astro-ph0107367/part1
1: 
2: %Version that Arnon sends me (25/4/02) as last one used
3: %          aa.dem 
4: % AA vers. 4.01, LaTeX class for Astronomy & Astrophysics 
5: % demonstration file 
6: %                                                (c) Springer-Verlag HD 
7: %------------------------------------------------------ 
8: % 
9: %\documentclass[referee]{aa} 
10: % for a referee version 
11: %             
12: \documentclass[article]{aa} 
13: \usepackage{epsfig,deluxe} 
14: %              
15: \begin{document} 
16:  
17:  
18: \newcommand{\gsim}{\hbox{\rlap{$^>$}$_\sim$}} 
19:   \thesaurus{06;  19.63.1} 
20: % A&A Section 6: Form. struct. and  evolut. of stars} 
21: %  \thesaurus{06     % A&A Section 6: Form. struct. and evolut. of stars 
22: %             (03.11.1;  % Cosmogony, 
23: %              16.06.1;  % Planets and satellites: general, 
24: %              19.37.1;  % Stars: formation of, 
25: %              19.53.1;  % Stars: oscillations of, 
26: %              19.63.1)} % Stars: structure of. 
27: % 
28: \authorrunning{S. Dado, A. Dar \& A. De R\'ujula} 
29: \titlerunning{Afterglows of GRBs} 
30: \title{On the Optical and X-ray Afterglows of Gamma Ray Bursts} 
31:  
32: \author{Shlomo Dado$^{^1}$, Arnon Dar$^{^1}$ and 
33: A. De R\'ujula$^{^2}$} 
34: \institute{1. Physics Department and Space Research Institute, Technion\\ 
35:                Haifa 32000, Israel\\ 
36:            2. Theory Division, CERN, CH-1211 Geneva 23, Switzerland} 
37:            
38:  
39: \maketitle           
40:  
41: \begin{abstract} 
42: We severely criticize the consuetudinary analysis of the 
43: afterglows of gamma-ray bursts (GRBs) in the conical-ejection 
44: fireball scenarios. We argue that, instead, recent 
45: observations imply that the long-duration GRBs 
46: and their afterglows are produced by highly relativistic jets of 
47: cannonballs (CBs) emitted in supernova explosions. 
48: The CBs are heated by their collision with the supernova shell. 
49: The GRB is the boosted surface radiation the CBs emit as they reach 
50: the transparent outskirts of the shell. 
51: The exiting CBs further decelerate by sweeping up 
52: interstellar matter (ISM). The early X-ray afterglow is dominated by thermal 
53: bremsstrahlung from the cooling CBs, the optical afterglow by synchrotron 
54: radiation from the ISM electrons swept up by the CBs. 
55: We show that this model fits simply and remarkably 
56: well all the measured optical afterglows of the 15 GRBs with known 
57: redshift, including that of GRB 990123, for which unusually prompt 
58: data are available. We demonstrate that GRB 980425 was a normal 
59: GRB produced by SN1998bw, with standard X-ray and optical afterglows. 
60: We find that the very peculiar afterglow of GRB 970508
61: can be explained if its CBs  encountered a significant
62: jump in density as they moved through the ISM. 
63: The afterglows of the nearest 8 of the known-redshift GRBs 
64: show various degrees of evidence for an association 
65: with a supernova akin to SN1998bw. In all other cases such an 
66: association, even if present, would have been undetectable 
67: with the best current photometric sensitivities. 
68: This gives strong support to the proposition that most, maybe all, 
69: of the long-duration GRBs are associated with supernovae. 
70: Though our emphasis is on optical afterglows, we also 
71: provide an excellent description of X-ray afterglows. 
72:  
73: \end{abstract} 
74:  
75: \keywords{gamma rays bursts, supernovae, optical afterglow, X-ray afterglow} 
76:  
77:  
78: \section{Introduction} 
79:  
80: Our information about the once totally mysterious gamma-ray bursts (GRBs) 
81: has increased spectacularly in the past few years. 
82: The rapid directional localization of gamma-ray bursts by the 
83: satellites BeppoSAX (e.g. Costa et al.~1997), Rossi (e.g. Levine et al.~1996) 
84: and by the Inter-Planetary Network (IPN) of the spacecrafts Ulysses, 
85: Konus-Wind and NEAR (e.g. Cline et al.~1999) led to 
86: a flurry of progress: the discovery of 
87: long-duration GRB afterglows (Costa et al.~1997; Van Paradijs et al.~1998); 
88: the discovery of the GRBs' host galaxies (Sahu et al.~1997a); 
89: the measurement of their redshifts (Metzger et al.~1997b) that verified their 
90: cosmological  origin (e.g. Paczynski 1986; Meegan et al.~1992); the 
91: identification of their birthplaces ---mainly star formation regions in 
92: normal galaxies (e.g. Paczynski 1998; Holland and Hjorth 1999)--- 
93: and the first evidence for a possible association between GRBs and supernova 
94: explosions (Galama et al.~1998a). 
95:  
96: The enormous isotropic energies inferred from the redshifts and fluences of GRBs 
97: and their short-time variability have indicated that 
98: the GRBs must be produced by gravitational stellar collapse (Goodman et 
99: al. 1987, Dar et al.~1992). The prevalent belief is that they are 
100: generated by synchrotron emission from relativistic fireballs produced by 
101: mergers of compact stars (Paczynski 1986; Goodman 1986; Goodman et al.~1987), 
102: by hypernova explosions (Paczynski 1998), or by relativistic 
103: ``firecones'' (e.g. Rhoads 1997, 1999) from collapsars or failed supernovae 
104: (Woosley 1993;  Woosley \& MacFadyen 1999; MacFadyen 
105: \& Woosley 1999; MacFadyen et al.~2001). But various observations, 
106: including in particular the ones 
107: we shall extensively discuss here, strongly suggest that most of the 
108: long-duration GRBs are produced in supernova events (Dar \& De R\'ujula 
109: 2000a and references therein) by highly collimated 
110: superluminal jets (e.g. Shaviv \& Dar 1995; Dar 1997; Dar \& Plaga 1999). 
111:  
112: Various authors (e.g. Rhoads 1997, 1999; MacFadyen and Woosley 1999; 
113: Sari et al.~1999) have merged the notion that GRBs are produced by 
114: highly relativistic jets (e.g. Brainerd 1992; Woosley 1993; 
115: Shaviv \& Dar 1995; Dar 1997; Dar 1998a; 
116: Dar \& Plaga 1999) with the popular fireball 
117: models of GRBs (see, e.g. Piran 1999  and references therein) 
118: to morphe the concept of ``firecones'' or similar denominations.  
119: Firecone considerations are used to analyze ``breaks'' in GRB 
120: afterglows (Rhoads 1997, 1999; Sari et al.~1999; Frail et al.~2001) and to 
121: extract properties of the GRB engine and ejecta (Frail et al.~2001, and 
122: references therein). 
123: The fireball idea that all radiation from GRBs originates from 
124: colliding shocks is certainly interesting and worth studying. 
125: But, concerning the evolution of AGs, 
126: the idea remains essentially untested, given the cavalier 
127: treatment it has received in much of the recent literature. In 
128: Section 2 we explain this harsh opinion. 
129:  
130: In recent papers the idea of jetted, supernova-associated 
131:  GRBs was made entirely explicit with the introduction of a 
132: relativistic cannonball (CB) model of GRB production in supernova 
133: explosions (Dar \& De R\'ujula 2000a, 2000b, 2001a, 2001b,  hereafter 
134: DD2000a, etc.). The CB model is completely different from the firecone scenarios, 
135: as we explain in Section 2. The CB model, we contend, explains the 
136: main observed features of long-duration GRBs and of their afterglows. In 
137: particular, in DD2000b, we have demonstrated that the CB model predicts 
138: the temporal and spectral properties of the bursts of 
139: $\gamma$-rays correctly. 
140:  
141: In this paper we derive the detailed predictions of the CB model for the GRB 
142: optical afterglows (AGs), which we only sketched in DD2000a. 
143: We compare the predictions, which are analytic in fair 
144: approximations, with the observed optical 
145: AG of all the GRBs with known redshift. We show that the CB model 
146: describes remarkably well these optical AGs, as well as the measured 
147: X-ray AGs of these GRBs. 
148:  
149: Our detailed analysis of the AGs allows us to show how the 
150:  nearest eight GRBs with measured redshift show varying degrees of 
151: evidence of an association 
152: with a supernova (SN) akin to SN1998bw: superimposed on the smooth 
153: AG of these GRBs one can discern the light curve of SN1998bw, 
154: adequately translated and red-shifted in luminosity distance, time-dependence 
155:  and spectrum (e.g. Dar 1998b; DD2000a). In all other 
156: cases, either there are no late-time measurements of the optical 
157: AG, or the SN contribution is too dim to be resolved from the late 
158: GRB afterglow or the host galaxy light, even by the HST or the most powerful 
159: ground-based telescopes. 
160:  
161:  
162:  
163: In spite of the fact that we use similar vocabulary in what concerns 
164: the GRB engine (which is quite irrelevant to the AG properties 
165: discussed here) the CB model is completely different from the collapsar 
166: model of GRBs (Woosley 1993; Woosley \& MacFadyen 1999; MacFadyen and 
167: Woosley 1999; MacFadyen et al.~2001). The crucial differences are 
168: explained in DD2000a and DD2000b.  One of them is that 
169: we use empirical facts as our guiding line, rather than the results of 
170: simulations that fail to explode SNe and lack a proper treatment 
171: of features that are no doubt relevant (relativistic dynamics, 
172: angular momentum and its transport, magnetic fields). 
173: Concerning AGs, the CB model does have a predecessor, 
174: the ``plasmoid model'', 
175: of Chiang \& Dermer (1997). The CB model, however, differs in many 
176: crucial details; it is more complete and, unlike the plasmoid model, it 
177: is very successful in describing the observations. 
178:  
179:  
180: We devote sections 3 to 6, to make the paper self-contained, 
181: to a brief review of the CB model. The novel theoretical core of 
182: the paper is in sections 7 to 10. We propose there, in particular, 
183: a simple mechanism governing the pace of radial expansion 
184: of a CB. The mechanism naturally explains the fact, observed in 
185: quasar and microquasar ejections, that {\it cannonballs}, faithful 
186: to their name, essentially stop expanding at some point of their 
187: voyage. We also derive predictions, which turn out to be in 
188: disagreement with observations, for the case of CBs that would 
189: continuously expand in an inertial manner. 
190: The remaining sections are devoted to the description of 
191: the AG data in the CB model, to a detailed comparison 
192: with observations, which turns out to be extremely successful, 
193: and to the extraction of conclusions.  
194:  
195: Three GRBs deserve special mention in their CB model interpretation. 
196: GRB 980425 turns out to be entirely ``normal'', it is uncommonly 
197: near ($\rm z=0.0085$) but its emitted CBs are observed at an unusually large 
198: angle, giving it a normal $\gamma$-ray fluence. This interpretation  (DD2000a) 
199: is strengthened by the fact that we successfully predict its 
200: optical AG (dominated up to 20 months by SN1998bw) from its 
201: X-ray AG, which is entirely due to the CBs, and normal. 
202: GRB 970508 has an extremely peculiar AG light curve, which 
203: can be easily explained, but only if its CBs encountered a significant jump
204: in density as they moved through the ISM.
205: There are uncommonly early 
206: data on the optical AG of GRB 990123, which 
207: fit the expectations for CBs that are moving through the wind 
208: of the parent star, in its Wolf-Rayet phase. 
209: 
210:  
211: \section{Uses and abuses of fireballs} 
212:  
213: In this section we place the CB model of GRB afterglows in the perspective of 
214: the generally accepted views on the subject, which are based on the ``fireball'' 
215: model and on modifications thereof. This serves a double purpose: 
216: it clarifies how completely different the CB model is from the fireball 
217: ones, and it shows how unconvincing a fraction of the fireball literature is. 
218:  
219: In the fireball model, reviewed in Piran (1999, 2000) and Meszaros (2001), 
220:  both the $\gamma$-rays and the AG of a GRB are made by 
221: synchrotron radiation in inward- and outward-moving shocks, which 
222: are produced as relativistically expanding shells collide with each other 
223: and with interstellar material. The attitude is often espoused 
224: that the actual engine producing these colliding 
225: ejecta need not always be explicitly discussed. The possibility 
226: that the ejecta may not be spherically distributed has been repeatedly 
227: studied in the literature, but, in the fireball model, this was not done 
228: in detail prior to the influential papers of Rhoads (1997, 1999), who predicted 
229: abrupt breaks in the power law of the AG light curves. 
230: With the advent of GRB 990123, with its record equivalent spherical 
231: energy (see Table I) and an AG light curve through which it is 
232: possible to draw a broken power law (e.g. Figs. 1--4 of Holland et al.~2000a) 
233: the fireball advocates (see, e.g. Frail et al.~2001) adopted the 
234: arguments in favour of collimated GRBs (e.g. in the case of 
235: GRBs from quasars, Brainerd 1992; in the case of a funnel in 
236: an explosion, Meszaros \& 
237: Rees 1992; and in the case of jets in gravitational collapses, 
238: Shaviv \& Dar 1995, Dar 1997, Dar 1998a; Dar \& Plaga 1999, DD2000a and 
239: references therein). So have fireballs evolved into ``collimated fireballs'', 
240: ``firecones'' or ``conical fireball jets'', while maintaining the ``fire'' 
241: lineage. 
242:  
243: Consider first a proper (i.e. spherical) fireball expanding in a homogeneous 
244: (or spherically symmetric) medium. A conical section 
245: of this fireball would expand as a fixed-angle cone: a {\it firecone}. 
246: Consider next material that is ejected with a conical distribution. 
247: If the cone expands laterally 
248: at a transverse velocity $\rm v_T$, its opening angle, as viewed 
249: from the origin of the ejecta, increases with time. 
250: As illustrated in Fig.~(\ref{figtrumpet}a) the edges of the material 
251: describe a trumpet-shaped curve, not a fixed-aperture {\it cone}, 
252: as some of the names given to it may induce one to think. 
253: We call these ``firecones'' or ``conical fireball jets'' 
254: {\it firetrumpets}, since, for $\rm v_t\neq 0$, that is what they are. 
255:  
256: Let $\rm\gamma(t)$ 
257: be the Lorentz factor of the ejecta, that diminishes with time 
258: as they collide with ambient material. The light emitted by an 
259: element of a firetrumpet's surface is collimated by its motion 
260: into an angle of aperture $1/\gamma$. 
261: If $\rm v_T/\gamma$ were constant, the firetrumpet's opening angle 
262: would vary as: 
263: \begin{equation} 
264: \rm \theta_j(t)=\theta_j(0)+{v_T\over c\,\gamma}\, . 
265: \label{jetangle} 
266: \end{equation} 
267: At the time $\rm t=t_b$ at which $\rm \theta_j(t)=1/\gamma(t)$, the angle of 
268: emission of light becomes broader than the angle of the cone (Rhoads 1997, 1999). Thereafter the forward light-collimation 
269: is less efficient, an on-axis observer would see up to the 
270: edge of the cone, and no longer an increasing fraction of the ejecta 
271: (Meszaros et al. 1999). 
272: At early times $\rm t<t_{exp}$, the lateral expansion of the firetrumpet 
273: may not be important and the ejecta's deceleration as it plunges 
274: through constant-density material results 
275: ---as it would for a fixed-angle cone--- in 
276: $\rm\gamma(r)\propto r^{-3/2}$, with $\rm r(t)$ the travelled distance, 
277:  while at later times 
278: $\rm\gamma(r)\propto exp[-r/r_{exp}]$. Rhoads assumes that these 
279: two transitions occur at the same time ($\rm t_b=t_{exp}$) and 
280: that they are abrupt, leading to a {\it break}: a sudden increase 
281: in the index $\alpha$ of an AGs' power-law evolution, $\rm F\propto t^{-\alpha}$. 
282: The break-time is estimated as the time at which a cone which 
283: is {\it not} laterally expanding decelerates to $\rm \gamma(t)=1/\theta_0$. 
284:  
285:  
286: Sari et al.~(1999) change some of the parameters used by Rhoads 
287: (notably $\rm v_t=c/\sqrt{3}$ to $\rm v_T=c$) and invert 
288: $\rm t_b(\theta_j(0))$ to obtain $\rm \theta_j(t_b)$: 
289: \begin{equation} 
290: \rm \rm \theta_j(t_b) \simeq \theta_j(0) 
291: =0.1\,\left[{t_b\over 6.2\;h}\,{n_1\over E_{52}}\right]^{3\over 8}\, 
292: \label{Sari} 
293: \end{equation} 
294: where $\rm n_1$ is the local density in cm$^{-3}$ and $\rm E_{52}$ 
295: is the ejecta's energy in $10^{52}$ erg units. 
296:  
297: While these theoretical developments were taking place, more than a dozen 
298: (mainly R-band) optical AGs were being observed. They 
299: did not have abrupt breaks. The observers 
300: (GRB 990123: Fruchter et al.~1999a, Castro-Tirado et al.~1999b, 
301: Kulkarni et al.~1999;  GRB 990510:  Stanek et al.~1999, 
302: Harrison et al.~1999, Israel et al.~1999;  GRB 990705: Masetti et al.~2000;  
303: GRB 991208: Castro-Tirado et al.~2001;  
304: GRB 991216:  Halpern et al.~2000a;  GRB 000301c: Sagar et al.~2000b, 
305: Jensen et al.~2000;  GRB 000418:  Berger et al.~2001;  GRB 000926:  Fynbo 
306: et al.~2001, Sagar et al.~2001a;  GRB 010222:  Masetti et al.~2001, 
307: Stanek et al.~2001) fitted the 
308: slow steepening of AG fluences to phenomenological formulae such as: 
309: \begin{equation} 
310: \rm F_\nu = {2\,F_\nu^b \over 
311: \left[ (t/t_b)^{\alpha_1\,s}+(t/t_b)^{\alpha_2\,s}\right]^{1\over s}}\, , 
312: \label{pheno} 
313: \end{equation} 
314: which interpolate 
315:  between two power laws with a tunable ``abruptness'' $\rm s$, 
316: often set at $\rm s=1$. 
317: The values of $\rm t_b$ extracted from these fits, and their 
318: distributions, have no clear 
319: meaning, since different groups use different parametrizations, 
320: and none of them is theoretically justified. 
321:  
322: Moderski et al.~(2000), Huang at al. (2000a,b), 
323: Kumar \& Panaitescu ( 2000) and 
324: Panaitescu \& Kumar (2001) have modelled the 
325: light emitted by a firetrumpet without some of the approximations 
326: introduced by Rhoads. The evolution of the ejecta's deceleration 
327: is treated continuously. The emission is computed from isochronous 
328: points in the firetrumpet, so that light simultaneously received 
329: is light that had been simultaneously emitted 
330: (lifting the prior ``approximation'' that the speed of  light is infinite). 
331: Not having an abrupt break put in by hand, 
332: no abrupt break is predicted. The fair conclusion is that 
333: the light curves are too smooth to allow for a determination 
334: of a break time $\rm t_b$ (Moderski et al. 2000). 
335:  
336: {\it All the firetrumpet advocates place the 
337: observer precisely on the jet's axis}, as in Fig.~(\ref{figtrumpet}a), 
338: for no stated reason. It is obvious that the viewing 
339: angle is a relevant parameter that cannot be unceremoniously dismissed. 
340: In particular a non-vanishing viewing angle would contribute 
341: to erase even further any trace of a sharp break. 
342: Moreover, a distribution of viewing angles would completely erase 
343: a possible meaning to the distribution of specific $\rm t_b$ values extracted 
344: from expressions such as Eqs.~(\ref{Sari}) and (\ref{pheno}). 
345:  
346: The hypothetical firetrumpet ejecta behave in a 
347:  different way from most of the highly relativistic jets observed 
348: in quasars (e.g. radio jets: 
349: Bridle 2000; optical jets: Cranc et al.~1993; X-ray jets: Wilson et al.~2000) 
350: and microquasars (e.g. Mirabel \& Rodriguez 
351: 1994, 1999). The ejecta of the real jets, as seen from their emission 
352: point up to the point where they eventually stop and expand, generally 
353: subtend angles that {\it decrease} with time, exactly the opposite 
354: of the assumed firetrumpet behaviour of Eq.~(\ref{jetangle}) 
355: and Fig.~(\ref{figtrumpet}a). 
356: In the analysis of these real objects (e.g. Pearsons \& 
357: Zensus 1987; Mirabel \& Rodriguez 1994, 1999; Ghisellini \& Celotti 
358: 2001) it is the angle of observation ---and not the angle subtended by 
359: the ejecta--- that plays a key role. 
360:  
361: Frail et al.~(2001) 
362: use the published values of $\rm t_b$ ---fit to expressions such as 
363: Eq.~(\ref{pheno})--- to extract a set of values of $\rm \theta_j(0)$. 
364: In so doing they use a modified Eq.~(\ref{Sari}), in which the dependence 
365: on redshift and on the efficiency of light production are not overlooked. 
366: In this way they reach a series of conclusions that their analysis 
367: does not justify. 
368:  
369: The firetrumpet model may, to some extent, be correct. We have seen that, 
370: alas, the consequences of its basic assumptions have not been properly 
371: extracted. In particular, the ``anthropo-axial'' view that the 
372: ejecta of the observed AGs always point to the observer has not been shown 
373: to be a fair approximation. 
374:  
375: In Fig.~(\ref{figtrumpet}b) we illustrate the geometry of the CB model. 
376: In the AG phase, the CBs are expanding very slowly, or 
377: not at all (DD2000a), like the observed ejecta 
378: in quasars and microquasars. In contradistinction to 
379: the firetrumpet case of Eq.~(\ref{jetangle}), 
380: the angle with which the CBs are viewed from the origin 
381: {\it decreases} with time. But this angle is irrelevant, and 
382: negligible relative to the opening angle $1/\gamma$ of the 
383: emitted light. The angle at which the ejected CBs are viewed 
384: is obviously relevant and we do not 
385: set it to zero by fiat. 
386:  
387:  
388:  
389: \section{The cannonball model of GRBs} 
390:  
391: In the CB model, long-duration GRBs and their AGs are produced in 
392: core collapse supernovae by jets of highly relativistic ``cannonballs'' 
393: that pierce through the supernova shell.  The detailed model is based 
394: essentially on the following analogies, hypotheses and explicit 
395: calculations: 
396:  
397: \subsection{Relativistic jets in astrophysics} 
398:  
399: Astrophysical systems, such as quasars and 
400: microquasars, in which periods of intense accretion onto a compact 
401: massive object occur, emit highly collimated relativistic jets of plasma. 
402: The Lorentz factor $\rm 
403: \gamma\equiv 1/\sqrt{1-v^2/c^2}$ of these jets ranges from mildly 
404: relativistic: $\gamma\sim 2.55$ for GRS 1915+105 (Mirabel \& Rodriguez 
405: 1994, 1999), to quite 
406: relativistic: $\gamma={\cal{O}}\,(10)$ for typical quasars (e.g. Ghisellini 
407: et al.~1993), and even to highly relativistic: $\gamma\sim 10^3$ for PKS 
408: 0405$-$385 (Kedziora-Chudczer et al.~1997). These jets are not continuous 
409: streams of matter, but consist of individual blobs, or ``cannonballs''.  
410: (e.g. Kraft et al. 2001). The mechanism producing these 
411: surprisingly energetic and collimated emissions is not understood, but it 
412: seems to operate pervasively in nature (In Section 7 we propose a 
413: mechanism capable of collimating CBs). We assume the CBs to be composed 
414: of ordinary ``baryonic'' matter (as opposed to $\rm e^+\, e^-$ pairs), as 
415: is the case in the microquasar SS 433, from which Ly$_\alpha$ and metal 
416: K$_\alpha$ lines have been detected (e.g. Margon 1984, Kotani et al.~1996). 
417:  
418: \subsection{The GRB/SN association} 
419: There is mounting evidence 
420: for an association of  supernova 
421: explosions of type Ib/Ic and GRBs (DD2000a). The first example was GRB 
422: 980425 (Soffitta et al.~1998; Kippen et al. 1998), within whose error circle 
423: SN1998bw was soon detected optically  (Galama et al.~1998a) 
424: and at radio frequencies (Kulkarni et al.~1998a). The chance probability 
425: for a spatial and temporal coincidence is less than $10^{-4}$ 
426: (e.g. Galama et al.~1998a), or much smaller if the revised BeppoSAX position 
427: (e.g. Pian et al. 1999) is used in the estimate 
428: (we shall show that the observed X-rays originate in the CBs of this 
429: GRB, and not on the associated SN). The unusual radio 
430: (Kulkarni et al.~1998a; Wieringa et al.~1999) and 
431: optical (Galama et al.~1998a; Iwamoto et al.~1998) properties of SN1998bw 
432: support this association. The exceptionally small fluence 
433: and redshift of GRB 980425 make this event very peculiar, though 
434: not in the CB model (DD2000a). 
435: The energy supply in a SN event similar to SN 1998bw is too small to 
436: accommodate the fluence of cosmological GRBs, unless their $\gamma$-rays 
437: are highly beamed. SN 1998bw is a peculiar supernova, perhaps 
438: because it is observed close to the axis of its GRB emission. 
439:  
440: Evidence for a SN1998bw-like contribution to a GRB afterglow (Dar 1999a; 
441: Castro-Tirado \& Gorosabel 1999) was first found by Bloom et al.~(1999a) for 
442: GRB 980326, but the unknown redshift prevented a quantitative analysis. 
443: The AG of GRB 970228 (located at redshift $\rm z=0.695$) appears to 
444: be overtaken by a light curve akin to that of SN1998bw (located at $\rm 
445: z_{bw}=0.0085$), when properly scaled by their differing redshifts (Dar 
446: 1999b; Reichart 1999; Galama et al.~2000).  Evidence of similar 
447: associations was found for GRB 990712 (Hjorth et al.~2000a; Sahu et 
448: al.~2000;  Bjornsson et al.~2001), GRB 980703 (Holland et al.~2000b), GRB 
449: 000418 
450: (DD2000a), GRB 991208 (Castro-Tirado et al.~2001) and GRB 990510 (Sokolov et 
451: al. 2001a). For the remaining cases in Table I, corresponding to 
452: %the observational data 
453: %preclude a conclusion, for one or more reasons: the late AG has not 
454: %been measured; $\rm F_{bw}[\nu']$ is not known for large $\rm \nu'\simeq 
455: %\nu\,(1+z)$; the GRB's afterglow or the host galaxy are much brighter than 
456: %the SN.  
457: GRBs with larger redshifts, either no late 
458: observations of the AG are available, or the expected 
459: SN bump is an unobservably small effect. These conclusions 
460: will be strengthened by our detailed analysis of AGs 
461: in the CB model. All of the nearest GRBs with 
462: measured redshifts show various degrees of evidence for a 
463: supernova in their AG, suggesting the possibility of an association of 
464: {\it all} the long-duration GRBs with core-collapse SNe. 
465:  
466: \subsection{The SN/GRB association} 
467:  
468: By a SN/GRB association ---as opposed to the GRB/SN association 
469: we have discussed--- we mean the converse statement to that ending 
470: the last subsection: that most SNe of certain relatively 
471: frequent types may be associated with GRBs. This 
472: appears at first sight to be entirely untenable. The total rate of type 
473: II/Ib/Ic SNe has been estimated from their observed rate in the local 
474: Universe (e.g. Van den Bergh \& Tammann 1991) and the star formation rate 
475: as function of redshift, to be $\rm R_{SN}= 12\pm 5\, s^{-1}$, or $\rm 
476: R_{SN}\sim 3.8\times 10^8\, y^{-1}$, 
477: in the observable Universe (Madau et 
478: al. 1998), while the observed rate of GRBs is a mere 
479: $\rm R_{GRB}\simeq 10^3\, y^{-1}$ (see, for example, Lamb 2001). 
480:  
481:  
482: The bolometric energy fluence from a CB 
483: moving with a Lorentz factor $\gamma\gg 1$ and 
484: seen by a stationary observer at an angle $\theta\ll 1$ 
485: relative to the CB's direction of motion  (e.g. DD 2000a) is: 
486: \begin{equation} 
487: \rm  {dF\over d\Omega} \propto 
488: \left[{2\,\gamma \over 1+\gamma^2\,\theta^2}\right]^3\, . 
489: \label{fluence} 
490: \end{equation} 
491: Barring the case of GRB 980425 (whose exceptionality and 
492: interpretation we shall discuss) the equivalent spherical 
493: energies of the GRBs with measured redshifts, as listed in Table I, range 
494: between, approximately, $2 \times 10^{54}$ erg (GRB 990123) and $6.6 \times 
495: 10^{51}$ erg (GRB 970508). The $\theta$ dependence is the steepest 
496: parameter dependence of the CB model (DD2000b). It is therefore 
497: reasonable to attribute the range of observed equivalent spherical energies 
498: mainly to the $\theta$ dependence (as if GRBs were otherwise approximately 
499: standard candles). The observed  spread in 
500: equivalent energy then corresponds, according to Eq.~(\ref{fluence}), 
501: to a spread of 
502: viewing angles between $\theta \approx 0$ and $\theta\approx 2.4/\gamma$. 
503: Thus the geometrical fraction of GRBs which are observable (with the 
504: current or past sensitivity) is approximately $\rm f(\gamma)=2\pi\, \theta^2 / (4 
505: \pi) \approx 2.84 / \gamma^2 $,  where we have taken two jets of 
506: CBs per event. Compare 
507: $\rm R_{SN}$ and $\rm R_{GRB}$ 
508: to conclude that an approximately one-to-one GRB/SN 
509: association would require beaming into a solid angle 
510: that is a fraction $\rm \sim\! 2.8\times 10^{-6}$ of $4\pi$. 
511: For CBs moving with $\gamma\!\sim\! 10^3$, $\rm f(\gamma)=2.84\times 10^{-6}$: 
512: precisely the required beaming factor. That is, 
513: for a one-to-one SN/GRB association: 
514: \begin{equation} 
515: \rm R_{GRB} = f(\gamma)\, R_{SN}= (1082 \pm 450)\, 
516: \left[{10^3\over \gamma}\right]^2 \; y^{-1}\,, 
517: \label{GRB} 
518: \end{equation} 
519: in agreement with observation. 
520: Moreover, if the recent claims that the $\rm\sim (1+z)^3$ dependence of the 
521: star formation rate continues to $\rm z\!>\! 1$ (Fenimore \& Ramirez-Ruiz 2000; 
522: Ramirez-Ruiz and Fenimore 2000; Reichart et al.~2001) 
523: were correct,  SN of types Ib/Ic 
524: would by themselves suffice to explain the observed GRB rate. Thus, 
525: relativistic beaming solves the energy crisis of GRBs and may allow 
526: an approximately one-to-one 
527: SNIc/GRB association (Dar 1999b;  Dar \& Plaga 1999; DD2000a). 
528: The above considerations leading to a GRB/SN association that may 
529: be as biunivocal as indicated by Eq.~(\ref{GRB}) are weakened by 
530: the fact that we have not taken into account effects such as the efficiency of 
531: GRB identification as a function of fluence. It is clear, however, 
532: that for the high beaming factors we have advocated, the GRB and SN 
533: rates are quite comparable. In the CB model, the ``special'' character of  
534: SN 1998bw is due to the fact that it is observed very close to the 
535: GRB axis. 
536:  
537: Previous analyses of GRB/SN associations, except that in DD2000a, 
538: were based on  a power-law extrapolation to late times 
539: of the early-time GRB afterglows.  Here we shall use the CB model, 
540: instead, to calculate the 
541: GRB-afterglow light curves at all times. This procedure leads, as we shall 
542: see, to a much better exposition of the GRB/SN association. 
543:  
544:  
545:  
546: \subsection{The GRB engine} 
547:  
548:  
549: We assume that in  core-collapse SN events 
550: a fraction of the parent star's 
551: material, external to the newly-born compact object, 
552: falls back in a time very roughly of the 
553: order of one day (De R\'ujula 1987, DD2000a). Given the considerable 
554: specific angular momentum of stars, 
555: it settles into an accretion disk and/or torus 
556: around the compact object. 
557: %\footnote{We choose to base our conjectures on analogies with known 
558: %processes, as opposed to computer simulations. The latter do not yet 
559: %realistically include rotation, magnetic fields, the transport of angular 
560: %momentum... Most noticeably, they do not produce SN explosions.}. 
561: The subsequent sudden episodes 
562: of accretion ---occurring with a time sequence that we cannot predict--- 
563: result in the emission of CBs. These emissions last till 
564: the reservoir of accreting matter is exhausted. 
565: The emitted CBs initially expand in the SN rest system at a speed 
566: $\rm\beta\,c/\gamma$, with $\rm\beta\,c$ presumably of the same order as 
567: the speed of sound in a relativistic plasma 
568: ($\beta=1/\sqrt{3}$), or smaller. 
569: The solid angle a CB subtends is so extremely small that presumably 
570: successive CBs do not hit the same point of the outgoing SN shell, 
571: as they catch up with it. These considerations 
572: are illustrated in Fig.~(\ref{model}). 
573:  
574: \subsection{The GRB}From this point onwards, the CB model is not 
575: based on analogies or assumptions, but on processes whose outcome can be 
576: approximately worked out in an explicit manner. The violent collision of 
577: the CB with the SN shell heats the CB (which is not transparent at this point 
578: to $\gamma$'s from $\pi^0$ decays) to a surface temperature that, by the 
579: time the CB reaches the transparent outskirts of the SN shell, is 
580: $\sim 150$ eV, further 
581: decreasing as the CB travels (DD2000b). The resulting CB surface 
582: radiation, Doppler-shifted in energy and forward-collimated by the CB's 
583: fast motion, gives rise to an individual pulse in a GRB (DD2000b). The GRB 
584: light curve is an ensemble of such pulses, often overlapping one another. 
585: The energies of the individual GRB $\gamma$-rays, as well as their typical 
586: total fluences, indicate CB Lorentz factors of ${\cal{O}}$(10$^3$), as the 
587: SN/GRB association does (DD2000a). In the CB model, unlike in 
588: the shocked-fireball models, the photons of the GRB proper 
589: are not made by synchrotron radiation which, as we shall see, 
590:  is subdominant at this stage of the evolution of a CB. 
591:  
592: \section{Afterglow components} 
593:  
594: In the CB model, the persistent radiation in the direction of an observed GRB 
595: has three origins: the ejected CBs, the concomitant SN explosion, and the 
596: host galaxy. These components are usually unresolved in the 
597: measured ``GRB afterglows'', so that the corresponding light curves and 
598: spectra refer to a cumulative energy flux density: 
599: \begin{equation} 
600: \rm    F_{AG}=F_{CBs}+F_{SN}+F_{HG}\, , 
601: \label{sum} 
602: \end{equation} 
603: with $\rm F\equiv\nu\,dN_\gamma/(dt\,d\nu\,dA)$. 
604:  
605: The emission of the GRB's host galaxy is usually determined from 
606: measurements at times late enough for the CB's afterglow and 
607: the SN light to have become comparatively weak (e.g. 
608: Sokolov et al.~2001b and references therein). This assumes that the host 
609: galaxy's emission is steady  on periods of a few months. 
610: There is no indication of GRB host-galaxy variability 
611: on such time scales. 
612:  
613: Core-collapse supernovae (SNII/Ib/Ic) 
614: are far from being standard candles. But if their explosions 
615: are fairly asymmetric ---as they would be if a fair fraction of 
616: them emitted jets of CBs---  much of the variability could be a reflection 
617: of the varying angles from which we see their 
618: non-spherically expanding shells. 
619: Exploiting this possibility to its extreme, we shall use 
620: SN1998bw as an ansatz standard candle, associated with every 
621: GRB of known $\rm z$ (Dar 1999b; DD2000a). The adequacy 
622: of this bold hypothesis can be judged from its rather surprising success. 
623:  
624: Let the energy flux density of SN1998bw be $\rm F_{bw}[\nu,t]$. 
625: For a similar SN placed at a redshift $\rm z$: 
626: \begin{eqnarray} 
627: {\rm F_{SN}[\nu,t] = } && 
628: {\rm{1+z \over 1+z_{bw}}\; 
629: {D_L^2(z_{bw})\over D_L^2(z)}}\, \times\nonumber \\ 
630: &&{\rm F_{bw}\left[\nu\,{1+z \over 1+z_{bw}},t\, 
631: {1+z_{bw} \over 1+z}\right]\; A(\nu,z)}\, , 
632: \label{bw} 
633: \end{eqnarray} 
634: where $\rm D_L(z)$ is the luminosity distance\footnote{The cosmological 
635: parameters we use in our calculations are: 
636: $\rm H_0=65$ km/(s Mpc), ${\rm \Omega_M}=0.3$ and 
637: ${\rm \Omega_\Lambda}=0.7$.} 
638: and $\rm A(\nu,z)$ is the extinction along 
639: the line of sight. The extinction in our Galaxy 
640: is reasonably well measured, but  for the GRBs' environments it must be 
641: estimated from the spectra of each particular AG and host galaxy.  
642:  
643: The contribution of CBs to the GRB afterglows requires 
644:  a much more detailed discussion. 
645:  
646:  
647: \section{Times and frequencies} 
648:  
649: Four ``clocks'' ticking at different paces and three different scales 
650: of frequency need be considered in the cannonball model of 
651: GRBs and their afterglows. 
652:  
653: Let $\rm \gamma=1/\sqrt{1-\beta^2}={E_{CB}/(M_{CB}c^2)}$ be 
654: the Lorentz factor 
655: of a CB, which diminishes with time as the CB hits the SN shell 
656: and as it subsequently ploughs through the interstellar medium. 
657: Let $\rm t_{SN}$ be the 
658: local time in the SN rest system, $\rm t_{CB}$ the time in the CB's 
659: rest system, $\rm t_{Ob}$ the time   measured by 
660: a nearby observer viewing the CB at an angle $\theta$ 
661: away from its direction of motion, and $\rm t$ the time 
662: measured by an earthly observer viewing the CB at 
663: the same angle, but from a ``cosmological'' distance 
664: ($\rm z\neq 0$). 
665: Let x be the distance travelled by the CB in the SN rest system. 
666: The relations between the above quantities are: 
667: \begin{eqnarray} 
668: &&\rm 
669: dt_{SN}=\gamma\,dt_{CB}=\rm{dx\over\beta\, c}\, ; 
670: \nonumber \\ 
671: &&\rm 
672: dt_{CB}=\delta\,dt_{Ob}\, ;\nonumber\\ 
673: &&\rm 
674: dt=(1+z)\,dt_{Ob}={1+z\over \gamma\,\delta}\;dt_{SN}\;, 
675: \label{times} 
676: \end{eqnarray} 
677: where the Doppler factor $\delta$ is: 
678: \begin{equation} 
679: \rm 
680: \delta\equiv\rm{1\over\gamma\,(1-\beta\cos\theta)} 
681: \simeq\rm {2\,\gamma\over (1+\theta^2\gamma^2)}\; , 
682: \label{doppler} 
683: \end{equation} 
684: and its approximate expression is valid for $\theta\ll 1$ and $\gamma\gg 1$, 
685: the domain of interest here. 
686: Notice that for large $\gamma$ and not large $\theta\gamma$, 
687: there is an enormous ``relativistic aberration'': 
688: $\rm dt\sim dt_{SN}/\gamma^2$, and the observer sees 
689: a long CB story as a film in extremely fast motion. 
690:  
691: The frequency of the photons radiated by a CB 
692: in its rest system, $\rm \nu_{CB}$, their frequency 
693: in the direction $\theta$ 
694: in the local SN system, $\rm \nu_{SN}$,  and the photon 
695: frequency $\nu$ measured by a cosmologically distant observer, 
696: are related by: 
697: \begin{equation} 
698: \rm \nu_{CB}=   {\nu_{SN}\over \delta} 
699: \, ;\;\;\;\;\;\nu_{SN}=(1+z)\,\nu\; , 
700: \label{energies} 
701: \end{equation} 
702: with $\delta$ as in Eq.(\ref{doppler}). 
703:  
704:  
705:  
706: \section{The cooling of CBs} 
707:  
708: As a CB pierces through the SN shell, its surface 
709: is heated by the collisions with the shell's constituents, and 
710: %(in the ``surface heating'' model of DD2000b we successfully describe 
711: %the properties of GRBs by assuming a dynamical equilibrium between 
712: %the heat deposited and reemitted within one 
713: %radiation length of the CB's surface). 
714: cools down from an early maximum temperature 
715: because of the decreasing density of the shell's material 
716: it collides with (a detailed description of the CB--shell 
717: collision can be found in DD2001b). At this early point of a CB's avatars, 
718: the internal radiation pressure is very large. Thus, 
719: in studying the properties of the $\gamma$ rays (DD2000b), we assumed 
720: the CBs to expand (in their rest system and at early times) at a speed 
721: comparable to that of sound in a relativistic plasma ($\rm c/\sqrt{3}$). 
722: This fast expansion implies that it is a good approximation 
723: to treat CBs, in their rest system, as spherical objects. 
724:  
725: Let $\rm N_{jet}$ be the baryon or electron number of 
726: the ensemble of CBs in a jet, which we have estimated 
727: to be $\rm N_{jet}\sim 6\times 10^{51}$ (e.g Eq.~(5) of 
728: DD2001b, for $\rm \gamma_{in}=10^4$), which is close to that of the 
729: Earth ($\rm N_\otimes\simeq 3.6\times 10^{51}$). 
730: On average, GRBs consist of 5 to 10 significant pulses, so 
731: that a single CB may have one order of magnitude fewer constituents. 
732: As they exit the shell and enter the interstellar medium (ISM), CBs 
733: become transparent to their enclosed radiation 
734: when they reach a radius: 
735: \begin{equation} 
736: \rm R_{trans}\sim \left[{3\over 4\pi}\,N_{CB}\,\sigma_T\right]^{1/2}\! 
737: \simeq (10^{13}\;cm)\;\left[{N_{CB}\over 6\times 10^{50}}\right]^{1/2}\!\!\! , 
738: \label{Rtrans} 
739: \end{equation} 
740: where $\rm\sigma_T=6.65\times 10^{-25}$ cm$^2$ is the Thomson 
741: cross section. We can use Eqs.~(\ref{times}) to conclude that, if 
742: the CBs are expanding at a fraction $\rm\beta_{trans}$  of the 
743: speed of light\footnote{The quantity $\rm\beta_{trans}$ is nearly 
744: identical to $\rm\beta_{out}$, the transverse speed as the 
745: CB exits the SN shell, introduced in our previous work on the CB model 
746: (e.g. DD200a,b).}, 
747: they reach a size $\rm R_{trans}$ in an observer's time: 
748: \begin{equation} 
749: \rm t_{trans}={1+z\over \delta}\, t_{trans}^{CB}= 
750: {(1+z)\,R_{trans}\over\delta\,\beta_{trans}\,c}. 
751: \label{ttrans} 
752: \end{equation} 
753: For the reference value of $\rm N_{CB}$ in Eq.~(\ref{Rtrans}), 
754: $\gamma=1/\theta=10^3$ and $\rm\beta_{trans}=1/(3\sqrt{3})$, 
755: CBs become transparent in a mere $\rm t_{trans}\sim 3.5$ s. 
756:  
757: The GRB is emitted by the CBs from a distance 
758: of ${\cal{O}}(1)$ radiation length from the 
759: exterior of the SN shell, 
760: when their temperature is $\rm T_\gamma\sim 150$ eV 
761: and their radius, for our typical parameters, is 
762: $\rm R_\gamma\sim 2.5\times 10^{11}$ cm (DD2000b, DD2001b). 
763: Soon thereafter, travelling in a thin environment and expanding fast, the CBs 
764: should cool in an approximately adiabatic way. Their temperature 
765: at $\rm t_{trans}$ is then: 
766: \begin{equation} 
767: \rm T_{trans} \sim {R_\gamma\over R_{trans}}\, T\simeq 4.0\, eV. 
768: \label{TTh} 
769: \end{equation}
770: >From about one third of $\rm t_{trans}$ onwards, the CBs would appear 
771: to be ``collisionless'' to the ISM hadrons piercing through  
772: them\footnote{ISM particles that get entangled in the CB's 
773: magnetic field would not be collisionless after such a very short time.}, 
774: since the high-energy nucleon--nucleon cross section ($\rm\sigma_N\sim 
775: 4\times 10^{-26}$ cm$^2$) is about one order 
776: of magnitude smaller than $\rm \sigma_T$ and the 
777: condition for ``transparency'' to the ISM particles is, up to 
778: a numerical factor of ${\cal{O}}(1)$, analogous to Eq.~(\ref{Rtrans}). 
779:  
780: \section{The expansion of CBs} 
781:  
782: When a CB, in a matter of (observer's) seconds, becomes transparent to 
783: radiation, it loses its internal radiation pressure. If it has been 
784: expanding up to that moment at a speed comparable to that of relativistic 
785: sound, should it not inertially continue to do so? The fact that it is 
786: collisionless makes the conclusion seem unavoidable. But the CBs 
787: emitted by many quasars appear, within the resolution of the 
788: observations, not to expand laterally for most of their trajectory, 
789: before their forward motion nearly stops. What may the reason be? 
790:  
791: The ISM the CBs traverse has been previously partially 
792: ionized by the forward-beamed GRB radiation. The neutral ISM fraction 
793: is efficiently ionized by Coulomb interactions as it enters the CB. 
794: In analogy to processes occurring in quasar and microquasar 
795: ejections, the bulk of the swept-up ionized ISM particles are multiply 
796: scattered, in a ``collisionless'' way, by the CBs' turbulent magnetic fields. 
797: As illustrated in Fig.~(\ref{CBrest}), in 
798: the rest system of the CB these particles are isotropically 
799: re-emitted into the ISM. In the rest system of the parent SN 
800: they are forward collimated and boosted to an energy 
801: $\rm\sim\! m\, c^2\gamma^2$ (Dar 1998b). The isotropic re-emission implies 
802: an inwards force on the CB's surface. Assume that the bulk of the 
803: ISM particles are {\it not} re-energized by the CB's turbulent fields. Let $\rm R$  
804: be the CB's radius and let $\rm n_p$ be the proton ISM number density. The 
805: rate at which the ISM protons impinge on the CB is $\rm 
806: r=\pi\,R^2\,c\,\gamma\,n_p$, with $\rm \gamma\,n_p$ the ISM proton density  
807: seen from the CB's rest system. The momentum (or, for large $\gamma$, the  
808: energy) 
809: of these protons isotropically leaving the CB is, per unit surface, 
810: \begin{equation} 
811: \rm P=r\,{E_p\,c\over 4\pi\,R^2}={1\over 4}\,m_p\gamma^2\,n_p\,c^2\, . 
812: \label{pressure} 
813: \end{equation} 
814:  
815: During the first hours after the GRB time, the CBs are still fully ionized 
816: and cooling rapidly by expansion and 
817: bremsstrahlung (DD2001a). Their full constituency of 
818: relatively cold ions, electrons, cosmic rays and entangled magnetic fields is 
819: electromagnetically coupled, and subject to the very large inwards 
820: pressure of Eq.~(\ref{pressure}). This stabilizes the CB's radius to an 
821: asymptotic value $\rm R_{max}$. To estimate it, since the initial expansion 
822: velocity is not fully relativistic ($\rm \beta_{trans}^2\ll 1$), we may 
823: use Newton's equation: 
824: \begin{equation} 
825: \rm P=-{M_{CB}\over 4\pi\,R^2}\;{d^2 R\over dt^2_{CB}} 
826: \label{Newton} 
827: \end{equation} 
828: with $\rm P$ as in Eq.~(\ref{pressure}), and 
829: integrate, to obtain, for a constant\footnote{A density distribution 
830: falling with distance to the progenitor 
831: star as $\rm x^{-2}$ (in a certain distance-domain) gives 
832: similar results, but in terms of more parameters. It suffices for the moment 
833: to use a constant value representing an average density close 
834: to the progenitor, particularly at the considerable distances from 
835: the progenitor at which the value of $\rm R_{max}$ is reached.} $\rm n_p$: 
836: \begin{equation} 
837: \rm R_{max}^3\sim 
838: R_{trans}^3+{3\,N_{CB}\,\beta_{trans}^2\over 2\,\pi\,n_p\,\gamma_0^2}\, , 
839: \label{Rinfinity} 
840: \end{equation} 
841: where we have approximated $\gamma$ by its initial value 
842: because the asymptotic radius, as we shall see, is reached much 
843: before the CB has had the time to decelerate significantly. 
844: For a value $\rm n_p=1\, cm^{-3}$ 
845: of the ISM density close to the progenitor, $\rm N_{CB}=6\times 10^{50}$, 
846: $\gamma_0=10^3$, Eq.~(\ref{Rinfinity}) gives 
847: $\rm R_{max}=2.2\times 10^{14}$ cm [$10^{14}$ cm] for 
848: $\rm\beta_{trans}=1/(3\sqrt{3})$ [$1/(10\sqrt{3})$]. 
849:  
850: The interval (in the CB's rest system) 
851: between the times when the CB has 
852: radius $\rm R_{trans}$ and radius $\rm R$ can be deduced from 
853: Eqs.~(\ref{pressure}) and (\ref{Newton}) to be: 
854:  
855: \begin{equation} 
856: \rm 
857: t(R)=\int^R_{R_{trans}}\;{dx\over c\,\beta_{trans}}\; 
858: \sqrt{{R_{max}^3-R_{trans}^3\over R_{max}^3-x^3}}\; . 
859: \label{tofR} 
860: \end{equation} 
861: The observer's time is shorter by a factor $\rm (1+z)/\delta$. 
862: In Fig.~(\ref{RCB}) we invert Eq.~(\ref{tofR}) to 
863: show the CB's radius as a function of 
864: observer's time (in minutes), for $\rm z=1$, $\delta=10^3$, and the other 
865: typical parameters quoted in the previous 
866: paragraph, for two choices of $\rm\beta_{trans}$. 
867: The CB is seen to expand linearly 
868: at a speed close to the initial $\rm \beta_{trans}\,c$, and then 
869: to settle fast into an approximately constant radius. 
870: To a good approximation the steady radius is reached 
871: in an observer's time: 
872:  
873: \begin{equation} 
874: \rm 
875: t_\infty\sim {1+z\over \delta}\;{R_{max}\over \beta_{trans}\,c}\, , 
876: \label{tapprox} 
877: \end{equation} 
878: which yields $\sim 1.2\;[1.9]$ minutes for 
879: $\rm \beta_{trans}=1/(3\sqrt{3})$ [$1/(10\sqrt{3})$], 
880: the examples in Fig.~(\ref{RCB}). 
881: Thus, typically, a few observer's minutes after the GRB, the CBs 
882: are expanding very very slowly. 
883:  
884: To estimate the internal magnetic field of the CB after it stopped 
885: expanding, $\rm B_\infty\, , $ we conjecture that the bulk of the kinetic  
886: energy of the CB's expansion after it becomes transparent is converted 
887: to internal magnetic energy. Since $\rm R_{max}^3 \gg R_{trans}^3$ and 
888: (by hypothesis) $\rm\beta_{trans}^2 \ll 1$, this means: 
889: \begin{equation} 
890: \rm {4\, \pi\over 3}\;  R_{max}^3\;  {B_{\infty}^2 \over 8\, \pi} 
891:         \sim {N_{CB}\, m_p\, \beta_{trans}^2\, c^2\over 2}  \, , 
892: \label{Bequiv} 
893: \end{equation} 
894: which, upon substitution of Eq.~(\ref{Rinfinity}), yields: 
895: \begin{equation} 
896: \rm B_{\infty } 
897: \sim 100\, 
898: \left[{n_p\over 1\, cm^{-3}}\right]^{1/2}\, 
899: \left[{\gamma\over 10^3}\right]\; \, Gauss\; . 
900: \label{Binfty} 
901: \end{equation} 
902: The very large field of Eq.~(\ref{Binfty}) is consistent with the fact 
903: that, in order to be able to sustain the inwards pressure of the isotropically 
904: re-emitted protons that it ejects, 
905: the magnetic field within the CB must have a pressure (or energy 
906: density) comparable to the pressure $\rm P$ of Eq.~(\ref{pressure}). 
907: The condition $\rm P=B^2/(8\pi)$ exactly reproduces Eq.~(\ref{Binfty}). 
908: We could also have added the building-up magnetic pressure 
909: to Eq.~(\ref{pressure}), to obtain a result for $\rm R_{max}$ differing from that 
910: of Eq.~(\ref{Binfty}) by $2^{1/3}$. This affects our conclusions insignificantly 
911: and, in any case, we cannot pretend to have a detailed understanding 
912: of the magnetohydrodynamics of turbulent plasmas. 
913:  
914: We have argued that the re-emission of the ISM 
915: protons, isotropic in the CB's rest frame, is what makes 
916: CBs stop expanding  at a speed that {\it ab-initio} was semirelativistic. 
917: This may also explain the surprising quasar and microquasar observations. 
918: There may be other reasons ---such as Coulomb-interaction ram pressure 
919: from the ambient material--- for CBs to stop expanding significantly at 
920: some point of their voyage. The strength of our conclusion that CBs 
921: expand slowly ---or not at all--- during the AG phase should be judged 
922: from the ability of the CB model to describe the AG observations. 
923:  
924: \section{The dominant afterglow mechanisms} 
925:  
926:  
927: In the CB model the GRB emission in $\gamma$ rays is mainly of 
928: thermal origin (although it does not have a thermal spectrum) 
929: and, in a fixed energy interval, it decreases exponentially 
930: with time (DD2000b). A few seconds after the last GRB pulse 
931: (the last CB), this pseudothermal emission becomes a 
932: subdominant effect. For the next few hours, the evolution 
933: of a CB is interestingly complicated. In particular, its originally 
934: ionized material should recombine into  hydrogen and 
935: emit Lyman-$\alpha$ lines that are seen Doppler-boosted 
936: to keV energies (DD2001a). Later, the 
937: CBs settle down to a much simpler phase, which typically lasts 
938: for months, till the CBs finally stop moving relativistically. 
939:  
940: Because of the CBs' large Doppler factors, radio emission in their rest 
941: frame is boosted to optical light in the observer's frame while their 
942: emitted optical light is boosted to the soft X-ray band. Radio emission 
943: from astrophysical plasmas at eV temperatures 
944: is mainly due to synchrotron radiation from 
945: relativistic electrons, whereas their optical glow is 
946: usually due to bremsstrahlung and line emission. For parameters in the 
947: general vicinity of the ones we have argued to be ``typical'' of the CB 
948: model, the X-ray AG is initially dominated by thermal 
949: bremsstrahlung (and line emission) and by synchrotron radiation 
950: thereafter, while the optical AG, generally observed later, 
951: is dominated by synchrotron 
952: radiation.  In this section we analyse these two dominant mechanisms, 
953: relegating to Appendix 1 the discussion of various subdominant ones. 
954:  
955: \section{Thermal bremsstrahlung: the early X-ray AG} 
956:  
957: When it becomes transparent, a CB cools down mainly by thermal-electron 
958: bremsstrahlung (TB) in $\rm e\,p$ collisions and by expansion. The comoving 
959: TB emission  rate (e.g. Peebles 1993) is: 
960: \begin{equation} 
961: \rm L_{brem}\simeq \eta\, {\overline n_e}^2\, T^{1/2} 
962: \, erg\, cm^3\, s^{-1}\, , 
963: \label{Lbrem} 
964: \end{equation} 
965: where $\rm \eta = 1.435\times 10^{-27}$, 
966: $\rm \overline n_e$ is the electron density 
967: of the CB and, here and in the rest of this section, 
968: $\rm T$ is its temperature in Kelvin and the remaining quantities 
969: are in c.g.s. units. Thus, as long as the CB is fully ionized, 
970: its total comoving TB energy-loss rate is: 
971: \begin{equation} 
972: \rm {dE_{CB}\over dt_{CB}} \simeq - \eta \, {3\, N_{CB}^2 
973: \, T^{1/2}\over 4\, \pi\, R^3} \, erg\,s^{-1}\, , 
974: \label{CBbrem} 
975: \end{equation} 
976: while its thermal energy is: 
977: \begin{equation} 
978: \rm E_{CB} \simeq 3\, N_{CB}\, k\, T\, , 
979: \label{Etotal} 
980: \end{equation} 
981: with $\rm k=1.38\times 10^{-16}\, erg/deg$ Boltzmann's constant. 
982: As long as TB dominates the cooling and for 
983: a constant expansion rate 
984: $\rm R(t_{CB})\sim  \beta_{trans}\, c\, t_{CB}$, 
985: Eqs.~(\ref{Rtrans}, \ref{ttrans}, \ref{CBbrem}, \ref{Etotal}) yield 
986:  $\rm T$ for $\rm t_{CB}\geq t_{trans}^{CB}$: 
987: \begin{eqnarray} 
988: \rm T^{1/2}(t_{CB})&\simeq&\rm 
989:  K+{\eta\, N_{CB} \over 16\, \pi\, k\, \beta_{trans}\, c\, R^2}\, 
990: \nonumber\\ 
991: \rm K &=& \rm T_{trans}^{1/2} - {\eta \over 12\, k\, \beta_{trans}\, 
992:                  c\, \sigma_T}\, . 
993: \label{Tbrem} 
994: \end{eqnarray} 
995:  
996: A distant observer at an angle 
997: $\theta$ relative to the CB's direction of motion receives this radiation 
998: in a Doppler-boosted, collimated and time-aberrant form 
999: (e.g. DD2000a). At a luminosity distance 
1000: $\rm D_L(z)$, the total power (integrated over frequencies) 
1001: per unit area is: 
1002: \begin{equation} 
1003: \rm {dF\over dt\, d\Omega}  \simeq {3\, \eta\,  N_{CB}^2\, T^{1/2}\, 
1004: \delta_0^4 \over 16\, \pi^2\, R^3\,  D_L^2}\,  erg\, cm^{-2}\, s^{-1}\, , 
1005: \label{fbrem} 
1006: \end{equation} 
1007: where $\delta_0=\delta[\gamma_0,\theta]$ is given by Eq.~(\ref{doppler}) 
1008: with the initial $\gamma$ value, which does not change in the 
1009: very short time during which TB dominates the X-ray AG. At 
1010: the transparency radius and reference baryon number of Eq.~(\ref{Rtrans}), 
1011: the temperature is $\sim 4.0$ eV $\simeq 4.6\times 10^4$ K, 
1012: as in Eq.~(\ref{TTh}), and for 
1013: $\delta=10^3$ and $\rm z=1$, the predicted energy flux, per CB, is 
1014: $\rm 1.7\times 10^{-8}\; erg\, cm^{-2}\, s^{-1}$. 
1015: This radiation's spectrum  is that of bremsstrahlung 
1016: (a flat $\rm E\,dn_\gamma/dE$), typically extending 
1017: to  $\rm E\sim 3\,T \,\delta/(1+z)\sim 12$ keV, in the X-ray domain. 
1018:  
1019: For our typical CB parameters, synchrotron emission takes over, as we 
1020: shall see in the next section, 
1021:  before the CB reaches its asymptotic radius and before the $\rm K$ term 
1022: in Eq.~(\ref{Tbrem}) becomes important. During this early phase 
1023: (for $\rm t\!\sim\! t_{trans}$), the TB emission by the CB's electrons 
1024: declines with time, if cooling is dominated by bremsstrahlung losses, as 
1025: $\rm R^{-3}\,T^{1/2}\!\sim\!R^{-5}\!\sim t^{-5}$.  
1026: Should the temperature decrease be dominated by adiabatic cooling 
1027: ($\rm 3\,N_{CB}\,k\,dT=-P\, dV$), the substitution 
1028: $\rm P=2\,\overline n_e\,k\,T$ results in $\rm T\propto R^{-2}$ and in 
1029: a TB emission declining as $\rm R^{-3}\,T^{1/2}\!\sim\!R^{-4}\!\sim t^{-4}$. 
1030: A decline as fast as $\rm t^{-4}$ or $\rm t^{-5}$ 
1031: has been observed in early X-ray afterglow observations, 
1032: e.g. GRB 920723: Burenin et al.~1999; GRB 970508: Piro et al. 1998;  GRB 
1033: 970828: Smith et al.~2001; GRB 990510:  Pian et al.~2001; GRB 010222: In 't 
1034: Zand et al.~2001. In less than a minute of observer's time, this emission 
1035: mechanism is overtaken by synchrotron radiation, whose decline is much 
1036: slower, as we proceed to discuss. 
1037:  
1038: \section{Synchrotron radiation: the optical afterglow} 
1039:  
1040: In this section we study the various effects resulting from 
1041: the interaction of a CB's entangled magnetic field with the 
1042: ISM particles that it sweeps as it travels. 
1043: The CBs lose momentum by sweeping the nuclei of the ionized ISM, and 
1044: re-emitting them isotropically (in the CB's rest system) at an energy 
1045: comparable to their incoming one. This allows us to predict the 
1046: law of CB deceleration: the behaviour of its decreasing 
1047: Lorentz factor $\rm\gamma(t)$. The incoming electrons, suffering 
1048: collisions with the magnetic domains, and losing energy effectively 
1049: by synchrotron radiation, acquire a predictable power-law energy 
1050: spectrum, which implies a given distribution of the emitted photons. 
1051: The emitted energy rate is equal to the rate at which the ISM electrons 
1052: bring energy into the CB in its rest system; this provides the 
1053: absolute normalization of the AG light curve. The dynamical time 
1054: for the energy supply by the swept-up ISM electrons is much longer than  
1055: the time it takes the electrons to acquire a power-law energy 
1056: distribution and to emit synchrotron radiation, justifying a 
1057: quasi-steady-state analysis. 
1058:  
1059: The rate at which the energy of the ISM electrons enters the CB 
1060: (in its rest frame) is: 
1061: \begin{equation} 
1062: \rm {dE_{CB}\over dt_{CB}}\simeq \pi \, R^2\, n_e\, m_e\, c^3\,\gamma^2\, , 
1063: \label{depo} 
1064: \end{equation} 
1065: where the incident 
1066: ISM electron energy is $\rm \gamma \, m_e\, c^2$, 
1067: and the extra power of $\gamma$ originates in the Lorentz 
1068: contraction of the ISM electron density,  $\rm n_e$. 
1069: An observer at a luminosity distance $\rm D_L$ and at an angle 
1070: $\theta$ relative to the CB's direction of motion receives a total 
1071: power per unit area (integrated over frequencies): 
1072: \begin{equation} 
1073: \rm {dF\over dt\, d\Omega} \simeq  {\pi \, R^2\, n_e\, m_e\,c^3\, 
1074: \gamma^2\,\delta^4  \over 4\, \pi\, D_L^2}\, . 
1075: \label{flux} 
1076: \end{equation} 
1077: At the very early time of CB transparency, for $\rm R=R_{trans}=10^{13}$ 
1078: cm, $\rm n_e=1$ cm$^{-3}$, 
1079: and for $\rm z=1$, $\gamma=\delta=10^3$, the above expression yields 
1080: $1.27\, \times 10^{-9}$ erg/(cm$^2$ s), which 
1081: is comparable to the bremsstrahlung emission of 
1082: Eq.~(\ref{CBbrem}). But, as we shall see anon, the synchrotron 
1083: radiation has a much softer spectrum than that of bremsstrahlung, and 
1084: the latter mechanism dominates at early times in the X-ray domain. 
1085: We have seen that in a matter of minutes the 
1086: asymptotic radius $\rm R_{max}$ of Eq.~(\ref{Rinfinity}) is reached, so 
1087: that thermal bremsstrahlung has decreased by more than three orders of 
1088: magnitude, while the synchrotron radiation has increased by two or more 
1089: orders of magnitude,  to become the dominant emission mechanism 
1090: at all frequencies. 
1091:  
1092: To estimate a ``dynamical time'', 
1093: $\rm \tau_{dyn}$, for the energy deposited by electrons in the CB, 
1094: we forestall that the observed afterglow fluences are the 
1095: ones expected in the CB model. We can then use Eqs.~(\ref{depo}) 
1096: and (\ref{flux}) to deduce a total typical $\rm E_{CB}\sim 3\times 10^{44}$ 
1097: erg and to conclude: 
1098: \begin{equation} 
1099: \rm \tau_{dyn}\equiv \left[{1\over E_{CB}}\,{dE_{CB}\over dt_{CB}}\right]^{-1} 
1100: \sim (8.2\times 10^7\; s)\;\left[{10^3\over\gamma}\right]^2, 
1101: \label{dyn} 
1102: \end{equation} 
1103: for $\rm R=R_{max}=2.2\times 10^{14}$ cm and $\rm n_e=10^{-3}$ cm$^{-3}$. 
1104:  
1105: In the CB's rest frame, the light crossing time for the asymptotic 
1106: radius $\rm R_{max}$ is $\rm\tau_{cr}\sim R_{max}/c\sim 10^4$ s. 
1107: The time for electrons that enter a CB to redistribute their energy 
1108: as they bounce off a few magnetic ``sub-domains'' is a fraction of $\rm\tau_{cr}$; this is much shorter  
1109: than $\rm \tau_{dyn}$, so that a cosmic-ray-like ``source'' 
1110: distribution of electrons (a power law in energy) is steadily generated. 
1111: The synchrotron cooling time of electrons in the CB's rest system is : 
1112: \begin{equation} 
1113: \rm \tau_{syn} \simeq {6\, \pi\, m_e\, c^2 \over \gamma_e\, c\, \sigma_T\, 
1114: B^2} \sim (80\; s)\;\left[{10^3\over\gamma_e}\right]\; 
1115: \left[{100\;\; Gauss\over B}\right]^2. 
1116: \label{tsyn} 
1117: \end{equation} 
1118: In the above equation, we have distinguished 
1119: $\rm\gamma_e$ (the Lorentz factor of an electron in the CB, 
1120: in the CB's rest system) from $\gamma$ (the CB's bulk-motion 
1121: Lorentz factor). Even for $\rm \gamma_e\to 1$, $\rm \tau_{dyn}\gg\tau_{syn}$. 
1122: The Larmor radius of electrons $\rm r_L=p_e/(e\,B)$ in a $\rm B=100$ 
1123: Gauss magnetic field is $\rm r_L\sim (1.5\times 10^5\; cm)(\gamma_e/10^3)$, 
1124: so that even for very high energies, 
1125: the residence time of the electrons in the CB 
1126: is much longer than $\rm\tau_{syn}$. The above inequalities imply that for 
1127: electrons of all energies, a spectrum of 
1128: ``Fermi-accelerated'', radiation-loss-modulated electrons is 
1129: steadily generated. 
1130:  
1131: The incoming ISM particles are Fermi-accelerated 
1132: by the turbulent magnetic fields inside the CBs to 
1133: a comoving ``cosmic-ray'' spectral distribution $\rm dn/dE\sim 
1134: E^{-\beta_p}$, with $\rm \beta_p\simeq 2.2$, as indicated 
1135: by simulations (Bendarz \& Ostrowski 1998), analytical estimates 
1136: and the interpretation of the 
1137: observations of cosmic rays (e.g. Dar \& Plaga 1999). 
1138: The acceleration being due to deflections by magnetic 
1139: fields, the spectral shape of the ``source'' distributions of protons 
1140: and electrons ought to be the same: $\rm \beta_e\simeq\beta_p$. 
1141: The index of the equilibrium electron spectrum, modulated by radiation 
1142: losses, is one unit higher: 
1143: $\rm \beta_e\simeq\beta_p+1\approx 3.2$ (Dar \& De R\'ujula 2001). 
1144: The emitted synchrotron radiation has a 
1145: power spectrum with index $\rm\alpha=(\beta_e-1)/2$, so that: 
1146: \begin{eqnarray} 
1147: \rm {dF \over d\nu_{_{CB}}}&\equiv&\rm 
1148:  \nu_{_{CB}}\,{dn_\gamma \over d\nu_{_{CB}}}\propto  
1149: \nu_{_{CB}}^{-\alpha}\nonumber\\ 
1150: \rm \alpha & \approx & \rm 1.1\; . 
1151: \label{spectrum} 
1152: \end{eqnarray} 
1153: The spectrum of Eq.~(\ref{spectrum}) should roughly 
1154: extend between two cutoff frequencies, $\rm \nu_{min}$ 
1155: and $\rm \nu_{max}$, that reflect the energy of the incoming electrons 
1156: and of the maximally accelerated ones. 
1157: The integrated spectrum of Eq.~(\ref{flux}) 
1158: is proportional to $\rm \nu_{min}^{1-\alpha}$ 
1159: and $\rm \nu_{min}\propto \gamma_c^2$ where $\rm \gamma_c$ is the electrons' 
1160: Lorentz factor above which they are in radiative equilibrium. 
1161: Since  the individual frequencies $\nu$ and the limiting frequency 
1162: $\rm\nu_{min}$ all refer to the CB's rest frame, and 
1163: are Doppler-shifted by its motion ($\rm \nu\propto \delta\; \nu_{CB}$) 
1164: as in Eqs.~(\ref{energies}), the non-frequency-integrated version 
1165: of Eq.~(\ref{flux}) ---that is, the 
1166: predicted spectral energy density for a GRB with 
1167: a number $\rm n_{_{CB}}$ of CBs--- is: 
1168: \begin{eqnarray} 
1169: \rm 
1170: F_\nu&\equiv& \rm{dF[\nu,t,\theta]\over dt\,d\nu\,d\Omega} \simeq 
1171: \rm  n_{_{CB}}\,(\alpha-1)\,\pi\, 
1172: [1+z]^{(1-\alpha)}\,m_e\,c^3\nonumber\\ 
1173: &\times&\rm {n_e(x[t])\;[R(t)]^2\,[\gamma(t)]^{2\,\alpha} 
1174: \over 4\,\pi\,[D_L(z)]^2\,\nu_c}\, 
1175: \left[{2\gamma(t)\over 1+[\theta\,\gamma(t)]^2}\right]^{n} 
1176: \,\left[{\nu\over \nu_c}\right]^{-\alpha};\nonumber\\ 
1177: \rm n&\equiv& 3+\alpha\simeq 4.1\nonumber\\ 
1178: \rm\nu_c&\sim&\rm {3 \,e\, B_\infty  \over 4 \pi \, m_e\, c} 
1179: \simeq 0.42\,{B_\infty\over 100\:Gauss} \;GHz, 
1180: \label{fluxdensity} 
1181: \end{eqnarray} 
1182: where we used the explicit form of $\delta$, Eq.~(\ref{doppler}), 
1183: $\rm n_e(x[t])$ is the density along the CB's trajectory 
1184:  $\rm x = \int \gamma \delta c \, dt/(1+z)$, 
1185: and the overall normalization is obtained by assuming 
1186: that relativistic electrons are in radiative equilibrium. 
1187: The predicted normalization for a GRB 
1188: is just an estimate, for  part of the energy 
1189: deposited in the CB by ISM protons, as well as a fraction 
1190: of its magnetic energy, may also be emitted as synchrotron 
1191: radiation\footnote{At very low frequencies, such as those 
1192: corresponding to radio waves in the observer's frame, 
1193: Eq.~(\ref{fluxdensity}) is expected to break down for a variety 
1194: of reasons: a deviation of the low-energy electron spectrum 
1195: from a universal power-law, inverse synchrotron and inverse 
1196: bremsstrhalung self-absorption, plasma frequency cutoff  
1197: and the effect of competing mechanisms other 
1198: than synchrotron radiation which all depend on the exact density profile  
1199: and ionization state of the CB.}. 
1200:  
1201: When a ``typical'' CB, within a minute or two after the end of its GRB, 
1202: reaches its final radius $\rm R_{max}\sim 2\times 10^{14}$ cm, and for 
1203: $\rm n_e=1$ cm$^{-3}$, $\rm z=1$, and $\gamma=\delta=10^3$, 
1204: Eq.~(\ref{fluxdensity}) yields $1.6\, \times 10^{-8}$ erg/(cm$^2$ s), 
1205: of which $6.4\, \times 10^{-10}$ erg/(cm$^2$ s) ($\sim\! 3.7\%$) is in 
1206: the 2--10 keV X-ray range and  $5.6\, \times 10^{-9}$ 
1207: erg/(cm$^2$ s) ($\sim\! 3.2\%$) 
1208: is in the visible range ($\rm 3900$ \AA $\leq \lambda\leq 7600$ 
1209: \AA)  corresponding to a spectral flux density of 2 Jansky (8 
1210: magnitude!) in the R band. 
1211: For the next few hours $\rm\gamma(t)$ 
1212: does not change significantly and the X-ray and optical 
1213: AGs vary as $\rm n_e(x[t])$. This variation should 
1214: in general be a decline, since the CBs are departing from a dense region.   
1215: Such a decline  may 
1216: have been observed both in the optical band in the 
1217: case of GRB 990123 (Akerlof et al.~1999), which rose above $\sim 9$th 
1218: magnitude tens of seconds after the GRB's onset, not far 
1219: from our estimate with ``typical parameters''  (this GRB, whose early 
1220: optical AG we shall discuss in detail, is at 
1221: $\rm z=1.6$, but its initial $\gamma$ and $\delta$ are very large, 
1222: see Table III). 
1223:  
1224: There may be an energy below which the synchrotron cooling time 
1225: of the electrons in the CB is longer than their acceleration time. 
1226: If so, the electron spectrum has an index $\rm \beta_e\simeq 
1227: \beta_p$ and the synchrotron radiation below a certain frequency 
1228: would have an index $\alpha\simeq 0.6$. We have implicitey assumed 
1229: that this frequency is below the smallest optically observed ones, 
1230: an assumption that the data generally  support.  
1231:  
1232: The flux of Eq.~(\ref{fluxdensity}) depends on $\rm \nu_c$
1233: as $\rm \nu_c^{\alpha-1}$, roughly the 10-th root of it.
1234: The actual value of $\rm \nu_c$ is therefore quite irrelevant
1235: to the optical and X-ray AGs discussed here. In our study of
1236: radio AGs (Dado et al. 2002) we find that, in the CB model,
1237: $\rm\nu_c$ is actually the characteristic synchrotron frequency
1238: emitted by the electrons that enter the CB with a Lorentz
1239: factor (in the CB's rest frame) $\rm \gamma_e=\gamma(t)$,
1240: that is $\rm \nu_c\sim 0.22\, \gamma_e^2\,\nu_L$, with 
1241: $\rm \nu_L\propto B\propto \gamma$ the Larmor radius in the 
1242: CB's magnetic field. The spectral index gradually changes from
1243: $\alpha\approx 0.5$ to $\alpha\approx 1.1$ at this frequency.
1244: For a spectrum with this transition in its 
1245: power law, Eqs.~(\ref{fluxdensity}) are to be modified as follows: 
1246: \begin{eqnarray} 
1247: \rm [\gamma(t)]^{2\alpha} &\rightarrow&\rm   [\gamma(t)]^2\nonumber\\ 
1248: \rm (\alpha-1) &\rightarrow&\rm {(\beta_p-2)(3-\beta_p)\over 2}\nonumber\\ 
1249: \rm \alpha & \approx & \rm 0.5~~if~~(1+z) \nu\leq \delta \,\nu_c \; ,\nonumber\\ 
1250: \rm \alpha & \approx & \rm 1.1~~if~~ (1+z)\nu\geq \delta \,\nu_c. 
1251: \label{modif} 
1252: \end{eqnarray} 
1253: The difference between Eqs.~(\ref{fluxdensity}) and (\ref{modif})
1254: is only relevant to the spectral shape of some very early optical AGs, 
1255: and to the ensemble of the radio AGs (Dado et al. 2002).
1256: 
1257: 
1258:  
1259: \subsection{The density of the Inter-Stellar Medium} 
1260:  
1261: The density of ISM protons very close to a GRB progenitor 
1262: plays a role in determining the asymptotic radius, $\rm R_{max}$, 
1263: of a CB, see Eq.~(\ref{Rinfinity}). The density of ISM protons 
1264: along a CB's trajectory controls, as we shall see, the evolution 
1265: in time of the CB's  Lorentz factor $\rm \gamma(t)$. 
1266: This function, and the density of interstellar electrons along 
1267: the CB's trajectory, determine, via Eq.~(\ref{fluxdensity}), the 
1268: AG properties. Clearly, we must discuss these densities in some detail. 
1269:  
1270: An analysis of historical SNe in the Galaxy (e.g. Higdon and 
1271: Lingenfelter, 
1272: 1980), of SNe in the LMC (e.g., Dune et al. 2001) and of SNe in 
1273: late-type 
1274: galaxies (Kennicut et al. 1989; van Dyke et al. 1996; 
1275: Higdon et al. 1998) indicates that $85\pm 10$\% of SNe occur in 
1276: {\it superbubbles} (e.g., Lingenfelter et al. 2001). 
1277: These are spaces 
1278: of typical size 0.1 to 0.5 kpc, surrounding star-formation regions, 
1279: that extend all the way into the galactic halo, and 
1280: from which the ISM has been swept away by massive-star winds and 
1281: previous SNe, resulting in an ISM with a low density  ($\rm n\sim 10^{-2}$ 
1282: to $10^{-3}$ cm$^{-3}$) comparable to that in a galactic halo. 
1283:  
1284: At their Wolf-Rayet phase, 
1285: massive stars that finally produce SNeII/Ib/Ic 
1286: emit strong winds with typical velocities, 
1287: $\rm v$,  close to $\rm 10^{3}\, km\, s^{-1}$, at a typical mass-loss rate 
1288: $\rm \dot{M}=$ a few $\rm 10^{-4}\, M_\odot\, yr^{-1}$, over the last 
1289: $\!\sim\! 10^5$ yr before the SN event. The density close to an 
1290: imminent SN is governed by the recent Wolf-Rayet wind and ejections, 
1291: and declines roughly quadratically with distance as:  
1292: \begin{eqnarray} 
1293: \rm n&\!\sim\!& \rm {\dot M \over 4\, \pi\, v\, x^2} \\ 
1294:     & \!\approx\! &\rm(0.18\,cm^{-3})\left[{\dot{M}\over 10^{-4}\,M_\odot\, yr^{-1}} 
1295: \right] 
1296: \left[{10^3\, km\, s^{-1}\over v}\right] \left[{1\,pc\over x}\right]^2\! , 
1297: \nonumber 
1298: \label{WRwind} 
1299: \end{eqnarray} 
1300: till $\rm x\!\sim\! 10$ pc, where the density becomes that of the surrounding 
1301: superbubble (e.g., Ramirez-Ruiz et al. 2000 
1302: and references therein). 
1303:  
1304: \subsection{Complications simplified} 
1305: To predict the explicit 
1306: time-dependence of the AG from a CB, one needs to know $\rm R(t)$, 
1307: $\rm \gamma(t)$ and the ISM density profile, $\rm n_e(x)$, along 
1308: the CB's trajectory. 
1309: Moreover, the various CBs that produce the different pulses in a 
1310: single GRB have slightly different physical parameters (baryon 
1311: number, Lorentz boost) that lead to the differences 
1312: between the individual $\gamma$-ray pulses of a given GRB. 
1313: The large powers of the Lorentz and Doppler factors 
1314: in Eq.~(\ref{fluxdensity}) favour the contribution of CBs with the 
1315: largest $\gamma$.  Given the extremely small fraction of solid angle that 
1316: a CB spans as viewed from the SN centre, we do not expect consecutive CBs to 
1317: hit the SN shell on the same spot (DD2000a). But it is in principle 
1318: possible that the initial expansion and slowing down of the CBs by the SN shell  
1319: and the ISM merges several of them into a single leading CB in the AG phase. 
1320: One seems to be faced with a plethora of parameters and possibilities. 
1321:  
1322: We shall find it sufficient to characterize the 
1323: various ISM densities that the CB encounters 
1324: by two constant densities. One of them is the average proton density very close 
1325: to the parent star, that determines the fast-reached 
1326: asymptotic radius of the CB. For its reference value 
1327: we adopt $\rm n_p^{SN}=1$ cm$^{-3}$. The other is the proton or electron 
1328: density in the superbubble and in the galactic halo, for which we 
1329: adopt as reference $\rm n=10^{-3}$ cm$^{-3}$. 
1330:  
1331: All other putative complications  previously quoted are 
1332: eased by the fact that the times over which AGs extend are 
1333: much longer than the typical intervals between GRB pulses, 
1334: so that the AG light curve is the sum of temporally unresolved 
1335: individual CB afterglows. We can therefore characterize, 
1336: as in Eq.~(\ref{fluxdensity}), the 
1337: AG with the parameters of a single CB, whose actual values 
1338: would represent a weighted average. 
1339:   
1340: \subsection{The slow down of a CB} 
1341:  
1342:  
1343: A CB ionizing and ploughing through 
1344: an ionized ISM of roughly constant density, would 
1345: lose momentum at a roughly constant rate, independent of whether the ISM 
1346: constituents are rescattered isotropically in the CB's rest frame, or 
1347: their mass is added to that of the CB. 
1348: Energy-momentum conservation for a highly 
1349: relativistic CB of initial mass $\rm M_{CB}\simeq N_{CB}\,m_p$ 
1350: results in the deceleration law (DD2000a): 
1351: \begin{equation} 
1352: \rm d\gamma=-{\pi\,R^2\,n_p\,\gamma^2\over N_{CB}}\, dx \, . 
1353: \label{dgamma} 
1354: \end{equation} 
1355: The element $\rm dx$ of travelled distance (for $\gamma\gg  1$) is, 
1356: according to Eqs.~(\ref{times}), related to the observer's 
1357: time interval $\rm dt$ as: 
1358: \begin{equation} 
1359: \rm dx={c\,\gamma\,\delta\,dt\over(1+z)}\, . 
1360: \label{dxsn} 
1361: \end{equation} 
1362: The law governing the CB's expansion rate is the differential 
1363: version of Eq.~(\ref{tofR}), to wit: 
1364: \begin{equation} 
1365: \rm dR=c\,\beta_{trans}\, 
1366: \sqrt{{R_{max}^3-R^3\over R_{max}^3-R_{trans}^3}}\;dt. 
1367: \label{Roft} 
1368: \end{equation} 
1369: The above set of three equations can be integrated 
1370: numerically for any given $\rm \gamma_0=\gamma(t=0)$, 
1371: $\rm R(t=0)=R_{trans}$ and ISM density along the CB's path 
1372: $\rm n_p(x)$. 
1373:  
1374: We limit our discussion to the case of CBs that, having reached 
1375: their asymptotic radius, are moving 
1376: through a constant-density medium, a case for which there are useful 
1377: analytical expressions for the solution $\rm \gamma(t)$ 
1378: of Eqs.~(\ref{dgamma}), (\ref{dxsn}) and (\ref{Roft}).  
1379: A constant density is a fair approximation, 
1380: for, in a very short observer's time,  the CB reaches the distance 
1381: from the SN at which the density is that of the surrounding 
1382: superbubble. To estimate this brief time, we may use the initial 
1383: Lorentz and Doppler factors to obtain: 
1384: \begin{equation} 
1385: \rm t\sim{(1+z)\over c\,\gamma_0\,\delta_0}\, x\sim (34\, min)\, 
1386:  \left[{x\over 10\;pc}\right]\, , 
1387: \label{tsn} 
1388: \end{equation} 
1389: where we used the typical parameters $\rm z=1$, $\gamma_0=10^3$ and 
1390: $\theta=1/\gamma_0$. 
1391: This constant density approximation is also justified 
1392: a posteriori by the agreement between our predicted AG 
1393: light curves and the observed ones. 
1394:  
1395: We have argued that CBs reach a steady radius $\rm R_{max}$ 
1396: in a few observer's minutes. To ascertain the 
1397: CB's slow-down law for constant radius and constant ISM density, we 
1398: may substitute Eqs.~(\ref{doppler}) and (\ref{dxsn}) into 
1399: Eq.~(\ref{dgamma}) and integrate, to obtain: 
1400: \begin{eqnarray} 
1401: \rm 
1402: &&{1\over\gamma^3}-\rm{1\over\gamma_0^3} 
1403: +3\,\theta^2\,\left[{1\over\gamma}-{1\over\gamma_0}\right]= 
1404: {6\,c\, t\over (1+z)\, x_\infty}\nonumber\\ 
1405: &&\rm x_\infty\equiv{N_{CB}\over\pi\, R_{max}^2\, n_p}\simeq (1.3\;Mpc)\times 
1406: \label{gamoft}\\ 
1407: &&\rm 
1408: \left[{N_{CB}\over 6\!\times\! 10^{50}}\right]^{1\over 3} 
1409: \left[{10^{-3}cm^{-3}\over n_p}\right]\, 
1410: \left[{n_p^{SN}\over 1\,cm^{-3}}\right]^{2\over 3} 
1411: \left[{\gamma_0\over 10^3}{1/(3\sqrt{3})\over\beta_{trans}}\right]^{4\over 3}\!\! , 
1412: \nonumber 
1413: %\label{gamoft} 
1414: \end{eqnarray} 
1415: where we have distinguished the average density $\rm n_p^{SN}$, close to 
1416: the parent  SN (that determines $\rm R_{max}$) from the density $\rm n_p$ 
1417: in the superbubble or the outer galaxy. 
1418: The function of interest,  $\rm\gamma(t)$, is the real root 
1419: of the above cubic equation, that is: 
1420: \begin{eqnarray} 
1421: \rm \gamma&=&\rm\gamma(\gamma_0,\theta,x_\infty;t) 
1422: =\rm {1\over B} \,\left[\theta^2+C\,\theta^4+{1\over C}\right]\nonumber\\ 
1423: \rm C&\equiv&\rm 
1424: \left[{2\over B^2+2\,\theta^6+B\,\sqrt{B^2+4\,\theta^6}}\right]^{1/3} 
1425: \nonumber\\ 
1426: \rm B&\equiv&\rm 
1427: {1\over \gamma_0^3}+{3\,\theta^2\over\gamma_0}+ 
1428: {6\,c\, t\over  (1+z)\, x_\infty} 
1429: \label{cubic} 
1430: \end{eqnarray} 
1431:  
1432: The distance travelled by the CB 
1433: is given by directly integrating Eq.~(\ref{dgamma}): 
1434: \begin{equation} 
1435: \rm x(\gamma)= x_\infty\, 
1436: \left[{1\over\gamma}-{1\over\gamma_0}\right]\, . 
1437: \label{range} 
1438: \end{equation} 
1439: The characteristic distance over which the Lorentz factor evolves 
1440: from $\gamma_0$ to  $\gamma_0/2$ is $\rm x_\infty/\gamma_0$: 
1441: roughly 1.3 kpc for  $\gamma_0=10^3$ and the reference value of 
1442: $\rm x_\infty$ in Eq.~(\ref{gamoft}). 
1443:  
1444:  
1445: In Fig.~(\ref{figflux}) we show, 
1446: for $\gamma_0=10^3$ and various viewing angles $\theta$, 
1447: the AG flux predicted by Eq.~(\ref{fluxdensity}), with $\rm n=4.1$ and 
1448: $\rm\gamma(t)$ as in Eq.~(\ref{cubic}) 
1449: with the reference value of $\rm x_\infty$ in Eq.~(\ref{gamoft}). 
1450: For $\theta\neq 0$, and particularly for sufficiently large $\gamma\,\theta_0$, 
1451: the AG curve described by Eqs.~(\ref{fluxdensity}, \ref{cubic}) 
1452: shows a very interesting behaviour. 
1453: Since $\rm \gamma(t)$ is a decreasing function of time, the Doppler 
1454: factor first increases with time, reaches a maximum value 
1455: at $\rm\gamma(t)\,\theta\sim 1$ and then declines. An observer 
1456: initially outside the beaming cone ($\gamma_0\,\theta >1$), 
1457: sees an AG that initially rises with time.  As 
1458: $\gamma$ decreases the cone broadens, and around $\gamma\theta\!\sim\! 1$ 
1459: beaming becomes less efficient, the AG declines. 
1460:  
1461: In Fig.~(\ref{figfluxother}) we show 
1462: the AG flux predicted by Eq.~(\ref{fluxdensity}), with $\rm n=4.1$ and 
1463: $\rm\gamma(t)$ as in Eqs.~(\ref{gamoft}, \ref{cubic}), 
1464: for $\gamma_0=1/\theta=10^3$, $\rm n_p=10^{-3}$ cm$^{-3}$, 
1465: and for various values of the density  $\rm n_p^{SN}$ close 
1466: to the progenitor SN. For the smaller $\rm n_p^{SN}$, the 
1467: limiting CB's radius $\rm R_{max}$ of Eq.~(\ref{Rinfinity}) is 
1468: larger. Consequently the CB, subsequently ploughing through the ISM, 
1469: loses momentum at a faster pace. The figure shows that this may 
1470: extinguish the AG very soon after the GRB, which would make it 
1471: much harder to observe. This effect can also be produced by 
1472: an increase in the ISM density $\rm n_p$, relative to 
1473: the reference choice in Eq.~(\ref{gamoft}), so that the AGs of events not 
1474: occurring in low-density superbubbles would also be hard to observe. 
1475: These may be (along with extinction) the reasons why, in some 50\% 
1476: of cases, GRBs appear not to have afterglows. 
1477:  
1478: For late times, when $\gamma\theta\ll 1$, Eq.~(\ref{gamoft}) implies that 
1479: $\rm\gamma\propto t^{-1/3}$. 
1480: According to Eq.~(\ref{fluxdensity}), then, 
1481: the AG light curve approaches: 
1482: \begin{eqnarray} 
1483: \rm F_\nu (t) &\propto& \rm t^{-\tau};\nonumber\\ 
1484: \tau&=&\rm 1+\alpha\simeq 2.1\, , 
1485: \label{lateAG} 
1486: \end{eqnarray} 
1487: while the $\nu$ dependence stays put at $\nu^{-\alpha}$. This is compatible 
1488: with what is  seen in various late-time AG observations (see e.g. 
1489: GRB 980326: Bloom et al.~1999a; 
1490: GRB 980519: Halpern et al.~1999; 
1491: GRB 990123: Holland et al.~2000a; 
1492: GRB 990510: Stanek et al.~1999, Harrison et al.~1999, Holland et al.~2000a;    
1493: GRB 991208: Castro-Tirado et al.~2001); 
1494: GRB 000301c: Masetti et al.~2000, Jensen et al.~2000; 
1495: GRB 000926: Fynbo et al.~2001, Harrison et al.~2001, Price et al.~2001;  
1496: GRB 010222: Masetti et al~2001, Stanek et al.~2001, Cowsik et al.~2001a). 
1497:  
1498:  
1499:  
1500:  
1501: \section{Comparison with optical observations} 
1502:  
1503: We compare our predictions with raw AG observations, i.e.  observations 
1504: not corrected for extinction. The first step in our procedure 
1505: is to work out what a raw ansatz-standard-candle supernova, SN1998bw, 
1506: would look like at the location of each GRB. For that, we use the 
1507: bare (unextinct) SN1998bw deduced by Galama et al.~(1998a), we 
1508: transport it to the GRB location by way of Eq.~(\ref{bw}) and correct it 
1509: for extinction in the host galaxy and in ours. 
1510: For the extinction in our galaxy 
1511: we use the estimates of Schlegel et al. (1998). 
1512: For the correction at the host (at the emitted 
1513: frequency $\rm (1+z)\,\nu$) we use the wavelength-dependent extinction 
1514: estimated by the observers from the AG spectrum and the colours of the 
1515: host galaxy.  The total extinction of the SN contribution for each 
1516: GRB's optical AG is given in Table II. 
1517:  
1518: The next step in our procedure is to fit the raw data minus the 
1519: SN contribution to a fixed host galaxy luminosity plus the CB's afterglow. 
1520: We do not correct the latter for extinction, which only affects 
1521: its fitted normalization. In the fits, 
1522: the afterglow's spectral energy density is given by Eq.~(\ref{fluxdensity}) 
1523: with $\rm R=R_{max}$ and a constant ISM density, that is: 
1524: \begin{equation} 
1525: \rm 
1526: F_\nu=F \; [\gamma(t)]^{2\alpha}\;[\delta(t)]^{3+\alpha}\, , 
1527: \label{fluxdensity2} 
1528: \end{equation} 
1529: with a normalization factor $\rm F$ which is one of the fitted parameters, 
1530: and with $\rm \gamma(t)$ 
1531: as in Eqs.~(\ref{gamoft})\footnote{Admittedly, our expression for 
1532: the AG light curve is not as simple as Eq.~(\ref{pheno}), 
1533: but it is also analytical. And it is justified.}. The contribution of the 
1534: host galaxy is fixed inside the $\pm 1\sigma$ error range of the photometry 
1535: measurements at the late times when the CB and the SN have 
1536: become sufficiently dim. 
1537:  
1538: The parameters to be fit are $\rm F$ and $\alpha$ in Eq.~(\ref{fluxdensity2}), 
1539: as well as $\rm x_\infty$, $\rm \gamma_0$ and $\theta$ entering the 
1540: expressions for $\delta$ and $\gamma$ as functions of time, Eq.~(\ref{cubic}). 
1541: The fit is done with the program MINUIT, checked over and 
1542: over (in the hunt of false local minima) with different input parameters. 
1543: The values of 
1544: $\gamma_0$, $\rm \alpha$, $\theta$ and $\rm x_\infty$ 
1545: for the different GRB afterglows 
1546: are listed in Table II,  the overall normalization $\rm F$ will be 
1547:  discussed separately. The results of these fits, which are very good, 
1548: are shown in Figs.~(\ref{fig228}) to (\ref{jump508}). Most of them refer to 
1549: R-band observations, which are the most extensive and accurate ones in the 
1550: optical band, and extend to very late times. In order to test that our 
1551: best fitted parameters are independent of frequency we have also fitted 
1552: other bands, when sufficiently accurate data are available, and obtained 
1553: very similar best fitted parameters. This can be seen in Table II 
1554: for GRBs 990510 
1555: and 990712, for which we also present V-band fits. 
1556: Before discussing the results in much more detail in section 13, 
1557: we pay attention to two very peculiar afterglows, and one 
1558: particularly well measured one. 
1559: 
1560: \subsection{GRB 970508, CBs exiting a superbubble?}
1561: 
1562: The optical AG of GRB 970508 is the only 
1563: one so far that has been seen to rise and fall very significantly. 
1564: In Fig.~(\ref{1CB508}) we show how 
1565: miserably a fit to this GRB fails, if it is made in the same way 
1566: as all of our other fits.
1567: 
1568: We have argued in Section 10.1 that GRB progenitors are presumably
1569: located in super-bubbles of 0.1 to 0.5 kpc size. There may be instances
1570: in which the jet of CBs, after travelling for such a distance, does not
1571: continue onwards to a similarly low-density halo region, but encounters
1572: a higher-density domain. To test whether this may explain the very
1573: peculiar shape of this AG, we have made a fit with two 
1574: values of the ISM particle-number density, instead of one,
1575: and a time (or distance from the progenitor) at which the
1576: transition occurs.  The result is shown in Fig.~(\ref{jump508})
1577: and it is fairly satisfactory. 
1578: 
1579: The fit parameters correspond to a density increasing by a factor 
1580: of $\sim 2.2$ at $\rm t\sim 1.1$ day after burst,
1581: at which point the CBs have travelled some $\sim 0.24$ kpc,
1582: a very reasonable radius for a superbubble. The remaining
1583: parameters are in the usual range, but for $\theta$, which,
1584: at $\sim 3.5$ mrad, is on the large side. Given this large value
1585: and $\gamma_0\sim 1123$, the time at which 
1586: $\rm\gamma(t)\,\theta=1$ is reached exceptionally late; this
1587: explains the rise and fall of the theoretical curve; see Dar
1588: and De R\'ujula (2000a) for an earlier version of this result.
1589: The relatively large $\theta$ is also in accordance with the
1590: fact that, in spite of a relatively large
1591: $\gamma$, the equivalent spherical energy of this GRB is
1592: particularly low, see Tables I and II.
1593: 
1594: In an earlier version of this paper, we attributed the shape
1595: of the AG of GRB 970508 to the effects of gravitational lensing
1596: by an intervening star or binary, of mass $\rm\sim 2\,M_\odot$. 
1597: That was an error. The required mass for an object placed mid-way
1598: to the GRB location is almost three orders of magnitude bigger.
1599: An effect of the observed size and shape
1600:  could also be due to an even heavier object 
1601: (such as a globular cluster) kiloparsecs away from the source,
1602: but the chance probability for that is negligible. The possibility
1603: of lensing by stars ---which has a few percent probability and
1604: would produce amplification effects typically lasting $\sim 1$ hour--- 
1605: is still interesting. We discuss it in Appendix III.
1606: 
1607:  
1608:   
1609: \subsection{GRB 980425, a very special case?} 
1610:  
1611: As reported in Table I, this GRB is by far the closest and yet, its measured 
1612: $\gamma$-ray fluence is not large. We have argued (DD2000a) that 
1613: in the CB model this is simply due to the fact that it is observed at a very 
1614: large $\theta$: the GRB fluence has the same angular dependence 
1615: as Eq.~(\ref{fluence}). The association of this GRB with SN1998bw is clear. 
1616: In fact, the ``afterglow'' is dominated by the SN. Only the very last 
1617: measured AG point is significantly above the $\rm ^{56}Co$ decay 
1618: trend of the SN ejecta and would be due to the proper afterglow: that 
1619: of the CBs (see Fig. 4  of D2000a). With only one point above the SN 
1620: ``background'' we cannot make, in the case of GRB 980425, 
1621: a detailed fit of the sort we have made for the other GRB afterglows. 
1622: But, as we shall see after we gain confidence on the success of the 
1623: CB model in describing X-ray afterglows, we can use the X-ray 
1624: measurements for this GRB to determine its parameters and to show 
1625: that, after all, it is not at all a very special case. 
1626:  
1627: \subsection{GRB 990123, the early optical data} 
1628:  
1629: In the case of this GRB, there are good optical data  starting 
1630: exceptionally early: during the $\gamma$-ray activity at $\rm t=22.18$ s after 
1631: its detected beginning (Akerlof et al. 1999).  
1632: The AG rises abruptly to a second point at $\rm t=47.38$ s, 
1633:  and decreases thereafter. 
1634: At the earlier stage, the CB is still hot and fully ionized, its  
1635: thermal bremsstrahlung (free-free) self absorption  
1636: %($\rm a_{\nu}\approx 0.018\, T^{-3/2}\, n_{CB}^2\, 
1637: %\nu^{-2}\, g_{ff}\, cm^{-1}$ where $\rm g_{ff}$ is the quantum mechanical 
1638: %gaunt factor, e.g. Rybicki and Lightman 1979)  
1639: is very large and fast decreasing, resulting in a fast rising AG 
1640: that turns into a declining light curve 
1641: when the CBs become transparent to short radio waves (corresponding to 
1642: optical light in the observer frame). It is possible to explain the 
1643: initial rise in detail, but the scarcity of the data and the surplus 
1644: of parameters make the exercise moot.  
1645: We choose to describe this AG from $\rm t=47.38$ s $\sim 5.5\times 10^{-4}$ 
1646: d onwards, 
1647: the first point shown of the measured decline and of Fig.~(\ref{early123}).  
1648: We can use the relation between local distance and observer's 
1649: time, Eq.~(\ref{dxsn}), 
1650: and the specific values of $\rm z$, $\gamma_0$ and $\delta_0$ 
1651: reported in the Tables for this GRB, to conclude that at the start of the 
1652: optical AG data the CBs are a mere 0.46 pc away from the progenitor star. 
1653: This is precisely in the domain where the density profile ought 
1654: to be that of Eq.~(\ref{WRwind}), $\rm n\propto r^{-2}$, induced 
1655: by the parent-star's wind and ejecta. Since 
1656: at these early times the CB's deceleration is negligible, an 
1657: $\rm r^{-2}$ density profile translates directly into an 
1658: optical AG that declines as $\rm t^{-2}$, see Eq.~(\ref{fluxdensity}).  
1659:  
1660: In Fig.~(\ref{early123}) we report the result of a fit to the AG 
1661: that includes a term proportional to $\rm t^{-2}$. The parameters 
1662: of the late ($\rm t\!>\! 1$ d) AG remain essentially unchanged 
1663: relative to the ones used before in constructing Fig.~(\ref{fig123}). 
1664: The normalization of the fitted $\rm t^{-2}$ term is very close to 
1665: that implied by Eqs.~(\ref{fluxdensity}) and (\ref{WRwind}): 
1666: the first point of the data is exactly reproduced for a density 
1667: $\rm n=0.54 \,cm^{-3}$ at that point, at which $\rm x\simeq 0.46$ pc. 
1668: Thus, the CB model also succeeds in describing this very early 
1669: AG in magnitude and shape\footnote{Even the most adamant defenders 
1670: of fireballs admit that, in their scenarios, the absence of ``windy'' AGs is  
1671: a problem, see, e.g. Piran (2001).}. 
1672:  
1673: \section{Comparison with X-ray observations} 
1674:  
1675:  
1676: The CBs enter their AG phase when they become transparent to radiation. 
1677: As we have seen, their X-ray AG is dominated by thermal bremsstrahlung 
1678: and first declines with time as $\rm \sim t^{-5}$, as described by 
1679: Eqs.~(\ref{Tbrem}, \ref{fbrem}). In a matter of observer's minutes, for typical 
1680: parameters, the CB's radius reaches its limiting value and 
1681: the synchrotron radiation of Eq.~(\ref{fluxdensity}),  
1682: which is proportional to $\rm n_p\, R^2$, becomes the dominant AG 
1683: source in both the X-ray and optical domains: the corresponding lightcurves 
1684: are approximately proportional. Once again for typical parameters, a CB 
1685: reaches the galactic halo in just a few hours. The light curves before that  
1686: time would be hard to model in minute detail, 
1687: since the ISM density is no doubt changing rapidly. 
1688: Coincidentally, there are not enough continuous X-ray measurements during 
1689: the first few hours after the GRBs to reliably extract a density 
1690: profile. 
1691:  
1692: The previous considerations justify a very simple 
1693: description of the X-ray light curves. Let $\rm  W$ 
1694: be the (constant) ratio of X-ray to R-band optical AGs. We expect: 
1695: \begin{equation} 
1696: \rm   F_X(t) \simeq F_X(t_{trans})\, \left[{t_{trans}\over t}\right]^5 
1697: + W\, F_R(t) \, . 
1698: \label{Xdensity} 
1699: \end{equation} 
1700: where $\rm t$ is the observer's time since the ejection of the (last) CB, and 
1701:  $\rm F_R(t)$ is the spectral energy density in the R-band, 
1702: i.e., Eq.~(\ref{fluxdensity}), that we have fitted to the R-band AG 
1703: observations.  The two parameters we fit to X-ray light curves 
1704: are $\rm F_X(t_{trans})\times [t_{trans}]^5$ and 
1705: $\rm W$, resulting in a very good description 
1706: of  the measured X-ray afterglows of the GRBs 
1707: of known redshift: 970228, 970828, 971214, 980613, 990123, 
1708: 990510, 000926 and 010222. The cases of GRBs 970508 
1709: and 991216 require a non-constant density profile. 
1710: The fits to the X-ray AGs are shown in 
1711: Figs.~(\ref{X228}-\ref{X222}).  
1712: GRB 980425 is discussed separately below. 
1713:  
1714: During a GRB the X-ray luminosity fluctuates as the $\gamma$-ray 
1715: luminosity does, changing abruptly at the end of the GRB  into 
1716: a very fast decline. This is expected in the CB model, in which the two 
1717: behaviours have slightly different origins: thermal bremsstrahlung from 
1718: the various CBs' surfaces during the GRB, thermal bremsstrahlung from the 
1719: rapidly cooling CBs' volumes as they become transparent to the radiation 
1720: they enclose, in a short time $\rm t_{trans}$ ---of the order of 
1721: seconds--- at the end of each individual CB. In 
1722: Figs.~(\ref{X228}-\ref{X222}) we have therefore 
1723: shown our fits to data beginning 
1724: at the start of the sharp X-ray decline.  
1725:  
1726:  
1727: The use in Eq.~(\ref{Xdensity}) of a more careful treatment of the 
1728: bremsstrahlung contribution (evolving from $\rm t^{-5}$ to $\rm t^{-3}$) 
1729: is unwarranted: the assumption of a constant ISM density 
1730: should be inappropriate between 
1731: $ \rm \sim\! 2\times 10^{-3}$ and $\sim\! 0.2$ days, and there are no 
1732: data in that domain except for GRB 991216 and perhaps 970508 that, like the 
1733: early optical AG of GRB 990123, suggest an initial density variation  
1734: $\rm\propto 1/r^2$, resulting in an observed $\rm\sim t^{-2}$ decline. 
1735: In these cases, we have fitted the data with 
1736: Eq.~(\ref{Xdensity}) for a $\rm \sim 1/r^2$ plus constant density along the 
1737: CB trajectory. For the particularly interesting case of GRB 970508,  
1738: shown in Fig.~(\ref{X508}), we have also subtracted from the data 
1739: the contribution of the X-ray line observed before 0.8 d.  
1740: This reduces the two points observed  between 
1741: 0.2 and 1 day by a factor $\sim 0.39$. 
1742: The overall result is compatible with an effect that, at late times, 
1743: is achromatic, since both the late optical and X-ray AGs are
1744: proportional to $\rm n_e$; see Fig.~(\ref{jump508}) 
1745: for the optical counterpart.
1746:  
1747: In what follows we refer to the initially rapidly-falling part of 
1748: an X-ray AG, as ``early'' and to the subsequent much flatter 
1749: AG as ``late''. 
1750:  
1751: We have chosen to write the prediction for X-rays 
1752: as in Eq.~(\ref{Xdensity}) to better expose the expected achromaticity 
1753: of the late AG. Moreover, the data for late X-ray AGs is generally more 
1754: sparse than for optical light; it is therefore advisable to exploit 
1755: the fact that the CB's parameters are better fit from the R-band data, 
1756: and to write the late X-ray AG as a rescaling of that data.  
1757: But the effects of absorption are much less severe 
1758: for X rays and there is one parameter, 
1759: the overall normalization, for which it is preferable to use the 
1760: X-ray fluence in testing the model. The values of 
1761: $\alpha$, $\theta$, $\gamma_0$, and $\rm x_\infty$, 
1762: fit for each GRB to the R-band AG, are sufficient to deduce,  
1763: via Eq.~(\ref{fluxdensity}), the shape ---but not the normalization--- of the 
1764: expected {\it late} X-ray flux in the 2-10 keV domain.  
1765: This flux depends, via $\rm n_e\,R_{max}^2$ at fixed $\rm x_\infty$, 
1766: on the number of CBs and their individual baryon number, as 
1767: $\rm F\propto n_{_{CB}}\,N_{CB}=N_{jet}$ (the dependence on 
1768: $\rm B_\perp$, via $\nu_0^{\alpha-1}$, is extremely weak). 
1769: Let $\rm q$, for a given GRB, be the ratio of the observed flux to the one  
1770: expected for our canonical $\rm N_{CB}=6\times 10^{50}$, and  
1771: for one dominant (largest $\gamma$) cannonball: $\rm n_{_{CB}}=1$. 
1772: The values of $\rm q$ are reported in Table III. They range from $\sim 1/3$ to 
1773: $\sim 3$, indicating that the CB model satisfactorily explains the  
1774: {\it late} X-ray AG normalizations 
1775: and that the total baryon number of the  
1776: ensemble of CBs that dominate the AG appears to be quite constant. 
1777:  
1778:  
1779: One can see in Figs.~(\ref{X228}) to (\ref{X222}), that the 
1780: X-ray fluences at the start of the X-ray decline are 
1781: $\rm \sim 10^{-8}-10^{-7}\; erg\, cm^{-2}\, s^{-1}$. These figures  
1782: for the beginning of the early X-ray AG compare 
1783: quite favourably with the typical prediction quoted after Eq.~(\ref{fbrem}): 
1784: $\rm 4.35\times 10^{-9}\; erg\, cm^{-2}\, s^{-1}$ for a single CB. 
1785: Once again, the parameters extracted from fitting the optical AGs 
1786: are not sufficient to fix case by case the overall {\it early} X-ray normalization in 
1787: Eq.~(\ref{fbrem}), which also depends directly on other parameters 
1788: (notably $\rm N_{CB}$, $\rm n_{CB}$ and the radii of the still-growing CBs).  
1789: Using these degrees of freedom we could fit not only the late, but also the 
1790: early absolute X-ray fluences. Suffice it to say that the magnitude of the  
1791: {\it early} X-ray afterglow is also the one expected in the CB model. 
1792:  
1793:  
1794: The fitted values of $\rm W$ in Eq.~(\ref{Xdensity}) 
1795: are in fair agreement with the absorption-dependent expectation 
1796: $\rm  [A(\nu_X)/A(\nu_O)]\,(\nu_X/ \nu_O)^{-\alpha_{_{\!OX}}}$ 
1797: from the spectrum of Eq.~(\ref{spectrum}). 
1798: Some examples of values of $\rm\alpha_{_{\!OX}}$ 
1799: extracted this way are 
1800: $1.06\pm 0.12$  for GRB 970228 (Frontera et al., 1998), 
1801: $1.12 \pm 0.07$  for GRB 970508 (Galama et al. 1998b), 
1802: $0.95\pm 0.1$ for GRB 971214 (Dal Fiume et al. 2000), 
1803: $>1$ for GRB 980703 (Vreeswijk et al. 1999) 
1804: $0.96 \pm 0.26$  for GRB 990510 (Pian et al. 2001), 
1805: 0.9 to 1.1 for GRB 000926 (Piro et al. 2001) and 
1806: $0.97 \pm 0.05$  for GRB 010222 (In `t Zand et al. 2001). 
1807: However, the inferred values of $\rm \alpha_{OX}$ are affected by large 
1808: and very uncertain extinctions in the CB and host galaxies. Evidence for 
1809: large extinction of optical AGs of GRBs in their host galaxies is provided 
1810: by the large column densities ($\rm N_H > 10^{22}\, cm^{-1}$) extracted 
1811: from the X-ray observations of some GRBs (e.g., GRBs 970228, 970508, 
1812: 970828, 971214, 980329, 980519: Owens et al. 1998;  GRB 980703: Vreeswijk 
1813: et al. 1999) and from the absorbed spectra of the optical AG of other GRBs 
1814: (e.g. GRB 990712: Vreeswijk et al. 2000; GRB 991216: Halpern et al. 2000a; 
1815: GRB 000131: Andersen et al. 2000). In fact, the failure to detect the 
1816: optical AG of many long duration GRBs with well localized X-ray AG ---like 
1817: GRB 970111, GRB 970616, GRB 970815, and 970828--- may be due to strong 
1818: extinction of their optical AG in the host galaxy (Djorgovski et al. 2001 
1819: and references therein). 
1820:  
1821:  
1822:  
1823: \subsection{GRB 980425 is not exceptional} 
1824:  
1825: The $\gamma$-ray fluence of this GRB is not atypical but 
1826: its redshift, $\rm z=0.0085$, is extremely small. 
1827: If it is not intrinsically exceptional, its CBs must be viewed from 
1828: an atypically large angle (DD2000a). 
1829: For large $\theta$ the CBs' afterglow is strongly 
1830: reduced, as can be seen from Fig.~(\ref{figflux}), allowing for the possibility 
1831: that the AG is dominated by the SN. 
1832: Consequently, the CB parameters cannot be derived 
1833: from the optical light curve of the blended SN 1998bw/GRB 980425 system. 
1834: But they can be deduced from the X-ray emission of the system if we 
1835: assume, unlike the observers do (e.g., Pian et al. 1999;  Pian et al. 2001) 
1836: that it was produced by the CBs and {\it not} by the conventional 
1837: quasi-spherical SN ejecta\footnote{We are indebted to E. Pian for 
1838: discussions on this point.}. Indeed,  
1839: significant X-ray emission from SNe has been detected only 
1840: at much later times after the event. Moreover, the exceptionally slow 
1841: decline of the X-ray AG in this GRB is what is expected from the 
1842: large viewing angle interpretation, see Fig.~(\ref{figflux}). 
1843:  
1844: Given all of the above, for this GRB 
1845: we have ``reversed'' our procedure by first fitting the X-ray AG of the 
1846: SN 1998bw/GRB 980425 system. The fit is shown in 
1847: the upper part of Fig.~(\ref{425}) and the 
1848: fitted parameters are listed in Table II. The data are not very 
1849: precise and the fit (for which we assumed $\alpha=1.1$, 
1850: as fitted for all other GRBs) is not one of our best,  
1851: but it inescapably requires a very large viewing angle $\theta$. 
1852: The best fitted angle is $\theta\sim 8.3$ mrad, 
1853: corresponding, for the fitted $\gamma\sim 750$, 
1854: to an initial Doppler factor  $\delta\sim 37$.  
1855: If the CBs of GRB 980425 had been viewed from a typical viewing 
1856: angle, $\theta\!\leq\! 1/\gamma_0 $, the equivalent isotropic energy 
1857: would have been in the range $ 7.3\times 10^{51}$ 
1858: to $5.8 \times 10^{52}$ erg, like that of all other GRBs. 
1859:  
1860: If we assume that for GRB 980425 
1861: the extinction, ISM density, and CB 
1862: radius were the same as for other GRBs well measured in X-rays, 
1863: such as GRB 990510 or GRB 010222, we can use Eq.~(\ref{fluxdensity}) 
1864: to derive the expected intensity of the X-ray AG plateau of GRB 980425, 
1865: see Fig.~(\ref{425}), and its caption. 
1866: The results are $\rm F_X[425] = 0.32\, F_X[510]$, and  $\rm F_X[425] = 0.15\, 
1867: F_X[222] $, both yielding $\rm F_X[425] 
1868: \!\sim\! 4\times 10^{-13}\, erg\, cm^{-2}\, s^{-1}$, in 
1869: agreement with the BeppoSAX observations (Pian, 1999; Pian et al. 2000). 
1870: The double success in deducing a ``normal''  GRB equivalent 
1871: isotropic energy, and the 
1872: intensity of the X-ray AG, constitutes a very strong support for 
1873: the alleged association of SN1998bw with (a not exceptional) GRB 980425. 
1874:  
1875: The  fitted parameters of the X-ray AG can be used to predict the 
1876: magnitude and shape of the 
1877: optical AG of the blended SN 1998bw/GRB 980425 system, 
1878: if we assume the same  V/X extinction ratio as in GRBs 990510, 000926 and 
1879: 010222. This we do in the lower part of Fig.~(\ref{425}). The 
1880: CBs' contribution dominates at very late time and, 
1881: remarkably, it is in perfect agreement with the HST 
1882: observation (Fynbo et al. 2000) on day 778 after the GRB.    
1883:  
1884: \section{Discussion of the results on optical AGs} 
1885:  
1886: With the exception of the AG of GRB 970508, which has the sharp ``break up'' 
1887: that we have explained via a sudden change in density, 
1888: a look at Figs.~(\ref{fig228}) to (\ref{jump508})  clearly reveals that the 
1889: observed AGs have absolutely no ``breaks''. 
1890: In the CB model, the gradual evolution of the proper 
1891: afterglow (that of the CBs) is simply a consequence of 
1892: the gradual decrease of the Lorentz factor $\rm \gamma(t)$. 
1893: We give in Table II the list of the parameters resulting from 
1894: our fits ($\alpha$, $\gamma_0$, $\theta$ and $\rm x_\infty$) 
1895: to optical R-band afterglows (and in two cases, to 
1896: the V-band afterglow as well). 
1897:  
1898: The fits to the CB model are satisfactory, particularly since the best-fit 
1899: parameters turn out to be precisely in the expected ranges. 
1900: On close inspection one notices that our curves occasionally 
1901: undershoot or overshoot some points by a small factor, as in 
1902: GRBs 990123 and 000301c\footnote{The feature at $\rm t\!\sim\! 4$ days in 
1903: GRB 000301c has been interpreted by Garnavich et al. (2000b) 
1904: as due to gravitational lensing.}. 
1905: This is not a surprise: the AG fluences 
1906: are proportional to the ISM number density, which we do not 
1907: expect to be exactly constant for kpc distances, even 
1908: in the halo of galaxies. If such ``defects'' were not present 
1909: in our fits, we would have concluded 
1910: that the data had been over-parametrized. For the same reasons, 
1911: and because of the systematic errors in the data, 
1912: the values of the parameters we extract from our fits should not 
1913: be taken entirely at face value, even though the minimization procedure 
1914: ---which attributes to the errors a counterfactual purely statistical origin--- 
1915: results in tiny 1 $\sigma$ spreads for the fitted parameters, and in 
1916: $\chi^2 $ values that are in most cases extremely satisfactory. 
1917:  
1918: All the  figures (\ref{fig228}) to (\ref{jump508}) refer to optical data 
1919: for $\rm t\!\geq\! 0.1$ days, for which it is reasonable 
1920: to approximate the ISM density by a constant value, 
1921: describing the density of the superbubble and/or the galactic halo. 
1922: We have already discussed the early observations of 
1923: GRB 990123, for which this approximation breaks down. 
1924:  
1925: \subsection{The distribution of fitted parameters} 
1926:  
1927: In the CB model, the parameter $\alpha$ of 
1928: Eqs.~(\ref{spectrum}, \ref{fluxdensity}) is 
1929: the only one for which we have no reason to expect 
1930: a range of different values. It is therefore extremely 
1931: satisfactory  that the fitted values of $\alpha$ are, within errors, 
1932: compatible with {\it all} of the GRBs having a universal 
1933: behaviour with the theoretically predicted value: $\alpha\approx 1.1$, 
1934: Eq.~(\ref{spectrum}). The narrow distribution of best fitted $\alpha$ 
1935: values is shown in Fig.~(\ref{alphadist}). 
1936: In the CB model, we have extracted the 
1937: values of $\alpha$ from 
1938: the temporal shape of the AG and ---adding consistency 
1939: to the picture--- they agree  well with the values 
1940: obtained from spectral observations, either in X-rays (with spectra 
1941: modified by a best-fit hydrogen column density) or in the 
1942: optical domain (with a galactic colour-dependent extinction). 
1943: Some examples are:\\ 
1944: GRB 970228: $\rm\alpha_X=1.06\pm 0.12$  (Costa et al.~1997).\\ 
1945: GRB 970508:  $\rm\alpha_{O}=1.12\pm 0.07$ (Galama et al.~1998c);\\ 
1946:  $\rm\alpha_X=1.11\pm 0.06$ (Galama et al.~1998c).\\ 
1947: GRB 990123: $\rm\alpha_{O}=1.29\pm 0.23$ (Holland et al.  1999a).\\ 
1948: GRB 990510:  $\rm\alpha_X=0.96 \pm 0.26 $ (Pian et al.~2001);\\  
1949:  $\rm\alpha_{O}=1.26\pm 0.15$ (Stanek et al.~1999).\\ 
1950: GRB 991208: $\rm\alpha_{O}=1.05\pm 0.09$ (Castro-Tirado et al. 2000).\\ 
1951: GRB 991208: $\rm\alpha_{O}\approx 1.1$ (Takeshima et al. 1999).\\ 
1952: GRB 000301c: $\rm\alpha_{O}=1.15\pm 0.26$  (Jensen et al.~2000).\\ 
1953: GRB 000926:  $\rm\alpha_X=1.2\pm 0.3$    (Piro et al.~2001);\\  
1954:  $\rm\alpha_{O}=1.02\pm 0.02$ (Sagar et al.~2001a).\\ 
1955: GRB 010222:  $\rm\alpha_X=0.97 \pm 0.15$ (In `t Zand et al.~2001);\\ 
1956:  $\rm\alpha_{O}=1.07\pm 0.09$  (Stanek  et al.~2001). 
1957:  
1958: There are also cases for which the reported value of $\alpha$ 
1959: differs significantly from 1.1. One notable instance is GRB 990510, 
1960: for which Beuerman et al. (1999) report $\alpha\sim 0.55$ and 
1961: Stanek et al. (1999) find $\alpha=0.61\pm 0.12$. To extract this value 
1962: the authors extrapolate the measured extinction: $\rm E(B-V)=0.20$ 
1963: (Schlegel et al. 1998). 
1964: If this measured extinction is used to correct {\it only} the measured  
1965: $\rm B-V$ 
1966: values for GRB 990510: $0.57\pm 0.02$ (Beuerman et al. 1999), 
1967: and $0.56\pm 0.03$ (Stanek et al. 1999), 
1968: one obtains $\rm \alpha_O=1.08\pm 0.12$. 
1969: The uncertainties entailed by absorption corrections are the 
1970: reason why we have chosen to de-emphasize results that are sensitive 
1971: to them, whether they do, or do not, agree with our expectations. 
1972:  
1973: The distribution in initial Lorentz factors, $\gamma_0$, shown in 
1974: Fig.~(\ref{gammadist}), agrees snugly with our expectation, 
1975: $\gamma_0\sim 10^3$, extracted from independent 
1976: information: the fluence and the individual-photon energies of GRBs 
1977: (DD2000a,b) and the energies of X-ray lines in their afterglow 
1978: (DD2001a). Notice how surprisingly narrow this distribution is. 
1979:  
1980: The distribution of viewing angles $\theta$ is shown 
1981: in Fig.~(\ref{thetadist}). 
1982: The AG data for GRB 000131 consist in only three points, 
1983: while for GRBs 991208  and 000301c the measurements start rather late. 
1984: The sensitivity to $\theta$ in our fit to these GRBs is not good. We reflect 
1985: this fact in Fig.~(\ref{thetadist}) by having the corresponding 
1986: entries unshaded. The distribution is compatible with the 
1987: expectation that the limited experimental sensitivity to GRBs introduces 
1988: a sharp cutoff as $\theta$ increases; see the steep fluence function, Eq.~(\ref{fluence}). 
1989:  
1990: The parameter $\rm x_\infty$ of Eq.~(\ref{gamoft}) is the only 
1991: one for which we expect a rather broad distribution. 
1992: Indeed, it depends on the densities close to the GRB 
1993: progenitor, which ought to be 
1994: quite variable, and in the region where 
1995: the CB light is emitted; see Eq.~(\ref{gamoft}). 
1996: The values of $\rm x_\infty$ reported in Table II show 
1997: a spread of a bit over one order of magnitude, 
1998: supporting the expectation. 
1999: In Fig.~(\ref{xdist}) we show the distribution of 
2000: $\rm Log_{10}[x_\infty(Mpc)]$, which peaks at 
2001: the reference value of Eq.~(\ref{gamoft}) and extends 
2002: to smaller values, as expected if the average density close 
2003: to the progenitor is sometimes much smaller than our rather large 
2004: reference value: $\rm n_p^{SN}=1\;cm^{-3}$; and/or the density 
2005: of the ISM is bigger than our rather low reference value: 
2006: $\rm n_p=10^{-3}\;cm^{-3}$. 
2007:  
2008: The values of $\alpha$, $\theta$, $\gamma_0$ and $\rm x_\infty$ 
2009: are not sufficient to predict the overall normalization 
2010: of an AG: $\rm F$ in Eq.~(\ref{fluxdensity2}), whose approximate 
2011: value is given by the absolutely normalized 
2012:  Eq.~(\ref{fluxdensity}). Indeed, 
2013: $\rm F$ is proportional to the product of the number of CBs 
2014:  and their baryon number. To skirt absorption 
2015: corrections we have discussed in section 12 the values of $\rm F$ 
2016: in connection with late X-ray AGs. There, we compared 
2017: the data and the naive expectation for  a single (highest-$\gamma$)  
2018: dominant CB ($\rm n_{_{CB}}=1$) and 
2019: our canonical $\rm N_{CB}$, to work out the ratio $\rm q$ 
2020: of fitted to predicted values of $\rm F$ in the X-ray band.  
2021: The same exercise can be carried along for the R-band AGs, 
2022: with the result that the optical values of $\rm q$ are not 
2023: within a factor of three, but within an order of magnitude 
2024: of $\rm q=1$. It is tempting to conclude that this may 
2025: be due to poorly-understood absorption. It is in any case clear that 
2026: the AG magnitude, in the CB model, is not a problem\footnote{In the 
2027: fireball model and its descendants, the 
2028: efficiency of conversion of kinetic energy to photons is claimed to be 
2029: high, and both the GRBs and their AGs are due to the same 
2030: mechanism: synchrotron radiation. It is therefore difficult to explain why, 
2031: at the end of the GRB, the radiation rate suddenly drops by two orders of 
2032: magnitude or more, and why there is more energy in the GRB than in the 
2033: afterglow (see, e.g. Burenin et al.~1999; Pian et al.~2001, In `t Zand et 
2034: al.~2001).}. 
2035:  
2036: The GRB consists of the photons emitted by the hot CBs as they 
2037: exit the SN shell. In the rest system of a CB, the individual energies, as in 
2038: Eq.~(\ref{energies}), are a fraction $\rm (1+z)/\delta_0$ of 
2039: the observed energies, with $\delta_0=\delta[\gamma_0,\theta]$. 
2040: The total energy emitted by a GRB, in the rest system of its CBs, is 
2041: in the form of isotropically distributed photons that appear to us as the collimated $\gamma$-ray burst. This comoving total energy 
2042:  is related to the observed fluence 
2043: $\rm F_\gamma$ by: 
2044: \begin{equation} 
2045: \rm E^{CB}_\gamma = {4\,\pi\,D^2_L\,F_\gamma\over(1+z)\,\delta_0^3}\, . 
2046: \label{ECBrest} 
2047: \end{equation} 
2048: For the GRBs all of whose parameters are well determined, 
2049: we list in Table III the values of $\rm E^{CB}_\gamma$. 
2050: Interestingly, their distribution ---shown in 
2051: Fig~(\ref{Edist})---  is also quite narrow. 
2052:  
2053: We have not discussed in this subsection the parameters of 
2054: GRB 980425, listed in Table II. 
2055: They are obtained from a fit to the X-ray ---as opposed to optical--- AG, 
2056: and they are imprecise. The deduced value of $\rm E_\gamma^{CB}$ 
2057: is $0.16 \times 10^{44}$ erg, a bit lower than those listed 
2058: in Table II. This is to be expected, the small $\delta$ of this GRB 
2059: makes its GRB softer, and less prominent within the BATSE energy window. 
2060:  
2061: To summarize, the distributions of parameters are in extremely good 
2062: agreement with the expectations of the CB model and, if anything, 
2063: they are astonishingly close to what they would be 
2064: for ``standard candle'' GRBs. 
2065:  
2066: \subsection{The GRB/SN association in view of our results} 
2067:  
2068: It is useful to discuss the evidence for a SN component in 
2069: the GRB optical AGs in order of decreasing redshift. 
2070: The fact that we have a consistent and successful description 
2071: of optical afterglows strengthens the interpretation of this 
2072: putative evidence. 
2073:  
2074:  
2075:  
2076: Examining Figs.~(\ref{fig228}) to (\ref{jump508}), we draw the 
2077: following conclusions. 
2078: In the six more distant GRBs, ranging from GRB 000131 
2079: at $\rm z=4.5$ to GRB 010222 at $\rm z\!>\!1.474$, there is 
2080: no evidence for {\it or against} a SN 1998bw-like component. 
2081: In GRB 000418, at $\rm z=1.119$, there is an indication of an excess, compatible 
2082: with the SN. In GRB 980613, at $\rm z=1.096$, the evidence, though based 
2083: mainly on just one point, is very strong. In GRB 991216, at $\rm z=1.02$, 
2084: there is a clear indication of a late excess over the CB's afterglow, though a 
2085: SN1998bw-like contribution does not describe it very well 
2086: (a slightly earlier bump would do a very good job, indicating 
2087: that the standard-candle hypothesis for the SN is good, but not perfect). 
2088: In GRB 980703, at $\rm z=0.966$, the SN excess is visible and well fitted, 
2089: but the errors are large. For GRB 970828, at $\rm z=0.957$, there 
2090: are no AG observations at optical wavelengths. In the peculiar
2091: case of GRB 970508, at $\rm z=0.835$, 
2092: there is in the data a clear excess at late times that is very 
2093: well fitted by our SN ansatz. For the next three closer GRBs 
2094: (991208, 970228, 990712), at $\rm z=0.706$ to 0.434, the evidence 
2095: is completely convincing that a SN1998bw-like contribution 
2096: is required to fit the data. In the case of  GRB 990712, once again, 
2097: a SN peak occurring slightly earlier than that of a redshift-corrected 
2098: SN1998bw would provide a better description of the AG. 
2099: Finally, GRB 980425, at $\rm z=0.0085$ is indeed associated to a SN: 
2100: our fairly satisfactory standard-candle choice. 
2101:  
2102: A clear trend is apparent in the last paragraph. The closer a GRB 
2103: is, the better the evidence for its association with a SN. The trend 
2104: is entirely consistent with the fact that, for the more distant GRBs, 
2105: a SN contribution to the AG could not be seen, even if it was there. 
2106: In all cases where the SN could be seen, it was seen, with the evidence 
2107: gaining in significance as the distance diminishes. 
2108: The temptation to conclude that all long-duration GRBs are 
2109: associated with SNe appears to us to be irresistible, even if an 
2110: irrefutable proof will never be possible. 
2111:  
2112: \subsection{Do cannonballs deserve their name?} 
2113:  
2114: We have argued that CBs should reach an asymptotic radius in 
2115: a very short time and travel thereafter as literal, i.e. non-expanding, 
2116: ``cannonballs''. Can this statement be contrasted with the data? 
2117: To answer this question we have analyzed the AGs produced by 
2118: CBs whose radius inertially increases at a fixed speed, 
2119: $\rm \beta_{exp}\, c$, in their rest system. The details are 
2120: given in Appendix II. The result is that the late AGs behave in this 
2121: case as $\rm F_\nu\propto t^{-\tau}$, with $\tau=9\,(1+\alpha)/5$. 
2122: For $\alpha\simeq 1.1$, as we have argued, $\tau\simeq 3.8$, 
2123: which completely disagrees with the AG light curves. For $\tau=2.1$, 
2124: in agreement with the latter, $\alpha\simeq 0.17$, which completely 
2125: disagrees with the measured spectra. Thus, the cannonballs 
2126: of the CB model do deserve their name. 
2127:  
2128:  
2129: \section{Summary and conclusions} 
2130:  
2131: We have previously argued that the cannonball 
2132: model offers a successful and simple explanation 
2133: of the fluence, energy spectrum, and temporal behaviour of 
2134: the prompt $\gamma$-rays of a GRB (DD2000b).
2135: >From these considerations we extracted a CB's 
2136: typical Lorentz factor, $\gamma_0\sim 10^3$, and typical 
2137: baryon number $\rm N_{CB}$: in the 
2138: vicinity of $6\times 10^{50}$, or ${\cal{O}}(10)$ times 
2139: as much for a jet of CBs in a multipulse GRB 
2140: (DD2000b, DD2001b). Using only these CB-related input parameters 
2141: and a reasonable initial CB expansion velocity, we have explicitly 
2142: worked out all of the properties of X-ray and optical afterglows. 
2143:  
2144: As an intermediate result, we 
2145: derived the temporal behaviour of the radius of 
2146: a CB and showed that, faithful to their name, CB radii tend to a constant 
2147: $\rm R_{max}$ in mere minutes of observer's time. The 
2148: value of $\rm R_{max}$ depends 
2149: on the ISM density close to the GRB progenitor. 
2150: GRBs whose CBs have large radii have afterglows whose temporal 
2151: decline is very fast. This is also the case for GRBs whose CBs travel through a 
2152: relatively dense ISM. 
2153: These may be the reasons, along with strong extinction in the host galaxy, 
2154: why the search for AGs is not always successful. 
2155:  
2156: We have shown how well the CB model describes 
2157: all the properties of the X-ray and optical afterglows of GRBs. 
2158: Our results do not ---and could not--- take into account possible variations 
2159: of the ISM density along a CB's path; they are in this sense 
2160: ``descriptions'' rather than fits. In spite of this, the descriptions 
2161: are excellent and the consistency of the results is impressive. 
2162:  
2163: The observed behaviour of both X-ray and optical AGs is the 
2164: predicted one. All the parameters extracted from the fits have 
2165: values or distributions close to the expected  ones. 
2166: This is the case for the integrated fluences in  
2167: the early and late X-ray and optical 
2168: bands, for the values of the Lorentz factor 
2169: $\gamma$, for the distribution of observation angles $\theta$, 
2170: for the spectral index $\alpha$ (that we determine, 
2171: not from the spectra themselves, but 
2172: from the time-dependence of the late afterglow), 
2173: and for the parameter $\rm x_\infty$ of Eq.~(\ref{gamoft}), 
2174: that governs the pace 
2175: of the CB's slowdown, and is a combination 
2176: of the ISM densities, and the CB's baryon number and radius. 
2177: Throughout, we have chosen to de-emphasize the results that are most  
2178: sensitive to systematic errors, such as absorption corrections. Thus, 
2179: we have not systematically extracted parameters from  
2180: flux normalizations, nor reported any $\chi^2$ tests of the quality of the fits 
2181: (which would be misleadingly good) or statistical-error estimates on the fitted 
2182: parameters (which would be misleadingly small). 
2183:  
2184:  
2185: The distribution of the derived 
2186: quantity $\rm E_\gamma^{CB}$ (the energy 
2187: of the GRB photons in the CBs' rest system) 
2188: has a range of less than a factor of four, 
2189: making long-duration GRBs not far from standard candles. 
2190: We have demonstrated that GRB afterglows can be 
2191: understood in detail. Perhaps this will pave the way to the 
2192: use of GRBs as cosmological beacons. 
2193:  
2194: GRB 980425 is at first sight a special case: its $\gamma$-rays 
2195: are rather soft, it is extremely near by cosmological standards, 
2196: and very clearly associated with a SN. In the CB model it is not 
2197: exceptional, only seen at a relatively large angle. 
2198: Its X-ray afterglow is normal and emitted 
2199: by its CBs, not by the isotropic ejecta of SN1998bw. 
2200: Its optical AG is dominated by the SN for almost two years, 
2201: but its last measured point is due to the CBs, and agrees with 
2202: the expectation. 
2203:  
2204: Very early optical AG data are only available for GRB 990123. 
2205: The CB model is also capable of describing them naturally: 
2206: their magnitude and time-dependence are those expected 
2207: for CBs moving through the density profile produced by winds in 
2208: the Wolf-Rayet phase of the progenitor. Future early observations 
2209: should test this feature of the CB model: a linear dependence 
2210: of the AG fluence on the varying local-density profile. 
2211:  
2212: For GRB 970508 we can provide a good fit 
2213: to its peculiar AG only if we assume that its CBs 
2214: encountered a sudden change in ISM density, 
2215: as they would if they are exiting a superbubble into
2216: a higher-density region.
2217:  
2218: The very early X-ray and optical AGs are not achromatic: 
2219: they are dominated by two mechanisms with different 
2220: time-dependence: thermal bremsstrahlung 
2221: and synchrotron radiation. 
2222: But in a matter of a few hours all of the AG is dominated by 
2223: synchrotron radiation, and the light curves
2224: should be achromatic, as observed. 
2225:  
2226: Our descriptions of the optical AGs indicate or require 
2227: the contribution to the light curve of a supernova akin 
2228: to SN1998bw, in all eight out of sixteen cases for which 
2229: the event occurred close enough for such a contribution 
2230: to be observable. The conclusion that there is a roughly one-to-one 
2231: association between core-collapse SNe (or perhaps just type Ib and Ic SNe) 
2232: and long-duration GRBs is very tempting. The enormous beaming 
2233: factor of the CB model makes this conclusion tenable and consistent. 
2234:  
2235: We are currently completing the study of the predictions 
2236: of the CB model for radio afterglows, for which self-absorption in the CBs
2237: is relevant, and requires a careful analysis. The results, soon to 
2238: be announced (Dado et al. 2002),
2239: are excellent. We also plan to discuss elsewhere  
2240: the interesting implications of the CB model 
2241: for cosmic-ray physics,  and for other 
2242: accreting compact objects that eject relativistic jets, such as 
2243: radiogalaxies, AGNs, quasars, microquasars, blazars and microblazars. 
2244:  
2245: It would be interesting to compare our results 
2246:  to those of conically ejected shocked fireballs. 
2247: We have argued that, for this, it would be 
2248: more convincing not to have the observed firetrumpets 
2249: pointing precisely to planet Earth. For the time being, 
2250: we have proved that the CB model ---based on simple 
2251: and definite hypothesis--- makes predictions that 
2252: are univocal (as opposed to multiple-choice), explicit, 
2253: analytical in fair approximations, quite simple, 
2254: very complete,  and very successful. 
2255:  
2256: \vskip .5cm 
2257: \noindent 
2258: {\bf Acknowledgements:} 
2259: We are very indebted to Rainer Plaga for numerous fruitful discussions. 
2260: We thank Elena Pian and Jean In `t Zand 
2261: for kindly providing us with the detailed data on the 
2262: BeppoSAX X-ray observations of GRBs 990510 and 010222; 
2263: Donald Smith and Atsunama Yoshida for the data on GRB 970828; 
2264: and Luigi Piro for the data on GRB 970508. 
2265: We are indebted to an anonymous referee for many 
2266: sensible suggestions and questions. 
2267: This research was supported in part  
2268: by the Helen Asher Fund for Space Reseach and by the  
2269: V.P.R. Fund - Steiner Research Fund at the Technion. 
2270:  
2271: \vskip .5cm 
2272: \noindent 
2273: \section*{Appendix I: Subdominant AG mechanisms} 
2274:  
2275: Three mechanisms declining less fast than thermal bremsstrahlung 
2276: ---but typically subdominant relative to synchrotron radiation--- 
2277: contribute to GRB afterglows: 
2278: relativistic bremsstrahlung  from Coulomb collisions 
2279: of CB electrons with the ISM constituents, 
2280: atomic transitions of CB atoms excited or ionized by these 
2281: same collisions, and inverse Compton scattering of 
2282: the CBs' electrons on the cosmic background radiation. 
2283: We discuss them here to substantiate our assertion that 
2284: thermal bremsstrahlung and synchrotron radiation 
2285:  dominate the AGs. 
2286:  
2287: \subsection*{Relativistic bremsstrahlung} 
2288:  
2289: The electrons of the CB, in their highly relativistic collisions 
2290: with the ISM nuclei, emit non-thermal bremsstrahlung 
2291: radiation. 
2292: The total power per unit area received by the observer 
2293: from this source  is: 
2294: \begin{equation} 
2295: \rm {dF_{RB}\over dt\, d\Omega} \approx { 45.7\, \alpha\, r_e^2\, N_{CB}\, 
2296: n_p\, 
2297: m_e\, c^3\, \gamma^2\,\delta^4  \over 4\, \pi\, D_L^2}\, , 
2298: \label{RBflux} 
2299: \end{equation} 
2300: where $\rm r_e=2.18\, fm$ is the classical radius of the electron 
2301: and $\alpha\approx 1/137$ is the fine structure constant. 
2302: For $\rm N_{CB}= 6\times 10^{50}$, $\gamma=\delta=10^3$, 
2303: $\rm z=1$ and $\rm n_p=1\, cm^{-3}$, the above expression yields 
2304: $3.8\times 10^{-11}$ erg/(cm$^2$ s). 
2305: This radiation has the same dependence on $\gamma$ and $\delta$ 
2306: as the synchrotron radiation of Eq.~(\ref{flux}), but it is some two orders 
2307: of magnitude smaller. 
2308: This spectrum extends to $\rm\!\sim\! m_e\,\delta/(1+z)\sim 250$ 
2309: MeV; it may be a contribution to the very high energy 
2310: $\gamma$-rays observed by EGRET and COMPTEL in some GRBs. 
2311:  
2312: \subsection*{Atomic transitions} 
2313:  
2314: The recombination and de-excitation of the CB atoms 
2315: excited and dissociated by Coulomb collisions with 
2316: the ISM particles contributes to the AG with a power: 
2317: \begin{equation} \rm 
2318: {dF_C\over dt\, d\Omega} \approx {8\, \pi \, r_e^2\, n_p\, m_e\, c^3\, 
2319: N_{CB}\, ln[\Lambda] \, \gamma\,\delta^4 \over 4\, \pi\, D_L^2}\, , 
2320: \label{Cflux} 
2321: \end{equation} 
2322: where $\rm ln[\Lambda]$ is the Coulomb logarithm, 
2323: \begin{equation} \rm 
2324: ln[\Lambda]= ln \left[ {\sqrt{2\, m_e\, c^2\, \gamma^2\, T_{max}} \over I} 
2325: \right ]\, , 
2326: \label{Lambda} 
2327: \end{equation} 
2328: $\rm I=13.6$ eV is the 
2329: ionization energy of hydrogen, and $\rm T_{max}$ is the maximum kinetic 
2330: energy that can be imparted to a stationary electron in a single collision 
2331: with a relativistic particle of mass $\rm M$: 
2332: \begin{equation} 
2333: \rm T_{max}= {2\, m_e\, c^2\, \gamma^2 \over 1+2\, \gamma^2\, m_e/M + 
2334:              (m_e/M)^2}\, . 
2335: \label{Tmax} 
2336: \end{equation} 
2337: For $\rm N_{CB}= 6\times 10^{50}$, $\gamma=\delta=10^3$, 
2338: $\rm n_p=1$ cm$^{-3}$ and 
2339: $\rm z=1$,  Eq.~(\ref{Cflux})  yields $7.15\times 10^{-11}$ erg/(cm$^2$ s), 
2340:  which 
2341: is negligible with respect to the synchrotron radiation of Eq.~(\ref{flux}). 
2342: This power is radiated mostly as synchrotron emission from the 
2343: knocked-on electrons and by line emission  from hydrogen recombination 
2344: (boosted to X-ray energies). 
2345:  
2346:  
2347: \subsection*{Inverse Compton scattering} 
2348:  
2349: The scattering of the CBs' electrons 
2350: on ambient starlight and the cosmic 
2351: background radiation (CBR) produces a radiated power: 
2352: \begin{equation} 
2353: \rm {dF_{CBR} \over dt\, d\Omega} \approx {32\, \pi \, r_e^2\, 
2354: \rho_\gamma\, 
2355: c\, N_{CB}\, \gamma^2\,\delta^4  \over 36\, \pi\, D_L^2}\, , 
2356: \label{ICSflux} 
2357: \end{equation} 
2358: where $\rho_\gamma$ is the energy density of the radiation field 
2359: ($\rm \rho_\gamma= 0.24\,(1+z)^4\, eV\, cm^{-3}$ for 
2360: the CBR). For $\rm N_{CB}= 6\times 10^{50}$, $\gamma=\delta=10^3$ 
2361: and $\rm z=1$, the above expression 
2362: yields $1.5\times 10^{-14}$ erg/(cm$^2$ s) for the CBR contribution, 
2363: which is negligible relative to synchrotron radiation 
2364: up to a very late afterglow phase, 
2365: when $\rm n_e$ may be very small. The contribution of 
2366: Compton scattering to a polarization of the signals, 
2367: on the other hand, may not be negligible (Shaviv and Dar, 1995). 
2368:  
2369: Inverse Compton scattering of the CBs' high energy electrons, with 
2370: a proper spectrum $\rm dn_e/dE \sim E_e^{-3.2}$, on their self-produced 
2371: synchrotron and thermal bremsstrahlung radiation produces very 
2372: high energy photons with a spectrum $\rm dn_\gamma/dE\sim E^{-2.1}$. 
2373: For distant GRBs, this 
2374: spectrum is cutoff at sub TeV energies by pair 
2375: production on the infrared background radiation. But, for very nearby 
2376: GRBs, it may be observable up to extremely high energies, larger than those 
2377: observed from blazars. 
2378:  
2379: \section*{Appendix II: Ever expanding ``cannonballs''} 
2380:  
2381: It is very instructive to study the possibility that CBs, 
2382: instead of reaching an asymptotic radius, would 
2383: continue to expand significantly. 
2384: To find $\rm\gamma(t)$ in this case, Eq.~(\ref{Roft}) 
2385: is to be substituted by: 
2386: \begin{equation} 
2387: \rm dR=\beta_{exp}\;{dx\over\gamma}, 
2388: \label{Rofx} 
2389: \end{equation} 
2390:  while Eqs.~(\ref{dgamma},  \ref{dxsn}) remain unchanged. 
2391: Insert Eq.~(\ref{Rofx}) into Eq.~(\ref{dgamma}) and integrate, to 
2392: obtain, for constant $\rm n_p$: 
2393: \begin{eqnarray} 
2394: \rm R^3(\gamma)&=&\rm R^3_{trans}+\widehat R^3_{max}, 
2395: \left[{1\over\gamma^2}-{1\over\gamma_0^2}\right]\nonumber\\ 
2396: \rm \widehat R^3_{max}&\equiv&\rm {3\,N_{CB}\,\beta_{exp}\over  
2397: 2\,\pi\,n_p}\, . \label{Rhat} 
2398: \end{eqnarray} 
2399: The CBs reach an asymptotic radius as $\gamma\to 1$, which bears 
2400: some resemblance to that of Eq.~(\ref{Rinfinity}), but it is reached much later 
2401: and it is much larger (e.g. $\rm \widehat R\sim 3.8\times 10^{17}$ cm for 
2402: $\rm n_p=10^{-3}$ cm$^{-3}$, $\rm\beta_{trans}=1/(3\sqrt{3})$, and 
2403:  $\rm N_{CB}=6\times 10^{50}$). 
2404:  
2405: Upon insertion of Eq.~(\ref{Rhat}) into Eq.~(\ref{dgamma}), we obtain 
2406: the analogue of Eq.~(\ref{gamoft}): 
2407: \begin{eqnarray} 
2408: -\rm\int_\gamma^{\gamma_0} &\rm d\gamma&\rm 
2409: {1+\theta^2\gamma^2\over \gamma^{8/3}} 
2410: \left[1-{\gamma^2\over\gamma_0^2}\right]^{-2/3} 
2411: ={2\,c\, t\over 3\, (1+z)\, \widehat x_\infty} 
2412: \nonumber\\ 
2413: \rm\widehat x_\infty&\equiv& 
2414: \rm{N_{CB}\over\pi\,\widehat R_{max}^2\, n_p} 
2415: \label{gamoftbis} 
2416: \end{eqnarray} 
2417: where we have neglected $\rm R^3_{trans}/\widehat R^3_{max}$. 
2418: The integral can be done analytically, but is not compact. 
2419:  
2420: For late times, when $\gamma\theta\ll 1$, Eq.~(\ref{gamoftbis}) implies that 
2421: $\rm\gamma\propto t^{-3/5}$.   According to Eq.~(\ref{fluxdensity}), then, 
2422: the AG light curve approaches: 
2423: \begin{eqnarray} 
2424: \rm F_\nu (t) &\propto& \rm t^{-\hat\tau};\nonumber\\ 
2425: \widehat\tau&=&\rm{9\over 5}\;(1+\alpha)\simeq 3.8. 
2426: \label{lateAGbis} 
2427: \end{eqnarray} 
2428: This behaviour cannot be reconciled with the data, 
2429: as explained in the text. 
2430: 
2431: \section*{Appendix III: gravitational lensing of moving CBs} 
2432: 
2433: The phenomenon of gravitational lensing is well known. 
2434: A lensing object of mass $\rm M$ has a Schwarzschild radius 
2435: $\rm R_S=2\,G_N\,M/c^2$. If $\rm D_A(z)$ is the angular distance 
2436: from source to observer and $\rm x$ is the fractional distance to the 
2437: lens, the Einstein radius of the system is $\rm R_E=[2\,R_S\,D_A\,x(1-x)]^{1/2}$. 
2438: As the lensing object crosses close to the line of sight (or, in our case, as 
2439: the line of sight to the fast-moving CB passes close to the lensing object) 
2440: the amplification $\rm A$ is: 
2441: \begin{eqnarray} 
2442: \rm A &=& \rm {2+u^2\over u\,\sqrt{u+u^2}}\, ,\nonumber\\ 
2443: \rm u(t)&\equiv&\rm \left[u_{min}^2+ 
2444: {[t-t_{max}]^2\over \tau^2}\right]^{1/2}\, ,\nonumber\\ 
2445: \rm \tau(t)&\equiv&\rm {R_E\over v_\perp(t)}\, ,\nonumber\\ 
2446: \rm v_\perp(t)&\simeq& \rm 
2447: c\,{\theta\,\gamma(t)\,\delta(t)\over 1+z}\, , 
2448: \label{Gravamp} 
2449: \end{eqnarray} 
2450: where $\rm u_{min}$ is the minimum distance to the lens, in Einstein radii, 
2451: of the line of sight to the CB during its motion. 
2452:  
2453: Gravitational lensing of a moving CB is peculiar in two ways: 
2454: the apparent velocity is superluminal, and 
2455: the time ``width''  of the effect, $\rm \tau(t)$, is itself  
2456: time dependent, since the CB is decelerating as the lensing occurs: 
2457: \begin{equation} 
2458: \rm \tau(t)=\tau(0)\;{v_\perp(0)\over v_\perp(t)}=\tau(0)\; 
2459: \left[{\gamma_0\over\gamma(t)}\right]^2\; 
2460: {1+[\theta\,\gamma(t)]^2\over 1+[\theta\,\gamma_0]^2}\, . 
2461: \label{tauoft} 
2462: \end{equation} 
2463: 
2464: For a solar-mass star placed halfway to a GRB at $\rm z=1$, the
2465: typical duration of a lensing event is $\tau(0)\sim 1$ hour.
2466: The average Einstein radius of a solar-mass star placed somewhere
2467: on the way to such a location is 
2468: $\rm R_E[\odot]\!=\!R_E\,(M_\odot/M)^{1/2}$ 
2469: $\rm 2\, \langle [x(1-x)]^{1/2}\rangle 
2470: \!\sim\! 1860$ AU. What is the optical depth (or apriori probability), $\epsilon$, 
2471: for an observable lensing by such an object? 
2472: Consider lensing during the first 10 days of an AG, when its fluence
2473: is relatively high and during which, for typical parameters,
2474: the CBs travelled a (local) distance $\rm x\!\simeq\! 2$ kpc. 
2475: The apparent transverse distance is 
2476: $\rm x_\perp\!=\! x\,\theta/(1+z)\!\simeq\! 1$ pc. 
2477: The average luminosity density of the local Universe is 
2478: $\rm \rho\sim (1.8\pm 0.2)\,h\times10^8\,L_\odot/Mpc^3$ and the mass to luminosity 
2479: ratio of star populations is $\rm M/L\sim 5$ to 10 in solar units, 
2480: so that the number density of ``typical'' solar-mass stars is 
2481: $\rm n_\odot\!\sim\! \rho\,M/L\!\sim\! 8.8\times 10^8/Mpc^3$ for 
2482: $\rm h\!\sim\! 0.65$. 
2483: The optical depth is: 
2484: \begin{equation} 
2485: \rm \epsilon=x_\perp\, R_\odot\,D_A\,n_\odot\,\langle (1+z)^3 \rangle\, . 
2486: \label{epsilon} 
2487: \end{equation} 
2488: In the interval extending to $\rm z=1$, and for 
2489: our adopted cosmological parameters, the volume average 
2490: $\langle (1+z)^3 \rangle$  is $\sim 5$. 
2491: Thus we obtain $\epsilon\sim (4\pm 2)$\%, which makes
2492: the lensing effects hopefully visible. 
2493:  
2494: A rough estimate of $\epsilon$ taking into account that stars gather in 
2495: galaxies gives a similar result. Let the surface-number density of stars in 
2496: a galaxy, as a function of distance to the centre, be approximated 
2497: by $\rm \Sigma_*(r)\!=\!\Sigma_*(0)\, e^{-r/h}$, with $\rm h\!\sim\! 5$ kpc. 
2498: For a reference galaxy with $\rm N_*\!=\!10^{11}$ stars $\Sigma(0)\!\simeq\! 640$  
2499: pc$^{-2}$. Define a galaxy's effective lensing radius so that 
2500: $\rm \Sigma_*(r_{eff})\,x_\perp\,R_E\!\sim\! 1$. For the quoted values of 
2501: $\rm x_\perp$ and $\rm R_E$ this means $\rm r_{eff}\!\sim\! 9$ kpc. 
2502: Approximate the surface density of galaxies at $\rm z\!<\! 1$ by 
2503: the observed value for galaxies with $\rm R$-magnitude below 25: 
2504: $\rm\Sigma_G\!\sim\! 4.6\times 10^8$ rad$^{-2}$ (Casertano et al. 2000). 
2505: The lensing probability at an angular distance $\rm D_A$ is then 
2506: $\rm \epsilon\!\sim\!\pi\,r_{eff}^2\,\Sigma_G/D_A^2\!\sim\! 4$\%. 
2507: 
2508:  
2509: \newpage 
2510: %\pagebreak 
2511: \vspace{.5  cm} 
2512:  
2513: {\bf 
2514: \noindent 
2515: Table I - Gamma-ray bursts of known  redshift} 
2516: %\vskip 0.2 true cm 
2517: \begin{table}[h] 
2518: %\vskip 0.1 true cm 
2519: \hspace{0.3cm} %if you want to center your table act on this argument 
2520: \begin{tabular}{|l|c|c|c|c|c|c|} 
2521: \hline 
2522:  
2523: % \multicolumn{8}{GRB with redshifts}\\ 
2524: \hline 
2525: \hline 
2526: GRB   &z & D$_{\rm L}$  & ${\rm F_\gamma}$ 
2527: &${\rm E_\gamma}$ &  ${\rm R[HG]}$\\ 
2528: \hline 
2529: 970228$^1$      &0.695  &4.55  &1.1   & 0.22      & 25.2$^{18}$  \\ 
2530: 970508$^2$      &0.835  &5.70  &0.49  & 0.10      & 25.0$^{19}$  \\ 
2531: 970828$^3$      &0.957  &6.74  &9.6   & 2.06      & 24.5$^{20}$  \\ 
2532: 971214$^4$      &3.418  &32.0  &0.94  & 2.11      & 25.6$^{21}$  \\ 
2533: 980425$^5$      &.0085  &.039  &0.44  &8.1E-6     & 14.3$^{22}$  \\ 
2534: 980613$^6$      &1.096  &7.98  &0.17  & 0.61      & 24.0$^{23}$  \\ 
2535: 980703$^7$      &0.966  &6.82  &2.26  & 1.05      & 22.6$^{24}$  \\ 
2536: 990123$^8$      &1.600  &12.7  &26.8  &19.80      & 23.9$^{25}$  \\ 
2537: 990510$^9$      &1.619  &12.9  &6.55  & 5.00      & 27.0$^{26}$  \\ 
2538: 990712$^{10}$   &0.434  &2.55  &6.5   & 0.53      & 21.8$^{27}$  \\ 
2539: 991208$^{11}$   &0.70   &4.64  &10.0  & 1.51      & 24.4$^{28}$  \\ 
2540: 991216$^{12}$   &1.020  &7.30  &19.4  & 5.35      & 24.8$^{29}$  \\ 
2541: 000131$^{13}$   &4.500  &44.4  & 4.2  &11.60      & 27.8$^{30}$  \\ 
2542: 000301c$^{14}$  &2.040  &17.2  &0.41  & 0.46      & 28.0$^{31}$  \\ 
2543: 000418$^{15}$   &1.119  &8.18  &2.0   & 0.82      & 23.9$^{32}$  \\ 
2544: 000926$^{16}$   &2.066  &17.4  &2.20  & 2.60      & 25.6$^{33}$  \\ 
2545: 010222$^{17}$   &1.474  &11.5  &12.0  & 7.80      & 25.9$^{34}$  \\ 
2546:  
2547: \hline 
2548: \hline 
2549: \end{tabular} 
2550: \end{table} 
2551: \vskip -0.3 true cm 
2552: \noindent 
2553: {\bf Comments:} $\rm z$: Redshift. $\rm D_L$: Luminosity distance in Gpc, 
2554: for $\rm \Omega_m=0.3, 
2555: \; \Omega_\Lambda=0.7$ and ${\rm H_0=65\, km\, s^{-1}\,Mpc^{-1}}$. 
2556: $\rm F_\gamma$: BATSE $\gamma$-ray fluences in units of 
2557: $10^{-5}$ erg cm$^{-2}$. $\rm E_\gamma$: (Equivalent spherical) energy in 
2558: units of  $10^{53}$ ergs. 
2559: $\rm R[HG]$: R-magnitude of the host galaxy, except for GRB 990510, for 
2560: which the V-magnitude is given, corrected for galactic extinction\\ 
2561: \noindent 
2562: {\bf References}: 
2563: 1: Djorgovski et al.~1999a; 
2564: 2: Metzger et al.~1997; 
2565: 3: Djorgovski et al.~2000; 
2566: 4: Kulkarni et al.~1998b; 
2567: 5: Tinney et al.~1998; 
2568: 6: Djorgovski et al.~1998a; 
2569: 7: Djorgovski et al.~1998b; 
2570: 8: Kelson et al.~1999; 
2571: 9: Vreeswijk et al.~1999a; 
2572: 10: Hjorth et al.~1999; 
2573: 12: Vreeswijk et al.~1999b; 
2574: 13: Andersen et al.~2000; 
2575: 14: Feng et al.~2000; 
2576: 15: Bloom et al.~2001; 
2577: 16: Fynbo et al.~2001; 
2578: 17: Jha et al.~2001; Fruchter et al.~2001; 
2579: 18: Fruchter et al.~1999b; 
2580: 19: Pian 2001; 
2581: 20: Djorgovskyet al.~2001 
2582: 21: Odewahn et al.~1998; 
2583: 22: Galama et al.~1998a; 
2584: 23: Djorgovski et al.~2000; 
2585: 24: Bloom et al.~1998b; 
2586: 25: Bloom et al.~1999b; 
2587: 26: Pian 2001; 
2588: 27: Hjorth et al.~1999; 
2589: 28: Diercks et al.~2000; 
2590: 29: Djorgovski et al.~1999b; 
2591: 30: Andersen et al.~2000; 
2592: 31: Smette et al.~2000; 
2593: 32: Bloom et al.~2000; 
2594: 33: Fynbo et al.~2001; 
2595: 34: Fruchter et al.~2001. 
2596: \newpage 
2597:  
2598:  
2599:  
2600: {\bf 
2601: \noindent 
2602: Table II -  The CB, host-galaxy, and extinction parameters} 
2603: %\vskip 0.2 true cm 
2604: \begin{table}[h] 
2605: %\vskip 0.1 true cm 
2606: \hspace{+.1cm} %if you want to center your table act on this argument 
2607: \begin{tabular}{|l|c|c|c|c|c|c|c|l|} 
2608: \hline 
2609: % \multicolumn{8}{GRB with redshifts}\\ 
2610: \hline 
2611: GRB   &$\gamma_0 $ &$\alpha$& $\theta\, $ &$\rm x_\infty\, $  
2612: &   R[HG]  & $\rm A_{SN}$ \\ 
2613: \hline 
2614: 970228        & 540   &1.10  &1.686    & 0.155    & 25.55   &0.50 \\ 
2615: 971214        & 999   &1.20  &0.708    & 0.373    & 25.69   &0.94 \\ 
2616: 980613        & 509   &1.09  &1.619    & 0.241    & 24.07   &0.82 \\ 
2617: 980703        & 779   &1.08  &0.953    & 0.344    & 22.54   &0.88 \\ 
2618: 990123        &1325   &1.09  &0.420    & 0.954    & 23.90   &0.96 \\ 
2619: 990510        & 991   &1.10  &0.261    & 0.777    & 27.80   &0.50 \\ 
2620: 990510$^a$    & 907   &1.08  &0.318    & 0.504    & 27.80   &0.40 \\ 
2621: 990712        & 948   &1.09  &0.750    & 1.191    & 21.93   &0.50 \\ 
2622: 990712$^a$    & 957   &1.08  &0.863    & 1.319    & 22.57   &0.40 \\ 
2623: 991208        &1034   &1.26  &0.100    & 1.357    & 24.81   &0.80 \\ 
2624: 991216        & 972   &1.09  &0.375    & 0.953    & 24.64   &0.80 \\ 
2625: 000131        &1200   &1.26  &0.100    & 0.793    & 27.80   &0.93 \\ 
2626: 000301c       &1040   &1.19  &2.223    & 0.141    & 28.00   &0.90 \\ 
2627: 000418        &1017   &1.17  &0.970    & 1.961    & 23.74   &0.92 \\ 
2628: 000926        & 760   &1.20  &0.740    & 0.133    & 25.63   &0.96 \\ 
2629: 010222        & 1178  &1.10  &0.465    & 1.026    & 25.76   &0.94 \\ 
2630:               &       &      &         &          &         &     \\ 
2631: 970508$^b$    & 1123   &1.10   & 3.51   & 0.293     & 24.69   &0.94 \\ 
2632: 980425$^b$    & 769   &1.10   & 8.30   & 0.252    & 14.30   &0.93 \\ 
2633: \hline 
2634: \hline 
2635:  
2636: \end{tabular} 
2637: \end{table} 
2638: \vskip -0.3 true cm 
2639: \noindent 
2640: {\bf Comments:} $\gamma_0$: Initial Lorentz factor. 
2641: $\theta$: Viewing angle relative to the CB line of motion, 
2642:  in milliradians. 
2643:  $\rm x_\infty$: CB slow-down parameter, in Mpc 
2644: ($\gamma=\gamma_0/2$ at $\rm x= x_\infty/\gamma_0$). 
2645: $\rm R[HG]$: Fitted value of the 
2646: R-magnitude of the host galaxy (except for GRB 990712, for 
2647: which also the V-magnitude is given) not 
2648: corrected for galactic extinction. 
2649: $\rm A_{SN}$: Attenuation of the SN1998bw-like contribution due to  
2650: galactic extinction. 
2651: $a$: V-band afterglow parameters.  
2652: $b$: Two GRBs are special: GRB 970508 is fit with two constant
2653: ISM densities, the  $\rm x_\infty$ quoted value corresponds to the
2654: initial one; in GRB 980425 
2655: the SN outshines the CBs in the optical, the fit is to the X-ray AG, 
2656: $\alpha=1.1$ was assumed, and the parameter determinations 
2657: are very imprecise. 
2658:  
2659:  
2660: \noindent 
2661: {\bf References}:\\ 
2662: {\bf GRB 970228}: 
2663: Castander, et al.~1999a; 
2664: Castander, et al.~1999b; 
2665: Djorgovski et al.~1999a; 
2666: Fruchter et al.~1997a; 
2667: Galama et al.~1997; 
2668: Galama et al.~2000; 
2669: Garcia et al.~1998; 
2670: Guarnieri, et al.~1997; 
2671: Metzger et al.~1997a; 
2672: Pedichini et al.~1997; 
2673: Sahu et al.~1997a; 
2674: Sahu et al.~1997b; 
2675: van Paradijs et al.~1998. \\ 
2676: {\bf GRB 970508}: 
2677: Bloom et al.~1998b; 
2678: Castro-Tirado et al.~1998b; 
2679: Chevalier \& Ilovaisky~1997; 
2680: Djorgovski et al.~1997; 
2681: Fruchter et al.~1997b; 
2682: Fruchter et al.~2000; 
2683: Galama et al.~1998b; 
2684: Metzger et al.~1997b; 
2685: Pedersen et al.~1998a; 
2686: Schaefer et al.~1997; 
2687: Sokolov et al.~1997; 
2688: Sokolov et al.~1998; 
2689: Zharikov et al.~1998. \\ 
2690: {\bf GRB 971214}: 
2691: Diercks et al.~1998; 
2692: Halpern et al.~1998b; 
2693: Kulkarni et al.~1998b.\\ 
2694: {\bf GRB 980613}: 
2695: Djorgovski et al.~1998a; 
2696: Djorgovski et al.~2000; 
2697: Halpern et al.~1998a; 
2698: Hjorth et al.~1998.\\ 
2699: {\bf GRB 980703}: 
2700: Bloom et al.~1998b; 
2701: Castro-Tirado et al.~1999a; 
2702: Holland et al.~2000b; 
2703: Holland et al.~2001; 
2704: Pedersen et al.~1998b; 
2705: Sokolov et al.~1998; 
2706: Vreeswijk et al.~1999c; 
2707: Zapatero Osorio et al.~1998.\\ 
2708: {\bf GRB 990123}: 
2709: Castro-Tirado et al.~1999b; 
2710: Fruchter et al.~1999a; 
2711: Galama et al.~1999a; 
2712: Holland et al.~2000a; 
2713: Kulkarni et al.~1999. \\ 
2714: {\bf GRB 990510}: 
2715: Beuermann et al.~1999; 
2716: Covino et al.~1999; 
2717: Fruchter et al.~1999c; 
2718: Galama et al.~1999b; 
2719: Harrison et al.~1999; 
2720: Holland et al.~2000a; 
2721: Marconi et al.~1999a,b; 
2722: Pietrzy{\'n}ski \& Udalski~1999a,b,c; 
2723: Stanek et al.~1999. \\ 
2724: {\bf GRB 990712}: 
2725: Hjorth et al.~2000a; 
2726: Sahu, et al.~2000\\ 
2727: {\bf GRB 991208}: 
2728: Castro-Tirado et al.~2001; 
2729: Sagar et al.~2000a.\\ 
2730: {\bf GRB 991216}: 
2731: Djorgovski et al.~1999b; 
2732: Garnavich et al.~2000a; 
2733: Halpern et al.~2000a; 
2734: Sagar et al.~2000a.\\ 
2735: {\bf GRB 000131}: 
2736: Andersen et al.~2000.\\ 
2737: {\bf GRB 000301c}: 
2738: Jensen et al.~2001; 
2739: Halpern et al.~2000b; 
2740: Garnavich et al.~2000b; 
2741: Masetti et al.~2000; 
2742: Rhoads \& Fruchter~2001; 
2743: Sagar et al.~2000b; 
2744: Veillet 2000a.\\ 
2745: {\bf GRB 000418}: 
2746: Berger et al.~2000; 
2747: Henden et al.~2000; 
2748: Klose et al.~2000;  
2749: Metzger \& Fruchter~2000.\\ 
2750: {\bf GRB 000926}: 
2751: Fynbo et al.~2001; 
2752: Halpern et al.~2000C 
2753: Harrison et al.~2001; 
2754: Hjorth et al.~2000b; 
2755: Price et al.~2001; 
2756: Sagar et al.~2001a; 
2757: Veillet~2000b. \\ 
2758: {\bf GRB 010222}: 
2759: Cowsik et al.~2001; 
2760: Jha et al.~2001; 
2761: Fruchter et al.~2001; 
2762: Garnavich et al.~2001; 
2763: Masetti et al.~2001; 
2764: Oksanen et al.~2001; 
2765: Orosz et al.~2001; 
2766: Price et al.~2001; 
2767: Sagar et al.~2001b; 
2768: Stanek et al.~2001; 
2769: Valentini et al.~2001; 
2770: Watanabe et al.~2001; 
2771: Veillet~2001.\\ 
2772:  
2773: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
2774:  
2775:  
2776: \vskip 0.3 true cm 
2777:  
2778: {\bf 
2779: \noindent 
2780: Table III - The rest frame GRB energy and X-ray AG of GRBs with  
2781: measured redshift and X-ray AG 
2782: from their observed $\gamma$-ray fluence and the optical AG parameters} 
2783: %\vskip 0.2 true cm 
2784: \begin{table}[h] 
2785: %\vskip 0.1 true cm 
2786: \hspace{-.1cm} %if you want to center your table act on this argument 
2787: \begin{tabular}{|l|c|c|c|c|c|c|c|c|c|} 
2788:  
2789: \hline 
2790: % \multicolumn{8}{GRB with redshifts}\\ 
2791: \hline 
2792: GRB   &z &$\rm D_L$  &${\rm F_\gamma}$  
2793: &$\gamma_0$&$\delta_0$&$\rm q $&$\rm E_\gamma^{CB}\,$  \\ 
2794: \hline 
2795: 970228   &0.695    &4.55  &1.10   & 540  & 591  & 0.86   &0.78 \\ 
2796: 970508   &0.835    &5.70  &1.10   & 1123  & 137  & 1.26   &1.47 \\ 
2797: 970828   &0.957    &6.74  &1.10   &1153  &1160  & 0.77   &1.34 \\ 
2798: 971214   &3.418    &32.0  &0.94   & 999  &1331  & 1.28   &1.11 \\ 
2799: 980613   &1.096    &7.98  &0.17   & 509  & 606  & 1.34   &1.74 \\ 
2800: 990123   &1.600    &12.7  &26.8   &1325  &2023  & 1.45   &1.84 \\ 
2801: 990510   &1.619    &12.9  &6.55   & 991  &1858  & 2.60   &0.78 \\ 
2802: 991216   &1.020    &7.30  &19.4   & 972  &1716  & 0.38   &1.54 \\ 
2803: 000926   &2.066    &17.4  &2.20   & 761  &1115  & 2.12   &0.60 \\ 
2804: 010222   &1.474    &11.5  &12.0   &1109  &1812  & 1.43   &1.31 \\ 
2805:  
2806: \hline 
2807: \hline 
2808: \end{tabular} 
2809: \end{table} 
2810: \vskip -0.3 true cm 
2811: \noindent 
2812: {\bf Comments:} $\rm D_L$: Luminosity distance, for $\rm \Omega_m=0.3, 
2813: \; \Omega_\Lambda=0.7$ and ${\rm H_0=65\, km\, s^{-1}\,Mpc^{-1}}$, in Gpc.\\ 
2814: $\rm F_\gamma$: BATSE/BeppoSAX  $\gamma$-ray fluences in units of 
2815: $10^{-5}$ erg cm$^{-2}$. $\rm \delta_0$: Initial Doppler factor. 
2816: $\rm q$: The ratio between observed and predicted late-time X-ray fluxes in   
2817: the 2-10 keV band, for a single unextinct standard CB with  
2818: $\rm N_{CB}=6\times 10^{50}\, .$  
2819: $\rm E_\gamma^{CB}$: Energy radiated by the ensemble of CBs 
2820: in its rest frame, in units of $\rm 10^{44}\,erg\,.$  \\ 
2821:  
2822: \newpage 
2823:  
2824: \begin{thebibliography}{} 
2825:  
2826:  
2827: \bibitem{} 
2828: Akerlof C., et al., 1999, Nature 398, 400 
2829: \bibitem{} 
2830: Andersen M.I., et al., 2000, A\&A 364, 54L 
2831: \bibitem{} 
2832: Bendarz J., Ostrowski M., 1998, Phys. Rev. Lett.  80, 3911 
2833: \bibitem {} 
2834: Berger E.,  et al., 2000, ApJ 545, 56 
2835: \bibitem {} 
2836: Berger E.,  et al., 2001, astro-ph/0102278 
2837: \bibitem {} 
2838: Beuermann K., et al., 1999, A\&A 352, L26 
2839: \bibitem{} 
2840: Bhargavi S.G., Cowsik R., 2000, ApJ 545, L77 
2841: \bibitem {} 
2842: Bjornsson G., et al., 2001, ApJ, 552, 121L 
2843: \bibitem{} 
2844: Bloom J.S., et al., 1998a, ApJ 507, L25 
2845: \bibitem{} 
2846: Bloom J.S., et al., 1998b, ApJ 508, L21 
2847: \bibitem{}  
2848: Bloom J.S., et al., 1999a, Nature 401, 452 
2849: \bibitem{}  
2850: Bloom J.S., et al., 1999b, ApJ 518, L1 
2851: \bibitem{} 
2852: Bloom J.S., et al., 2000, GCN Circ. 661 
2853: \bibitem{} 
2854: Bloom J.S., Djorgovski S.G., Kulkarni S.R., 2001, ApJ 554, 678 
2855: \bibitem{} 
2856: Brainerd J.J., 1992, ApJ 394, 33L 
2857: \bibitem{} 
2858: Bridle A., 2000,  http://www.cv.nrao.edu/~abridle/images.htm 
2859: \bibitem{} 
2860: Burenin R.A., et al., 1999, A\&A 344, L53 
2861: \bibitem{} 
2862: Casertano S., et al., 2000, astro-ph/0010245 
2863: \bibitem{} 
2864: Castander F., et al., 1999a, ApJ 523, 593 
2865: \bibitem{} 
2866: Castander F., et al., 1999b, ApJ 523, 602 
2867: \bibitem{} 
2868: Castro-Tirado A.J., et al., 1998a, Science 279, 1011 
2869: \bibitem{} 
2870: Castro-Tirado A.J., et al., 1998b, IAU Circ. 6848 
2871: \bibitem{} 
2872: Castro-Tirado A.J., Gorosabel J., 1999, A\&AS 138(3), 449 
2873: \bibitem{} 
2874: Castro-Tirado A.J., et al., 1999a, ApJ, 511, L85 
2875: \bibitem{} 
2876: Castro-Tirado A.J., et al., 1999b, Science 283, 2069 
2877: \bibitem{} 
2878: Castro-Tirado A.J., et al., 2001, A\&A 370, 398 
2879: \bibitem{} 
2880: Chiang J., Dermer C.F., 1997, astro-ph/9708035 
2881: \bibitem{} 
2882: Chevalier C., Ilovaisky S.A., 1997, IAU Circ. 6663 
2883: \bibitem{} 
2884: Cline T.L., et al., 1999, A\&A 138(3), 557 
2885: \bibitem{} 
2886: Corbet R., Smith D., 1999 GCN Circ. 506 
2887: \bibitem{} 
2888: Costa E., et al., 1997, Nature 387, 783   
2889: \bibitem{} 
2890: Covino S., et al., 1999, GCN Circ. 330 
2891: \bibitem{} 
2892: Cowsik R., et al., 2001, astro-ph/0104363 
2893: \bibitem{} 
2894: Cranc P., et al., 1993, ApJ 402, L37  
2895: \bibitem{}
2896: Dado S., Dar A., and De R\'ujula 2002, {\it in preparation}
2897: \bibitem{} 
2898: Dal Fiume D., et al., 2000, A\&A 355, 454 
2899: \bibitem{} 
2900: Dar A., et al., 1992, ApJ 388, 164 
2901: \bibitem{} 
2902: Dar A., 1997, astro-ph/9704187 
2903: \bibitem{} 
2904: Dar A., 1998a, ApJ 500, L93 
2905: \bibitem{} 
2906: Dar A., 1998b, Proc. Les Rencontres de Physique de la Vall\'ee d'Aoste, 
2907: (ed. M. Greco) p. 23 
2908: \bibitem{} 
2909: Dar A. 1999a, A\&AS  138(3), 505  
2910: \bibitem{} 
2911: Dar A., 1999b, GCN Circ. 346 
2912: \bibitem{} 
2913: Dar A., De R\'ujula, A., 2000a, astro-ph/0008474,\\ submitted to A\&A 
2914: \bibitem{} 
2915: Dar A., De R\'ujula, A., 2000b, astro-ph/0012227,\\ submitted to A\&A 
2916: \bibitem{} 
2917: Dar A.,  De R\'ujula, A.,  2001 MNRAS 323, 391 
2918: \bibitem{} 
2919: Dar A.,  De R\'ujula, A., 2001a, astro-ph/0102115,\\ submitted to A\&A 
2920: \bibitem{} 
2921: Dar A.,  De R\'ujula, A., 2001b, astro-ph/0105094,\\ submitted to Phys. Rev. 
2922: \bibitem{} 
2923: Dar A., Plaga R., 1999, A\&A 349, 259 
2924: \bibitem{} 
2925: De R\'ujula A., 1987, Phys. Lett.  193, 514 
2926: \bibitem{} 
2927: Diercks A.H., 1998, ApJ 503, L105 
2928: \bibitem{} 
2929: Diercks A., et al., 2000,  GCN  Circ. 764 
2930: \bibitem{} 
2931: Djorgovski S.G.,  et al., 1997, Nature, 387, 876 
2932: \bibitem{} 
2933: Djorgovski S.G., et al., 1998a, GCN Circ. 117, 189 
2934: \bibitem{} 
2935: Djorgovski S.G., et al., 1998b, ApJ 508, L17 
2936: \bibitem{} 
2937: Djorgovski S.G., et al., 1999a, GCN Circ. 289 
2938: \bibitem{} 
2939: Djorgovski S.G., et al., 1999b, GCN Circ. 510 
2940: \bibitem{} 
2941: Djorgovski S.G. et al., 2000,  astro-ph/0008029 
2942: \bibitem{} 
2943: Djorgovsky, S.G., et al., 2001, astro-ph/0107539 
2944: \bibitem{} 
2945: Djorgovsky, S.G., et al., 2001, astro-ph/0107539 
2946: \bibitem{} 
2947: Dodonov S.N., et al., 1999, GCN Circ. 475 
2948: \bibitem{} 
2949: Dunne B.C., Points, S.D., Chu, Y., 2001, astro-ph/0104212 
2950: \bibitem{} 
2951: Feng M.L., Wang L., Wheeler J., 2000, GCN Circ. 607 
2952: \bibitem{} 
2953: Fenimore E., Ramirez-Ruiz E., 2000,  astro-ph/0004176 
2954: \bibitem{} 
2955: Frail D.A., et al., 2001, astro-ph/0102282 
2956: \bibitem{} 
2957: Frontera F.,  et al., 1998, ApJ 493, 67L 
2958: \bibitem{} 
2959: Fruchter A.S., et al., 1997a, IAU Circ. 6747 
2960: \bibitem{} 
2961: Fruchter A.S., et al., 1997b, IAU Circ. 6647 
2962: \bibitem{} 
2963: Fruchter A.S., et al., 1999a, ApJ 519, L13 
2964: \bibitem{} 
2965: Fruchter A.S., et al., 1999b, ApJ 516, 683 
2966: \bibitem{} 
2967: Fruchter A.S., et al., 2000, ApJ 545, 664 
2968: \bibitem{} 
2969: Fruchter A.S., et al,, GCN Circ. 1087 
2970: \bibitem{} 
2971: Fynbo J.U., et al., 2000, ApJ 542, 89L 
2972: \bibitem{} 
2973: Fynbo J.U., et al., 2001, A\&A 373, 796 
2974: \bibitem{} 
2975: Gabudza D.D., et al., 1993, ApJ 410, 39 
2976: \bibitem{} 
2977: Galama T.J., et al., 1997, Nature 387, 479 
2978: \bibitem{} 
2979: Galama T.J., et al., 1998a, Nature 395, 670 
2980: \bibitem{} 
2981: Galama T.J., et al., 1998b, ApJ, 497, L13 
2982: \bibitem{} 
2983: Galama T.J., et al., 1998c, ApJ 500, L97 
2984: \bibitem{} 
2985: Galama T.J., et al., 1999a, GCN Circ. 313 
2986: \bibitem{} 
2987: Galama T.J., et al., 1999b, GCN Circ. 388 
2988: \bibitem{} 
2989: Galama T.J., et al., 2000, ApJ 536 185 
2990: \bibitem{} 
2991: Garcia M.R., et al., 1998, ApJ 500, L105 
2992: \bibitem{} 
2993: Garnavich P.M., et al.,  2000a, ApJ 543, 61 
2994: \bibitem{} 
2995: Garnavich P.M., et al., 2000b, ApJ 544, L11 
2996: \bibitem{} 
2997: Garnavich P.M., Loeb A., Stanek K.Z., 2000c, ApJ 544, L1 
2998: \bibitem{} 
2999: Garnavich P., Quinn, J., Stanek, K.Z., 2001, GCN Circ. 1009 
3000: \bibitem{} 
3001: Ghisellini G., et al. 1999, ApJ 517, 168 
3002: \bibitem{} 
3003: Ghisellini G., Celotti A., 2001, astro-ph/0103007 
3004: \bibitem{} 
3005: Kraft R.P., et al., astro-ph/0111340 
3006: \bibitem{} 
3007: Goodman J., 1986, ApJ 308, 47L 
3008: \bibitem{} 
3009: Goodman J., Dar A., Nussinov S., 1987, ApJ 314, L7 
3010: \bibitem{} 
3011: Guarnieri A., et al., 1997, A\&A, 328, L13 
3012: \bibitem{} 
3013: Halpern J.P., et al., 1998a, GCN Circ. 134 
3014: \bibitem{} 
3015: Halpern J.P., et al., 1998b, AAS, 192.3311 
3016: \bibitem{} 
3017: Halpern J.P., et al., 1999, ApJ, 517, 105L 
3018: \bibitem{} 
3019: Halpern J.P., et al., 2000a, ApJ 543, 697 
3020: \bibitem{} 
3021: Halpern J.P., et al., 2000b, GCN Circ. 578, 582, 585, 604 
3022: \bibitem{} 
3023: Halpern J.P., et al., 2000c, GCN Circ. 824, 829 
3024: \bibitem{} 
3025: Harrison F.A., et al., 1999, ApJ 523, L121 
3026: \bibitem{} 
3027: Harrison F.A., et al., 2001, astro-ph/0103377 
3028: \bibitem{} 
3029: Heger A., Langer N., Woosley S.E., 2000, ApJ 528, 368 
3030: \bibitem{} 
3031: Henden A., et al., 2000, GCN Circ. 652 
3032: \bibitem{} 
3033: Higdon J.C., Lingenfelter R.E., 1980, ApJ 239, 867 
3034: \bibitem{} 
3035: Higdon J.C., Lingenfelter R.E., Ramaty R., 1998, ApJ  509, L33 
3036: \bibitem{} 
3037: Hjorth J., et al., 1998, GCN Circ. 109 
3038: \bibitem{} 
3039: Hjorth J., et al., 1999, GCN Circ. 403 
3040: \bibitem{} 
3041: Hjorth J., et al., 2000a, ApJ 534, 147L 
3042: \bibitem{} 
3043: Hjorth J., et al., 2000b, GCN Circ. 809, 814 
3044: \bibitem{} 
3045: Hjorth J., et al., 2001, ApJ 552, 121L 
3046: \bibitem{} 
3047: Holland S., Hjorth J., 1999, A\&A 344, 67L 
3048: \bibitem{} 
3049: Holland S., et al., 2000a, A\&A 364, 467           
3050: \bibitem{} 
3051: Holland S., et al., 2000b, AAS 197.6303H            
3052: % \bibitem{} 
3053: % Holland S., et al., 2000c, A\&A 364, 54L 
3054: \bibitem{} 
3055: Holland S., et al., 2001a, A\&A 371, 52              
3056: \bibitem{} 
3057: Holland S., et al., 2001b, GCN Circ. 1002                 
3058: \bibitem{} 
3059: Huang Y.F., et al., 2000a, ApJ 543, 90 
3060: \bibitem{} 
3061: Huang Y.F., et al., 2000b,  MNRAS, 316, 943 
3062: \bibitem{} 
3063: In `t Zand J., et al., 2001, astro-ph/0104362 
3064: \bibitem{} 
3065: Israel G.L., et al. 1999, A\&A 348, 5L 
3066: \bibitem{} 
3067: Iwamoto K., et al., 1998, Nature 395, 672 
3068: \bibitem{} 
3069: Jensen B.L., et al., 2000, astro-ph/0005609  
3070: \bibitem{} 
3071: Jensen B.L., et al., 2001, A\&A 370, 909 
3072: \bibitem{} 
3073: Jha S., et al., 2001a, ApJ, 554, 15L 
3074: \bibitem{} 
3075: Jha S., et al., 2001b, Astro-ph/0103081 
3076: \bibitem{} 
3077: Kedziora-Chudczer L., et al., 1997, ApJ 490, 9L 
3078: \bibitem{} 
3079: Kelson D.D., et al., 1999, IAU Circ. 7096 
3080: \bibitem{} 
3081: Kennicut, R.C., Edgar, B.K., Hodge, P.W., 1989, ApJ 337, 761 
3082: \bibitem{} 
3083: Kippen R.M., et al., 1998, GCN Circ. 67 
3084: \bibitem{} 
3085: Klose S., et al., 2000, ApJ 545, 271              
3086: \bibitem{} 
3087: Kotani T., et al., 1996, PASJ 48, 619 
3088: \bibitem{} 
3089: Kulkarni S.R., et al., 1998a, Nature 395, 663 
3090: \bibitem{} 
3091: Kulkarni S.R., et al., 1998b, Nature 393, 35 
3092: \bibitem{} 
3093: Kulkarni S.R., et al., 1999 Nature 398, 389 
3094: \bibitem{} 
3095: Kumar P., Panaitescu A., 2000,  ApJ 541,9L 
3096: \bibitem{} 
3097: Lamb D.Q., 2001, AAS 198.3511L 
3098: \bibitem{} 
3099: Lingenfelter R.E., Higdon J.C., Ramaty R., 2000, astro-ph/0004166 
3100: \bibitem{} 
3101: Loeb A., Perna R., 1998, ApJ, 495, 597 
3102: \bibitem{} 
3103: Levine A.M., et al., 1996, ApJ 469, L33 
3104: \bibitem{} 
3105: MacFadyen A.I.,  Woosley S.E., 1999, ApJ 524, 262 
3106: \bibitem{} 
3107: MacFadyen A.I., Woosley S.E.,  Heger A., 2001, ApJ 550, 410 
3108: \bibitem{} 
3109: Madau P., 1998, astro-ph/9801005 
3110: \bibitem{} 
3111: Madau P., Della Valle, M., Panagia, 1998, N., astro-ph/9803284) 
3112: \bibitem{} 
3113: Marconi G., et al., 1999a, GCN Circ. 329 
3114: \bibitem{} 
3115: Marconi G., et al., 1999b, GCN Circ. 332 
3116: \bibitem{} 
3117: Margon B.A., 1984, ARA\&A 22, 507 
3118: \bibitem{} 
3119: Masetti N., et al., 2000, A\&A 359, 23L 
3120: \bibitem{} 
3121: Masetti N., et al., 2001, astro-ph/0103296 
3122: \bibitem{} 
3123: Meegan C.A., et al., 1992, Nature 355, 143 
3124: \bibitem{} 
3125: Meszaros P., 2001, Science, 291, 79 
3126: \bibitem{} 
3127: Meszaros P., Rees M.J., 1992, MNRAS 257, 29P 
3128: \bibitem{} 
3129: Meszaros P., Rees M.J., Wijers R.A.M.J., 1999, New Astron. 4, 303 
3130: \bibitem{} 
3131: Metzger M.R., et al., 1997a IAU Circ. 6631 
3132: \bibitem{} 
3133: Metzger M.R., et al., 1997a, Nature 387, 878 
3134: \bibitem{} 
3135: Metzger M.R., Fruchter, A., 2000, GCN Circ. 669 
3136: \bibitem{} 
3137: Metzger M.R., et al., 2000, GCN Circ. 733 
3138: \bibitem{} 
3139: Mirabel I.F.,  Rodriguez L.F., 1994, Nature 371, 46 
3140: \bibitem{} 
3141: Mirabel I.F., Rodriguez L.F., 1999, ARA\&A 37, 409 
3142: \bibitem{} 
3143: Moderski R., Sikora M., Bulik T., 2000, ApJ, 529, 151 
3144: \bibitem{} 
3145: Odewahn S.C., et al., 1998, ApJ 509, L5 
3146: \bibitem{} 
3147: Orosz J.A., 2001, GCN Circ. 976 
3148: \bibitem{} 
3149: Oksanen A., et al., 2001, GCN Circ. 990 
3150: \bibitem{} 
3151: Owens A., et al.,  1998, A\&A 339, L37 
3152: \bibitem{} 
3153: Paczynski B., 1986, ApJ 308, L43   
3154: \bibitem{} 
3155: Paczynski B., 1998, ApJ 494, L45   
3156: \bibitem{} 
3157: Panaitescu A., Kumar, P., 2000,  astro-ph/0010257 
3158: \bibitem{} 
3159: Pearsons  T.J., Zensus J.A., 1987, Superluminal 
3160: \newline 
3161: Radio Sources, p. 1 (Cambridge Univ. Press 1987) 
3162: \bibitem{} 
3163: Pedersen H., et al., 1998a, ApJ, 496, 311 
3164: \bibitem{} 
3165: Pedersen H., et al., 1998b, GCN Circ. 142 
3166: \bibitem{} 
3167: Pedichini F., et al. 1997 A\&A 327, 36L 
3168: \bibitem{} 
3169: Peebles P.J.E., 1993, {\it Principles of Physical Cosmology} 
3170: (Princeton University Press) 
3171: \bibitem{} 
3172: Pian E.,  1999, astro-ph/9910236 
3173: \bibitem{} 
3174: Pian E., et al., 1999, A\&AS 138(3), 463  
3175: \bibitem{} 
3176: Pian E., et al., 2001, A\&A 372, 456 
3177: \bibitem{} 
3178: Pian E., et al., 2001, astro-ph/0110051 
3179: \bibitem{} 
3180: Pietrzy{\'n}ski G., Udalski A., 1999a, GCN Circ. 316 
3181: \bibitem{} 
3182: Pietrzy{\'n}ski G., Udalski A., 1999b, GCN Circ. 319 
3183: \bibitem{} 
3184: Pietrzy{\'n}ski G., Udalski A., 1999c, GCN Circ. 328 
3185: \bibitem{} 
3186: Piran T., 1999, Phys. Rep.  314, 575 
3187: \bibitem{} 
3188: Piran T., 2000, Phys. Rep. 333, 529 
3189: \bibitem{} 
3190: Piran, T., 2001, astro-ph/0104134 
3191: \bibitem{} 
3192: Piro L., et al., 1998, A\&A 331, L41 
3193: \bibitem{} 
3194: Piro L., 2000,  astro-ph/0001436 
3195: \bibitem{} 
3196: Piro L., et al., 2001, astro-ph/0103306 
3197: \bibitem{} 
3198: Price P.A., et al., 2000, GCN Circ. 811 
3199: \bibitem{} 
3200: Price P.A., et al., 2000, astro-ph/0012303 
3201: \bibitem{} 
3202: Price P.A., et al., 2001, ApJ 549, 7L 
3203: \bibitem{} 
3204: Price P.A., et al., 2001, GCN Circ. 973 
3205: \bibitem{} 
3206: Ramirez-Ruiz, E., Fenimore E., 2000, astro-ph/0010588 
3207: \bibitem{} 
3208: Ramirez-Ruiz E., et al., 2000,  astro-ph/0012396 
3209: \bibitem{} 
3210: Rees M.J., 1966, Nature 211, 468 
3211: \bibitem{} 
3212: Reichart D.E., 1999, ApJ 521, L111  
3213: \bibitem{} 
3214: Reichart D.E., et al., 2001, ApJ 552, 57 
3215: \bibitem{} 
3216: Rhoads J.E., 1997, ApJ 488, 579 
3217: \bibitem{} 
3218: Rhoads J.E., et al., 1998, GCN Circ. 144 
3219: \bibitem{} 
3220: Rhoads J.E., 1999, ApJ 525, 737 
3221: \bibitem{} 
3222: Rhoads J.E., Fruchter A.S., 2001, ApJ, 546, 117 
3223: \bibitem{} 
3224: Rodriguez L.F.,  Mirabel I.F., 1999, ApJ 511, 398 
3225: \bibitem{} 
3226: Sagar R., et al., 2000a, BASI 28, 15 
3227: \bibitem{} 
3228: Sagar R., et al., 2000b, BASI 28, 499 
3229: \bibitem{} 
3230: Sagar, R. et al., 2001a, BASI 29, 1 
3231: \bibitem{} 
3232: Sagar, R. et al., 2001b, astro-ph/0104249 
3233: \bibitem{} 
3234: Sahu K.C., et al., 1997a, Nature 387, 476 
3235: \bibitem{} 
3236: Sahu K.C., 1997b, ApJ 489, L127 
3237: \bibitem{} 
3238: Sahu K.C., et al., 2000, ApJ 540, 74 
3239: \bibitem{} 
3240: Sari R., Piran P., Halpern J.P.,  et al., 1999, ApJ 524, L43 
3241: \bibitem{} 
3242: Schaefer B., et al., 1997, IAU Circ. 6658 
3243: \bibitem{} 
3244: Schlegel D.J., Finkbeiner D.P., Davis M., 1998, ApJ, 500, 525 
3245: \bibitem{} 
3246: Shaviv N.J.,  Dar A., 1995, ApJ 447, 863  
3247: \bibitem{} 
3248: Smette A., et al.,  2000, GCN Circ. 603 
3249: \bibitem{} 
3250: Smith D.A., et al., 2001, astro-ph/0103357 
3251: \bibitem{} 
3252: Sokolov V.V., et al., 1997, astro-ph/9709093 
3253: \bibitem{} 
3254: Sokolov V.V., et al., 1998, A\&A 334, 117 
3255: \bibitem{} 
3256: Sokolov V.V., et al., 2000, astro-ph/0001357 
3257: \bibitem{} 
3258: Sokolov V.V., et al., 2001a,  astro-ph/0102492 
3259: \bibitem{} 
3260: Sokolov V.V., et al., 2001b, A\&A 372, 438 
3261: \bibitem{} 
3262: Sollerman J., et al., 2000, astro-ph/0006406 
3263: \bibitem{} 
3264: Soffitta P., et al., 1998, IAU Circ. 6884 
3265: \bibitem{} 
3266: Spruit H.C.,  Phinney, E.S., 1998, Nature 393, 189 
3267: \bibitem{} 
3268: Stanek K.Z., et al., 1999, ApJ 522, L39 
3269: \bibitem{} 
3270: Stanek K.Z. et al., 2001, astro-ph/0104329 
3271: \bibitem{} 
3272: Takeshima T. et al., 1999, GCN Circ. 478  
3273: \bibitem{} 
3274: Tingay S.J., et al., 1995, Nature 374, 141 
3275: \bibitem{} 
3276: Tinney C., et al., 1998, IAU Circ. 6896 
3277: \bibitem{} 
3278: Valentini G. et al., 2001, GCN Circ. 992 
3279: \bibitem{} 
3280: van den Bergh S. Tammann G.A., 1991,  ARA\&A 29, 363 
3281: \bibitem{} 
3282: van  Dyk, S.D., Hamuy, M., Filippenko, A.V., 1996, AJ 111, 2017 
3283: \bibitem{} 
3284: van Paradijs J., et al., 1998, A\&A 192, L147 
3285: \bibitem{} 
3286: Veillet C., 2000a, GCN Circ. 588, 598, 610, 611, 623 
3287: \bibitem{} 
3288: Veillet C., 2000b, GCN Circ. 831 
3289: \bibitem{} 
3290: Veillet C., 2001, GCN Circ. 998, 1000, 1003 
3291: \bibitem{} 
3292: Vrba, F., Canzian, B., 2000, GCN Circ. 819 
3293: \bibitem{} 
3294: Vreeswijk P.M., et al., 1999a, GCN Circ. 310, 324 
3295: \bibitem{} 
3296: Vreeswijk P.M., et al., 1999b, GCN Circ. 496 
3297: \bibitem{} 
3298: Vreeswijk P.M., et al., 1999c, ApJ 523, 171 
3299: \bibitem{} 
3300: Vreeswijk P.M., et al., 2000, ApJ 546, 672 
3301: \bibitem{} 
3302: Watanabe K., et al., 2001, GCN Circ. 993 
3303: \bibitem{} 
3304: Wieringa M.H., et al., 1999, A\&AS 138, 467 
3305: \bibitem{} 
3306: Wijers R.A.M.J., et al., 1997,  MNRAS 288, L5 
3307: \bibitem{} 
3308: Wilson A.S., et al., 2000, astro-ph/0008467 
3309: \bibitem{} 
3310: Woosley S.E., 1993, ApJ 405, 273 
3311: \bibitem{} 
3312: Woosley S.E., MacFadyen, A.I.,  1999, A\&AS 138, 499 
3313: \bibitem{} 
3314: Yoshida A., et al., 2001, astro-ph/0107331 
3315: \bibitem{} 
3316: Zapatero Osorio M.R., et al., 1998, IAU Circ. 6967 
3317: \bibitem{} 
3318: Zharikov S.V., et al. 1999, A\&AS 138(3), 485 
3319:  
3320:  
3321: \end{thebibliography}{} 
3322:  
3323:  
3324: \newpage 
3325:  
3326:                              
3327: \clearpage 
3328:  
3329:  
3330: %%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3331: %%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3332: \begin{figure} 
3333: \begin{center} 
3334: \vspace*{.003cm} 
3335: \hspace*{-0cm} 
3336: \epsfig{file=coneCB.ps,width=8.7cm} 
3337: %\vspace*{-14.6cm} 
3338: \caption{(a) A firecone or, more properly, a {\it firetrumpet}. 
3339: In the scenario discussed in the text (initiated by Rhoads 1997), 
3340: the cone expands 
3341: conically for a distance, after which the jet angle 
3342: $\rm\theta_j$ widens as its front travels. 
3343: (b) Cannonballs (shown here, somewhat pedantically, 
3344: a bit Lorentz-contracted) subtend decreasing angles as they travel. 
3345: The only relevant angle in the CB model is the observer's 
3346: viewing angle $\theta$.} 
3347: \vspace*{-0.5cm} 
3348: \label{figtrumpet} 
3349: \end{center} 
3350: \end{figure} 
3351:  
3352:  
3353:  
3354: \begin{figure} 
3355: \begin{center} 
3356: \vspace*{.003cm} 
3357: \hspace*{-0cm} 
3358: \epsfig{file=model3.ps,width=8.7cm} 
3359: %\vspace*{-14.6cm} 
3360: \caption{An ``artist's view'' (not to scale) of the CB model 
3361: of GRBs and their AGs. A core-collapse SN results in 
3362: a compact object and a fast-rotating torus of non-ejected 
3363: fallen-back material. Matter (not shown) catastrophically accreting 
3364: into the central object produces 
3365: a narrowly collimated beam of CBs, of which only some of 
3366: the ``northern'' ones are depicted. As these CBs pierce the SN shell, 
3367: not precisely on the same spot, 
3368: they heat and re-emit photons, that are 
3369: Lorentz-boosted and collimated by the CBs' relativistic motion.} 
3370: \vspace*{-0.5cm} 
3371: \label{model} 
3372: \end{center}  
3373: \end{figure} 
3374:  
3375: \clearpage 
3376: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3377: \begin{figure} 
3378: \begin{center} 
3379: \vspace*{.003cm} 
3380: \hspace*{-0cm} 
3381: \epsfig{file=CBrest.ps,width=8.7cm} 
3382: %\vspace*{-14.6cm} 
3383: \caption{A CB, in its rest system, sees the constituents of the ionized 
3384: ISM impinge in one direction. The CB's chaotic magnetic field 
3385: disperses these particles so that they come out isotropically. 
3386: Electrons, but not protons, lose their energy by synchrotron radiation.} 
3387: \vspace*{-0.5cm} 
3388: \label{CBrest} 
3389: \end{center}  
3390: \end{figure} 
3391:  
3392:  
3393: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3394: \begin{figure} 
3395: \begin{center} 
3396: \vspace*{.003cm} 
3397: \hspace*{-0cm} 
3398: \epsfig{file=RCB.ps,width=8.7cm} 
3399: %\vspace*{-14.6cm} 
3400: \caption{The radius of a CB as a function of observer's time, 
3401: after the CB becomes transparent to radiation, for two choices 
3402: of the initial transverse expansion velocity $\rm\beta_{trans}\,c$.} 
3403: \vspace*{-0.5cm} 
3404: \label{RCB} 
3405: \end{center}  
3406: \end{figure} 
3407:  
3408:  
3409: %%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3410: \begin{figure} 
3411: \begin{center} 
3412: \vspace*{.003cm} 
3413: \hspace*{-0cm} 
3414: \epsfig{file=fluxnovel2.ps,width=7.5cm} 
3415: %\vspace*{-14.6cm} 
3416: \caption{Afterglow flux, in arbitrary units, as a function of observer's 
3417: time, for $\gamma_0=10^3$ and various viewing angles $\theta$, 
3418: as given by Eqs.~(\ref{fluxdensity}) with $\rm n=4.1$ and 
3419: $\rm\gamma(t)$ as in Eq.~(\ref{cubic}) 
3420: with the reference value of $\rm x_\infty$ in Eq.~(\ref{gamoft}).} 
3421: \vspace*{-0.5cm} 
3422: \label{figflux} 
3423: \end{center}  
3424: \end{figure} 
3425: %%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3426:  
3427: %%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3428: \begin{figure} 
3429: \begin{center} 
3430: \vspace*{.003cm} 
3431: \hspace*{-0cm} 
3432: \epsfig{file=fluxofnpnovel.ps,width=7.5cm} 
3433: %\vspace*{-14.6cm} 
3434: \caption{ 
3435: Afterglow flux, in arbitrary units, as a function of observer's 
3436: time, for $\gamma_0=1/\theta=10^3$ and various values 
3437: of the (constant) ISM density, $\rm n^{SN}$, close to the GRB progenitor, 
3438: as given by Eqs.~(\ref{fluxdensity}), with $\rm n=4.1$ and 
3439: $\rm\gamma(t)$ as in Eq.~(\ref{cubic}).} 
3440: \vspace*{-0.5cm} 
3441: \label{figfluxother} 
3442: \end{center}  
3443: \end{figure} 
3444: %%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3445:  
3446:  
3447: \clearpage 
3448:  
3449: \begin{figure}[t] 
3450: \begin{tabular}{cc} 
3451: \hskip 2truecm 
3452: \vspace*{2cm} 
3453: \hspace*{-1.7cm} 
3454: \epsfig{file=ag1n0228.eps, width=8cm} \\ 
3455: %\hskip 1truecm 
3456: \hspace*{.5cm} 
3457: \epsfig{file=ag2n0228.eps, width=8cm} 
3458: \end{tabular} 
3459: \caption{Comparisons between our fitted R-band afterglow 
3460: (upper curves) and the observations listed in Table II, 
3461: not corrected for extinction, 
3462: for GRB 970228, at $\rm z=0.695$. 
3463: Upper panel: Without subtraction of the host 
3464: galaxy's contribution (the straight line). 
3465: Lower panel: With the host galaxy subtracted, and the 
3466: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3467: (\ref{cubic}). The contribution 
3468: from a 1998bw-like supernova placed at the GRB's 
3469: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3470: is indicated in both panels by a line of crosses. 
3471: The SN bump is clearly discernible.} 
3472: \label{fig228} 
3473: \end{figure} 
3474:  
3475: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3476:  
3477: \begin{figure}[t] 
3478: \begin{tabular}{cc} 
3479: \hskip 2truecm 
3480: \vspace*{2cm} 
3481: \hspace*{-1.7cm} 
3482: \epsfig{file=ag1n1214.eps, width=8cm} \\ 
3483: %\hskip 1truecm 
3484: \hspace*{.5cm} 
3485: \epsfig{file=ag2n1214.eps, width=8cm} 
3486: \end{tabular} 
3487: \caption{Comparisons between our fitted R-band afterglow 
3488: (upper curves) and the observations listed in Table II, 
3489: not corrected for  extinction, 
3490: for GRB 971214, at $\rm z=3.418$. 
3491: Upper panel: Without subtraction of the host 
3492: galaxy's contribution (the straight line). 
3493: Lower panel: With the host galaxy subtracted, and the 
3494: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3495: (\ref{cubic}). The contribution 
3496: from a 1998bw-like supernova placed at the GRB's 
3497: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3498: is indicated by a line of crosses. The expected supernova 
3499: contribution is too weak to be observed.} