 960530 and GRB 980125 are 

335: illustrated in figure~\ref{fig1} and figure~\ref{fig2}. We have measured 

336: quiescent and emission episodic durations for these GRBs via the summed 

337: four-channel sliding temporal window technique of \cite{rm01}. As shown 

338: below, the localization measurements verify that the multiple emission 

339: episodes are indeed associated with the same sources for both GRB 960530 

340: and GRB 980125.

341: 

342: 

343: 

344: GRB 960530 (BATSE trigger 5478) has two distinct emission episodes separated 

345: by 233 s. Episode (a) is a single-pulse FRED (Fast Rise Exponential Decay) of 

346: duration 23 s. Episode (b) is a smooth single-pulse episode of 37 s. The BATSE 

347: locations of these episodes are consistent with the same source; episode (a) 

348: is located at $\rm{RA} = 143.0^\circ$ and $\rm{Dec} = -53.8^\circ$ with error 

349: $\pm 1.4^\circ$, while episode (b) is located at $\rm{RA} = 143.0^\circ$ and 

350: $\rm{Dec} = -53.0^\circ$ with error $\pm 1.9^\circ$.

351: 

352: 

353: 

354: GRB 980125 (BATSE trigger 6581) has two distinct emission episodes with a 

355: faint preburst episode. The preburst episode is faint and lasts 13 s. It 

356: is followed by a short quiescent period of 18 s. Although the preburst 

357: episode satisfies our rigorous definition of a distinct emission episode, 

358: we have chosen not to include it in our analysis because it is too faint 

359: for its properties to be accurately measured. Complex emission episode (a) 

360: follows and lasts 31 s. Episode (b) is a smooth episode with two distinct 

361: pulses and a duration of roughly 45 s, it is preceded by a quiescent period 

362: of 161 s. The first of these two distinct pulses is a smooth broad pulse 

363: similar in temporal structure to GRB 960530's emission episode (b), while 

364: the second pulse (which overlaps the first) is more indicative of a standard 

365: FRED. The BATSE locations of these episodes are also consistent with the same 

366: source; the faint preburst pulse is located at $\rm{RA} = 344.2^\circ$ and 

367: ${\rm Dec} = 41.0^\circ$ with an error of $\pm 3.8^\circ$, emission episode 

368: (a) is located at ${\rm RA} = 350.6^\circ$ and $\rm{Dec} = 35.1^\circ$ with 

369: an error of $\pm 0.5^\circ$, and emission episode (b) is located at 

370: $\rm{RA} = 354.6^\circ$ and $\rm{Dec} = 34.8^\circ$ with error $\pm 1.4^\circ$.

371: 

372: 

373: We demonstrate that the smooth quiescent pulses have other peculiar properties 

374: in addition to their peculiar temporal properties. To study these quiescent 

375: GRBs in more detail, we apply three techniques: (1) spectral lags, (2) 

376: Color-Color Diagrams (CCDs), and (3) the Internal Luminosity Function (ILF).

377: 

378: 

379: 

380: \subsection{Spectral Lags}

381: 

382: 

383: Spectral lags are energy-dependent delays in the GRB temporal structure. In 

384: general, soft pulses lag behind hard pulses. Pulses also appear to broaden 

385: at softer energies as $E^{-0.4}$ \citep{nor96}. Spectral lags apparently 

386: correlate with burst parameters such as beaming angles and peak luminosities 

387: (as measured on the 256 ms timescale) \citep{nor02}.

388: 

389: 

390: 

391: Spectral lags are defined as the maximum value of the Cross-Correlation 

392: Function (CCF) of two energy channels \citep{ban97}. Poisson background 

393: variations in the time histories complicate the identification of the 

394: CCF peak, so prior studies have modeled the CCF with a fitting function 

395: to identify the lag. CCFs are typically asymmetric because GRBs generally 

396: have longer decay times than rise times. A cubic fitting function has been 

397: found to be a more accurate characterization of CCF lags \citep{nor00} than 

398: a quadratic function \citep{wu00}. 

399: 

400: Rather than use a cubic or quadratic fitting function, we have instead 

401: chosen to use the pulse model of \citet{nor96}. This function is a more 

402: natural choice than the cubic or quadratic since the CCF describes a 

403: normalized comparison between episodes whose primary structures are 

404: pulses; it thus accounts for the CCF time-asymmetry in a natural way. 

405: 

406: Application of the CCF to the different emission components of GRB 960530 

407: and GRB 980125 leads to intriguing results. The smooth pulse appearing after 

408: the quiescent period in GRB 960530 has a significantly longer lag (in all 

409: energy channel combinations) than the first FRED pulse. The lags are shown 

410: in table~\ref{tbl-1}. The results are similar and are shown in 

411: table~\ref{tbl-2}) for GRB 980125: the smooth pulse in emission episode (b) 

412: has a broader CCF and a significantly longer lag than that found during the 

413: rest of the burst. It should be noted that the lag of emission episode (a) 

414: is consistent with the lag of the second (FRED) pulse found in emission 

415: episode (b). 

416: 

417: 

418: \subsection{The Internal Luminosity Function}

419: 

420: The Internal Luminosity Function (ILF) $\psi(L)$ is the distribution of 

421: luminosity within a GRB; $\psi(L) \Delta L$ represents the fraction of 

422: time during which a burst's luminosity lies between $L$ and $L+ \Delta L$ 

423: \citep{hor97, hak04}. For the purposes of this manuscript, we discuss the 

424: ILF calculated in the 50 to 300 keV energy range using four-channel 64-ms 

425: data. The ILF is normalized to the number of Poisson background-subtracted 

426: time intervals during which emission is observed $2 \sigma$ or more above 

427: the background and by the requirement that $\Sigma \psi (L) \Delta L =  1$. 

428: It is best fit with a quasi power-law form for most bursts such that 

429: \begin{equation}

430: \psi(L) \propto L^{\alpha} \times 10^{\beta \times [\log L]^2}.

431: \end{equation}

432: We refer to $\alpha$ as the power-law index and $\beta$ as the curvature 

433: index. The variables $\alpha$ and $\beta$ appear to be indicators of burst 

434: temporal morphology (they are highly-correlated such that GRBs with large 

435: $\alpha$ also have large $\beta$ while GRBs with small $\alpha$ also have 

436: small $\beta$); smooth, FRED-like bursts typically have large $\alpha$ 

437: (roughly $-2 \leq \alpha < 3$) while spiky (narrow-pulsed) bursts typically 

438: have small $\alpha$ (approximately $-7 \leq \alpha < -2$).

439: 

440: We have measured the ILF for the different components of quiescent GRBs 

441: 960530 and 980125; these are indicated in table~\ref{tbl-1} and 

442: table~\ref{tbl-2}. We note that the late-appearing, smooth, long-lag 

443: components of both bursts are fit by noticeably steeper ILF values than 

444: are the other episodes. The ILF values of these smooth pulses thus indicate 

445: a different morphology than that normally associated with FREDs. The ILF 

446: thus supports the hypothesis that the smooth broad pulses are different 

447: in nature than the pulses found elsewhere in these two quiescent GRBs.

448: 

449: The second (FRED) pulse in emission episode (b) (at time t=255 s) of GRB 

450: 980125 is somewhat peculiar. It is smoother than a typical FRED and has a 

451: correspondingly steep value of $\alpha$, but it has a short lag consistent 

452: with the burst's emission episode (a). It is possible that overlap with 

453: the broad, long-lag pulse has affected its ILF measurement. 

454: 

455: 

456: \subsection{Color-Color Diagrams}

457: 

458: Color-color diagrams (CCDs) \citep{gib00} can be used to describe GRB 

459: spectral evolution. Several colors can be created from BATSE four-channel 

460: data; \cite{gib00} defines the hard color index ${\rm HC} = F3/F2$ where 

461: F3 is the BATSE channel 3 photon flux in units of photons cm$^{-2}$ s$^{-1}$ 

462: and F2 is the BATSE channel 2 photon flux. The soft color index SC is defined 

463: similarly as ${\rm SC} = F2/F1$ where F1 is the BATSE channel 1 photon flux. 

464: The timescales over which these color indices are measured vary in order to 

465: maintain a roughly constant signal-to-noise ratio. The BATSE color indices 

466: in general have large enough signal-to-noise ratios for reliable spectral 

467: evolution studies. Given the detector response, the observed colors of 

468: cooling external shocks have been predicted by \cite{gib00}. With this 

469: information, CCDs can be used to search GRBs for signatures of external 

470: shocks. Having ascertained that the temporal and temporally-shifted 

471: spectral structures of the broad pulses are different than those of 

472: typical GRB pulses, we now explore the spectral evolution of these 

473: pulses using CCDs. 

474: 

475: GRBs typically undergo hard to soft spectral evolution. Thus, the terminal 

476: GRB emission tends to be softer than the initial emission. This general 

477: spectral evolution can be observed in a CCD.

478: 

479: Although the emission appearing towards the end of GRBs tends to be soft, 

480: not all terminal emission exhibits the colors predicted by the external 

481: shock model. Theoretical synchrotron cooling models are predicted to 

482: occupy a fairly narrow range of hard vs. soft colors. Both the fast 

483: cooling and slow cooling models are predicted to have their color indices 

484: evolve in the regimes $0.6 \leq SC \leq 1.0$ and $0.7 \leq HC \leq 1.1$ 

485: \citep{gib00,gib02}.

486: 

487: GRB 960530 has previously been considered to be an afterglow candidate 

488: because it exhibits a soft gamma-ray component that was coincident with 

489: the onset of the afterglow phase \citep{gib02}. The CCD of this soft 

490: gamma-ray component was additionally in agreement with the theoretical 

491: prediction for synchrotron cooling. It is difficult to firmly establish 

492: the connection between the CCD of late emission from GRB 980125 and the 

493: synchrotron shock model, but figure~\ref{fig3} indicates that this late 

494: emission is consistent with both fast- (solid line), and slow-cooling 

495: (dashed line) synchrotron models. Furthermore, the second episode is 

496: in better agreement with synchrotron afterglow emission than the first 

497: episode \citep{fre03}.

498: 

499: The concluding time periods of GRB 960530 and GRB 980125 thus both exhibit 

500: spectral behavior that evolves from hard to soft (such as is typically found 

501: in GRBs). However, this behavior is also consistent with a particular spectral 

502: evolution expected from external shocks initiating an afterglow.

503: 

504: 

505: \section{Discussion}

506: 

507: The late, long-lag pulses of quiescent GRBs 960530 and 980125 have different 

508: temporal and spectral characteristics than those found earlier in the bursts. 

509: More notably, the properties are also different than those generally observed 

510: in GRBs. These properties indicate a separate pulse component which has been 

511: conveniently isolated from the standard emission in these two GRBs by the 

512: quiescent interval.

513: 

514: 

515: \subsection{Are the Long-Lag Components Internal or External Shocks?}

516: 

517: Despite the evidence linking GRB pulses to internal shocks, the long-lag 

518: pulses of GRBs 960530 and 980125 appear to be inconsistent with internal 

519: shocks. As can be seen in figure~\ref{fig1} and figure~\ref{fig2}, the 

520: long-lag pulses appearing late in the bursts are broader than the 

521: short-lag pulses appearing early in the bursts, as expected from 

522: external shock models \citep{fen96,sar97,fen99,frr99}. We thus suspect 

523: that this newly-identified emission component is an external shock 

524: signature.

525: 

526: Further evidence supporting the external shock hypothesis can be found 

527: from the relationship observed between the duration of the post-quiescent 

528: emission episode and the quiescent time \citep{rm01,rmr01}. These authors 

529: found a linear, one-to-one ratio which they interpreted as a metastable 

530: configuration in the outflow such that a significant energy fraction still 

531: stored in the fireball is available for liberation through additional 

532: internal shocks at a later time. Therefore a longer quiescent interval will

533: be followed by a longer post-quiescent emission episode. GRB 960530 and GRB 

534: 980125 do not obey this relation. In GRB 960530, the ratio of the 

535: post-quiescent emission episode duration to the quiescent time is 

536: $37 {\rm s}/233 {\rm s} = 0.16$, while in GRB 980125 the ratio is 

537: $45 {\rm s}/161 {\rm s} = 0.28$. These values deviate considerably 

538: from those considered ``standard'' for quiescent GRBs. Because these 

539: values are not as expected for internal shocks, they could be examples 

540: of external shocks.

541: 

542: The newly-identified long-lag pulses have been found to occur towards the 

543: end of these GRBs; however, we note that the long-lag pulse in GRB 980125 

544: occurred prior to an additional concluding FRED pulse. (NOTE: both 

545: quiescent pulses have nearly the same peak flux.) The long-lag component 

546: may be present at even earlier times in other bursts. If so, then the 

547: newly-identified pulse type may be difficult to isolate because it is 

548: smooth, broad, and faint: it could overlap with narrower internal shock 

549: pulses to create complex temporal structures and complex spectral evolution. 

550: We note that many GRBs can be characterized by complex combinations of 

551: overlapping broad and narrow pulses. 

552: 

553: 

554: \subsection{Impact on the Lag vs. Peak Luminosity Relation}

555: 

556: The lag vs. peak luminosity relation \citep{nor00} is

557: \begin{equation}

558: L_{53} \approx 1.3 \times \tau^{-1.14}

559: \end{equation}

560: where $\tau = ({\rm lag_{31}}/0.01 {\rm s})$, ${\rm lag}_{31}$ is the 

561: lag between the 100-300 keV channel and the 25-50 keV channel, and $L_{53}$ 

562: is the isotropic peak luminosity on the 256 ms timescale in units of $10^{53}$ 

563: erg s$^{-1}$. The relation was originally calibrated for a half-dozen GRBs of 

564: known redshift. However, the anomalously-underluminous long-lag GRB 980425 was 

565: not included in the calibration. 

566: 

567: The long-lag emission pulses of GRBs 960530 and 980125 provide a dilemma for 

568: the lag vs. peak luminosity relation, since that relation implicitly assumes 

569: that each GRB can be characterized by a single lag. However, very long 

570: post-quiescent lags do not put the relation at risk since there are normal 

571: bursts having even longer lags. According to figure 2 of \citet{nor02}, 

572: roughly 300 individual GRBs have ${\rm lag}_{21} > 0.43$ s, where 

573: ${\rm lag}_{21}$ is the lag between the 50-100 keV channel and the 25-50 keV 

574: channel (${\rm lag}_{21} = 1.17 {\rm s}$ for GRB 960530 and 

575: ${\rm lag}_{21} = 0.42 {\rm s}$ for GRB 980125). The long post-quiescent 

576: lags of GRBs 960530 and 980125 are common enough that they appear to 

577: represent a normal GRB behavior. 

578: 

579: It is possible that some bursts exhibit {\it only} the smooth, long lag 

580: (external shock) component. The variable, short lag (internal shock) 

581: component would most likely be undetected in these bursts. There is a 

582: distinct possibility, then, that \citet{nor02} has applied the lag vs.\ 

583: peak luminosity relation to bursts dominated by short-lag components as 

584: well as to potentially different bursts dominated by long-lag components. 

585: Quiescent GRBs 960530 and 980125 can provide a consistency check for GRBs 

586: that exhibit both behaviors. One would expect the post-quiescent to 

587: pre-quiescent luminosity ratios of these GRBs $as obtained from the lag

588: measurements$ to be equal to their peak flux ratios. 

589: 

590: For pre-quiescent episode $a$ and post-quiescent episode $b$, the lag vs.\ 

591: peak luminosity relation predicts a ratio of peak luminosities

592: \begin{equation}

593: \frac{L_b}{L_a} = (\frac{\tau_b}{\tau_a})^{-1.14}.

594: \end{equation}

595: For GRB 960530, $\tau_{\rm b}/\tau_{\rm a} = 8.7$, from which 

596: $L_{\rm b}/L_{\rm a} = 0.065$. However, the ratio of 256 ms 

597: peak fluxes gives $P_{\rm b}/P_{\rm a} = L_{\rm b}/L_{\rm a} = 0.46$. 

598: For GRB 980125, $\tau_{\rm b}/\tau_{\rm a} = 0.048$, which predicts 

599: $L_{\rm b}/L_{\rm a} = 0.024$. The ratio of 256 ms 

600: peak fluxes is $P_{\rm b}/P_{\rm a} = L_{\rm b}/L_{\rm a} = 0.15$.

601: 

602: The peak luminosities of the short-lag (a) and long-lag (b) components 

603: are plotted in figure~\ref{fig4} together with the six short-lag GRBs 

604: (filled diamonds) used to calibrate the lag vs.\ peak luminosity relation 

605: \citep{nor00}. The luminosities of the long-lag components are predicted 

606: using both the lag vs.\ peak luminosity relation and peak flux ratios; 

607: the values predicted by the lag vs. peak luminosity relation lie on the 

608: theoretical (dashed) line while those predicted from peak flux ratios are 

609: shifted vertically from this line. The luminosities calculated from peak 

610: fluxes are larger than those calculated from the lag vs. peak luminosity 

611: relation, suggesting that the luminosities of these long-lag components 

612: are overluminous relative to the lag vs. peak luminosity relation.

613: 

614: This is difficult to reconcile with long-lag GRB 980425, which is 

615: underluminous relative to the lag vs. peak luminosity relation (see 

616: figure~\ref{fig4}) rather than overluminous. Even though GRB 980425 

617: and the post-quiescent emission of GRBs 960530 and 980125 all have 

618: long lags of similar durations, it appears that these lags are not 

619: good predictors of peak luminosity. The long-lag GRBs do not all 

620: obey the same lag vs. peak luminosity relation as short-lag GRBs, 

621: and it is possible that they do not obey a lag vs.\ peak luminosity 

622: relation at all.

623: 

624: 

625: \subsection{How Common is the Long-Lag Emission Component?}

626: 

627: The long-lag emission component is likely common in many GRBs, even though 

628: it has, thus far, only been expicitly identified in only two quiescent GRBs. 

629: The lag calculations of \cite{nor00} use an apodization technique to remove 

630: low-intensity emission from the lag calculation because a long-lag signature 

631: hidden in the low-intensity emission is capable of distorting the lag 

632: measurement (e.g. GRB 990123). The authors thus implicitly recognize the 

633: presence of a long-lag component arising from low-intensity emission. 

634: In addition to this signature, we have seen evidence of additional 

635: long-lag emission in the decaying tail of GRB 991216 (BATSE trigger 7906). 

636: As indicated by \cite{nor02}, it is also possible that many GRBs consist 

637: of a long-lag component only.

638: 

639: 

640: \subsection{Impact on Theoretical Models}

641: 

642: The two different pulse types observed here presumably arise from the same 

643: collimated relativistic outflow. The narrow pulses originate from colliding 

644: shells in the high-$\Gamma$, tightly-collimated jet center (e.g. 

645: \cite{mrw98,kum00}), where $\Gamma$ is the bulk Lorentz factor. The external 

646: shocks initiating the onset of afterglow are produced by a low-$\Gamma$, 

647: slowly-expanding shock found at the interface between the jet and the external 

648: medium (i.e. the external shock). The opening angle of the jet should increase 

649: as deceleration sets in (e.g. \citet{rrm99}). The detailed structure of GRB 

650: relativistic jets is still unclear, but determination of this structure is 

651: currently being performed using numerical hydrodynamic codes (e.g. 

652: \citep{zha03,alo03}). In fact, hydrodynamic simulations indicate that the 

653: beaming angle of the high-$\Gamma$ component is limited to $\sim 5^\circ$, 

654: whereas the low-$\Gamma$ component can have a beaming angle exceeding 

655: $15^\circ$ \cite{zha03}. The angular expansion of the external shock can be 

656: aided by the presence of a ``cocoon jet'', which is a secondary jet structure 

657: that forms when energy is deposited in a stellar cavity prior to breakout 

658: \cite{rcr02}.

659: 

660: 

661: 

662: In the standard internal/external shock model, deceleration of the 

663: relativistic ejecta sets in at some deceleration distance 

664: $R_{dec} \sim \Gamma^{-2/(3-s)}$ in the surrounding environment with 

665: density $n(r) \sim r^{-s}$, resulting in a temporal delay between the 

666: prompt internal shock $\gamma$-ray emission and the afterglow \citep{rcr02}. 

667: Thus, for a given density profile, large $\Gamma$ yields a small $R_{dec}$. 

668: A temporal overlap between the internal and external shock emission would 

669: be expected in these cases, producing complex $\gamma$-ray light curves. 

670: The optical flash that occurred some $\sim 40$ s into the prompt $\gamma$-ray 

671: emission of GRB 990123 has been interpreted as deceleration during the burst 

672: due to the reverse external shock \citep{sap99,mr99,srr02}. Interestingly, 

673: the optical flash is temporally coincident with the long-lag component of 

674: GRB 990123. \citet{sap99} note that the x-ray emission peaked some 

675: $\sim 60$ s into the burst. This overlaps in time with the long-lag 

676: emission, suggesting that the long-lag component may represent the onset 

677: of the afterglow, e.g. the long-lived forward shock emission. On the other 

678: hand, the low-intensity long-lag emission exhibits some short timescale 

679: variability. \citet{frb99} demonstrated that the widths of these low-intensity 

680: pulses do not evolve with time, suggesting no deceleration and an internal 

681: shock origin. However, this does not preclude the possibility of narrow 

682: late-time pulses overlapping a broad long-lag emission component. We note 

683: that these behaviors are quite unlike the behaviors observed in that latter 

684: parts of GRBs 950530 and 980125.

685: 

686: 

687: 

688: As noted before, the BATSE database contains many long-lag GRBs. These 

689: often have characteristics of FREDs, and many are initiated by what appear 

690: to be narrow, short-lag pulses. It is tempting to interpret these as GRBs 

691: produced by jets with small $R_{dec}$; few internal shocks are observed, 

692: and the onset of afterglow is thus the defining characteristic of their 

693: prompt emission. However, this hypothesis is complicated by the menagerie 

694: of other phenomena such as X-ray rich GRBs (XRRs), X-Ray Flashes(XRFs), 

695: and orphan afterglows, that are also thought to result from off-axis GRB 

696: viewing (e.g. \citet{mrw98,zha04}). This is because XRRs and XRFs are 

697: spectrally soft, but otherwise contain many pulses; some of these pulses 

698: are narrow and some are broad (R. Vanderspek, private communication). Our 

699: explanation of the long-lag bursts would imply that XRRs and XRFs are jets 

700: with small viewing angles (to explain the variable temporal structure) seen 

701: at large redshift, rather than being GRB jets viewed off-axis. However, 

702: \citet{blo03} have shown that two recent XRFs originated at redshifts 

703: less than $z = 3.5$.

704: 

705: 

706: \section{Conclusions}

707: 

708: At least two quiescent GRBs exhibit multiple lags. The short-lag component 

709: is consistent with internal shocks while an external shock explanation is 

710: preferred for the smooth long-lag component. The long-lag component is broad 

711: and faint, and is possibly present but unrecognizable in many bursts. The 

712: apodization of \citep{nor00} implicitly recognizes its existence in order 

713: to remove it. It is possible that many or all of the long-lag bursts in the 

714: sample of \citet{nor02} exhibit only the smooth and broad long-lag component.

715: 

716: Because the short and long lags found in these quiescent GRBs are so different,

717: we have assumed that there are two distinct GRB pulse types. This hypothesis 

718: needs to be verified. The best way to approach this problem is via statistical 

719: and/or data mining studies of GRB pulse characteristics. It is important to 

720: determine the existence of one continuous pulse distribution or poossibly 

721: distinctly different distributions. Detailed analyses of BATSE and $Swift$ 

722: GRB data are planned to corroborate this work.

723: 

724: The lag vs. peak luminosity relation apparently requires re-calibration. It 

725: is not clear that long-lag pulses obey the same relation as short-lag pulses, 

726: or if they even obey a lag vs. peak luminosity relation. Since there is a 

727: correlation between lag and variability, the variability vs. peak luminosity 

728: relation \citep{rei01} might require similar re-calibration.

729: 

730: The simplest explanation for the large number of long-lag GRBs found in the 

731: BATSE database is that they are GRBs viewed from off the jet axis. The 

732: tightly-beamed, variable, short-lag internal shocks would not be seen from 

733: large viewing angles, whereas the more widely-beamed, smooth, long-lag 

734: external shocks would be. However, it is difficult to reconcile this 

735: explanation at the present time with some of the other similar phenomena 

736: associated with GRBs (e.g. x-ray rich GRBs, Gamma-Ray Flashes, and Orphan 

737: Afterglows).

738: 

739: 

740: %% If you wish to include an acknowledgments section in your paper,

741: 

742: %% separate it off from the body of the text using the \acknowledgments

743: 

744: %% command.

745: 

746: 

747: 

748: %% Included in this acknowledgments section are examples of the

749: 

750: %% AASTeX hypertext markup commands. Use \url without the optional [HREF]

751: 

752: %% argument when you want to print the url directly in the text. Otherwise,

753: 

754: %% use either \url or \anchor, with the HREF as the first argument and the

755: 

756: %% text to be printed in the second.

757: 

758: 

759: 

760: \section{Acknowledgments}

761: 

762: We are grateful to the anonymous referee for a careful reading of this 

763: manuscript, and for suggestions that strengthened the theoretical 

764: discussion. We acknowledge R. D. Preece for calculating burst locations, 

765: and we are extremely grateful to T. Piran, E. E. Fenimore, E. Ramirez-Ruiz, 

766: J. Norris, J. Bonnell, and R. Vanderspek for valuable discussions. This 

767: research was sponsored by NASA grant NRA-98-OSS-03 and NSF grant AST00-98499.

768: 

769: 

770: %% The reference list follows the main body and any appendices.

771: 

772: %% Use LaTeX's thebibliography environment to mark up your reference list.

773: 

774: %% Note \begin{thebibliography} is followed by an empty set of

775: 

776: %% curly braces.  If you forget this, LaTeX will generate the error

777: 

778: %% "Perhaps a missing \item?".

779: 

780: %%

781: 

782: %% thebibliography produces citations in the text using \bibitem-\cite

783: 

784: %% cross-referencing. Each reference is preceded by a

785: 

786: %% \bibitem command that defines in curly braces the KEY that corresponds

787: 

788: %% to the KEY in the \cite commands (see the first section above).

789: 

790: %% Make sure that you provide a unique KEY for every \bibitem or else the

791: 

792: %% paper will not LaTeX. The square brackets should contain

793: 

794: %% the citation text that LaTeX will insert in

795: 

796: %% place of the \cite commands.

797: 

798: 

799: 

800: %% We have used macros to produce journal name abbreviations.

801: 

802: %% AASTeX provides a number of these for the more frequently-cited journals.

803: 

804: %% See the Author Guide for a list of them.

805: 

806: 

807: 

808: %% Note that the style of the \bibitem labels (in []) is slightly

809: 

810: 

811: %% different from previous examples.  The natbib system solves a host

812: 

813: %% of citation expression problems, but it is necessary to clearly

814: 

815: %% delimit the year from the author name used in the citation.

816: 

817: %% See the natbib documentation for more details and options.

818: 

819: 

820: 

821: \begin{thebibliography}{}

822: 

823: \bibitem[Aloy et al.(2003)]{alo03} Aloy, M.-A., Mart\'{i}, J.-M., G\'{o}mez, J.-L., Agudo, I., M\"{u}ller, E. \& Iban\'{e}z, J.-M.\ 2003, \apj, 585, L109

824: \bibitem[Band(1997)]{ban97} Band, D.~L.\ 1997, \apj, 486, 928

825: \bibitem[Bloom et al.(2003)]{blo03} Bloom, J.~S., Fox, D., 

826: van Dokkum, P.~G., Kulkarni, S.~R., Berger, E., Djorgovski, S.~G., \& 

827: Frail, D.~A.\ 2003, \apj, 599, 957

828: \bibitem[Burenin et al.(1999)]{bur99} Burenin, R.~A.~et al.\ 

829: 1999, \aap, 344, L53

830: \bibitem[Connaughton(2002)]{con02} Connaughton, V.\ 2002, 

831: \apj, 567, 1028 

832: \bibitem[Costa(2000)]{cos00} Costa, E.\ 2000, AIP 

833: Conf.~Proc.~526: Gamma-ray Bursts, 5th Huntsville Symposium, 365

834: \bibitem[Daigne \& Mochkovitch(1998)]{dm98} Daigne, F.~\& 

835: Mochkovitch, R.\ 1998, \mnras, 296, 275

836: \bibitem[Fenimore, Madras, ~\& Nayakshin(1996)]{fen96} 

837: Fenimore, E.~E., Madras, C.~D., \& Nayakshin, S.\ 1996, \apj, 473, 998 

838: \bibitem[Fenimore et al.(1999)]{fen99} Fenimore, E.~E., 

839: Cooper, C., Ramirez-Ruiz, E., Sumner, M.~C., Yoshida, A., ~\& Namiki, M.\ 

840: 1999, \apj, 512, 683 

841: \bibitem[Fenimore \& Ramirez-Ruiz(1999)]{frr99} Fenimore, 

842: E.~E.~\& Ramirez-Ruiz, E.\ 1999, ASP Conf.~Ser.~190: Gamma-Ray Bursts: The 

843: First Three Minutes, 67 

844: \bibitem[Fenimore, Ramirez-Ruiz, \& Wu(1999)]{frb99} Fenimore, E.~E., Ramirez-Ruiz, E., \& Wu, B.\ 1999, \apjl, 518, L73

845: \bibitem[Frail et al.(2001)]{fra01} Frail, D.~A.~et al.\ 

846: 2001, \apjl, 562, L55

847: \bibitem[Freismuth, Giblin, ~\& Hakkila(2003)]{fre03} 

848: Freismuth, T.~M., Giblin, T., ~\& Hakkila, J.\ 2003, American Astronomical 

849: Society Meeting, 202

850: \bibitem[Giblin et al.(1999)]{gib99} Giblin, T.~W., van 

851: Paradijs, J., Kouveliotou, C., Connaughton, V., Wijers, R.~A.~M.~J., 

852: Briggs, M.~S., Preece, R.~D., ~\& Fishman, G.~J.\ 1999, \apjl, 524, L47 \bibitem[Giblin(2000)]{gib00} Giblin, T.~W.\ 2000, Ph.D.~Thesis

853: \bibitem[Giblin et al.(2002)]{gib02} Giblin, T.~W., 

854: Connaughton, V., van Paradijs, J., Preece, R.~D., Briggs, M.~S., 

855: Kouveliotou, C., Wijers, R.~A.~M.~J., ~\& Fishman, G.~J.\ 2002, \apj, 570, 

856: 573 

857: \bibitem[Hakkila et al.(2004)]{hak04} Hakkila, J., Giblin, T.~W., Young, K.~C., Fuller, S.~P., Stallworth, A.~D., \& Sprague, A.~P., 2004, in proceedings of the 2003 Santa Fe Gamma-Ray Burst Conference (ed. E.~E. Fenimore), submitted.

858: \bibitem[Horack \& Hakkila(1997)]{hor97} Horack, J.~M.~\& Hakkila, J.\ 1997, \apj, 479, 371

859: \bibitem[Kobayashi, Piran, \& Sari(1997)]{kss97} Kobayashi, 

860: S., Piran, T., ~\& Sari, R.\ 1997, \apj, 490, 92 

861: \bibitem[Kumar \& Piran(2000)]{kum00} Kumar, P.~\& Piran, T.\ 

862: 2000, \apj, 535, 152

863: \bibitem[Meegan et al.(2003)]{mee03} Meegan, C.~A. et al. 2003, Current BATSE Gamma-Ray Burst Catalog. Available at gammaray.msfc.nasa.gov/batse/grb/catalog

864: \bibitem[Nakar \& Piran(2002)]{np02} Nakar, E.~\& Piran, T.\ 

865: 2002, \apjl, 572, L139

866: \bibitem[M\'{e}sz\'{a}ros, Rees, \& Wijers(1998)]{mrw98} M\'{e}sz\'{a}ros, P., Rees, M.~J. \& Wijers,~R.~A.~M.~J.\ 1998, \apj, 499, 301

867: \bibitem[M\'{e}sz\'{a}ros, \& Rees(1999)]{mr99} M\'{e}sz\'{a}ros, P., \& Rees, M.~J.\ 1999, \mnras, 306, L39

868: \bibitem[Norris(2002)]{nor02} Norris, J.~P.\ 2002, \apj, 579, 386 

869: \bibitem[Norris et al.(1996)]{nor96} Norris, J.~P., Nemiroff, 

870: R.~J., Bonnell, J.~T., Scargle, J.~D., Kouveliotou, C., Paciesas, W.~S., 

871: Meegan, C.~A., \& Fishman, G.~J.\ 1996, \apj, 459, 393 

872: \bibitem[Norris, Marani, ~\& Bonnell(2000)]{nor00} Norris, 

873: J.~P., Marani, G.~F., ~\& Bonnell, J.~T.\ 2000, \apj, 534, 248

874: \bibitem[Norris, Scargle, \& Bonnell(2001)]{nor01} Norris, 

875: J.~P., Scargle, J.~D., ~\& Bonnell, J.~T.\ 2001, AIP Conf.~Proc.~587: Gamma 

876: 2001: Gamma-Ray Astrophysics, 176

877: \bibitem[Paciesas et al.(1999)]{pac99} Paciesas, W.~S.~et al.\ 1999, \apjs, 122, 465

878: \bibitem[Panaitescu \& Kumar(2002)]{pan02} Panaitescu, A.~\& 

879: Kumar, P.\ 2002, \apj, 571, 779

880: \bibitem[Ramirez-Ruiz, Celotti, \& Rees(2002)]{rcr02} Ramirez-Ruiz, E., Celotti, A., ~\& Rees, M.~J.\ 2002, \mnras, 337, 1349

881: \bibitem[Ramirez-Ruiz \& Fenimore(1999)]{rrm99} Ramirez-Ruiz, 

882: E.~\& Fenimore, E.~E.\ 2000, \aaps, 138, 521

883: \bibitem[Ramirez-Ruiz \& Fenimore(2000)]{rrf00} Ramirez-Ruiz, 

884: E.~\& Fenimore, E.~E.\ 2000, \apj, 539, 712

885: \bibitem[Ramirez-Ruiz \& Merloni(2001)]{rm01} Ramirez-Ruiz, 

886: E.~\& Merloni, A.\ 2001, \mnras, 320, L25

887: \bibitem[Ramirez-Ruiz, Merloni, ~\& Rees(2001)]{rmr01} 

888: Ramirez-Ruiz, E., Merloni, A., ~\& Rees, M.~J.\ 2001, \mnras, 324, 1147

889: \bibitem[Reichart et al.(2001)]{rei01} Reichart, D.~E., Lamb, 

890: D.~Q., Fenimore, E.~E., Ramirez-Ruiz, E., Cline, T.~L., ~\& Hurley, K.\ 

891: 2001, \apj, 552, 57

892: \bibitem[Rees \& M\'{e}sz\'{a}ros(1994)]{ree94} Rees, M.~J.~\& 

893: M\'{e}sz\'{a}ros, P.\ 1994, \apjl, 430, L93

894: \bibitem[Sari(1999)]{sar99} Sari, R.\ 1999, \apjl, 524, L43

895: \bibitem[Sari \& Piran(1997a)]{sar97a} Sari, R.~\& Piran, T.\ 

896: 1997a, \mnras, 287, 110

897: \bibitem[Sari \& Piran(1997b)]{sar97} Sari, R.~\& Piran, T.\ 

898: 1997b, \apj, 485, 270

899: \bibitem[Sari \& Piran(1999)]{sap99} Sari, R.~\& Piran, T.\ 

900: 1999, \apj, 517, L109

901: \bibitem[Soderberg \& Ramirez-Ruiz(2002)]{srr02} Soderberg, A.~M., \& Ramirez-Ruiz, E.~\ 2002, \mnras, 330, L24

902: \bibitem[Stern, Poutanen, \& Svensson(1999)]{ste99} Stern, 

903: B., Poutanen, J., \& Svensson, R.\ 1999, \apj, 510, 312

904: \bibitem[Wu \& Fenimore(2000)]{wu00} Wu, B.~\& Fenimore, E.\ 

905: 2000, \apjl, 535, L29

906: \bibitem[Zhang et al.(2004)] {zha04} Zhang, B., Dai, X., Lloyd-Ronning, N. ~M., ~\& M\'{e}sz\'{a}ros, P.\ 2004, \apj, 601, L119

907: \bibitem[Zhang, Woosley, \& MacFadyen(2003)]{zha03} Zhang, 

908: W., Woosley, S.~E., \& MacFadyen, A.~I.\ 2003, \apj, 586, 356

909: \end{thebibliography}

910: 

911: 

912: 

913: 

914: 

915: \clearpage

916: 

917: 

918: 

919: %% Use the figure environment and \plotone or \plottwo to include 

920: %% figures and captions in your electronic submission.

921: 

922: 

923: \begin{figure}

924: %\plotone{5478_hist.eps}

925: \plotone{f1.ps}

926: \caption{Four-channel time history of GRB 960530 (BATSE trigger 5478). \label{fig1}}

927: \end{figure}

928: 

929: 

930: \clearpage 

931: 

932: 

933: \begin{figure}

934: %\plotone{6581_hist.eps}

935: \plotone{f2.ps}

936: \caption{Four-channel time history of GRB 980125 (BATSE trigger 6581). \label{fig2}}

937: \end{figure}

938: 

939: 

940: \clearpage

941: 

942: 

943: 

944: \begin{figure}

945: \plottwo{f3a.ps}{f3b.ps}

946: \caption{Color-Color diagram for the first and second emission episodes of GRB 980125 (BATSE trigger 6581).\label{fig3}}

947: \end{figure}

948: 

949: 

950: 

951: \clearpage

952: 

953: 

954: 

955: \begin{figure}

956: %\plotone{laglum_sn_arrows.ps}

957: \plotone{f4.ps}

958: \caption{The lag vs. peak luminosity relation applied to GRBs 960530 and 980125).\label{fig4}}

959: \end{figure}

960: 

961: 

962: 

963: %% If you are not including electonic art with your submission, you may

964: 

965: %% mark up your captions using the \figcaption command. See the 

966: 

967: %% User Guide for details.

968: 

969: %%

970: 

971: %% No more than seven \figcaption commands are allowed per page, 

972: 

973: %% so if you have more than seven captions, insert a \clearpage 

974: 

975: %% after every seventh one. 

976: 

977: 

978: 

979: 

980: 

981: 

982: %% Tables should be submitted one per page, so put a \clearpage before

983: 

984: %% each one.

985: 

986: 

987: 

988: 

989: 

990: \clearpage

991: 

992: 

993: \begin{deluxetable}{lccccc}

994: \tabletypesize{\scriptsize}

995: \tablecaption{Properties of Quiescent GRB 960530 (BATSE trigger 5478). \label{tbl-1}}

996: \tablewidth{0pt}

997: \tablehead{

998: \colhead{Time Period} & \colhead{morphology}   & \colhead{ILF $\alpha$}   &

999: \colhead{Ch 21 lag (sec)} &

1000: \colhead{Ch 31 lag (sec)} & \colhead{Ch 32 lag (sec)}

1001: }

1002: \startdata

1003: before 200 sec & spiky FRED & $0.9 \pm 0.4$ & $0.1 \pm 0.11$ & $0.58 \pm 0.12$ & $0.58 \pm 0.04$ \\

1004: after 200 sec & smooth gradual & $-7.5 \pm 1.2$ & $1.17 \pm 0.26$ & $5.04 \pm 0.06$ & $2.77 \pm 0.06$ \\

1005: \enddata

1006: \end{deluxetable}

1007: 

1008: 

1009: 

1010: \clearpage

1011: 

1012: 

1013: 

1014: \begin{deluxetable}{lccccc}

1015: \tabletypesize{\scriptsize}

1016: \tablecaption{Properties of Quiescent GRB 980125 (BATSE Trigger 6581). \label{tbl-2}}

1017: \tablewidth{0pt}

1018: \tablehead{

1019: \colhead{Time Period} & \colhead{morphology}   & \colhead{ILF $\alpha$}   &

1020: \colhead{Ch 21 lag (sec)} &

1021: \colhead{Ch 31 lag (sec)} & \colhead{Ch 32 lag (sec)}

1022: }

1023: \startdata

1024: before 150 sec & complex + FRED & $-1.3 \pm 0.5$ & $0.04 \pm 0.02$ & $0.08 \pm 0.01$ & $0.10 \pm 0.01$ \\ 

1025: 150 to 250 sec & smooth gradual & $-3.1 \pm 0.5$ & $0.43 \pm 0.32$ & $1.66 \pm 0.19$ & $0.90 \pm 0.40$ \\

1026: after 250 sec & smooth FRED & $-3.9 \pm 0.4$ & $0.07 \pm 0.03$ & $0.13 \pm 0.08$ & $0.05 \pm 0.08$ \\

1027: \enddata

1028: \end{deluxetable}

1029: 

1030: 

1031: 

1032: %% The following command ends your manuscript. LaTeX will ignore any text

1033: 

1034: %% that appears after it.

1035: 

1036: 

1037: 

1038: \end{document}ipt. LaTeX will ignore any text

1039: 

1040: %% that appears after it.

1041: 

1042: 

1043: 

1044: \end{document}

1045: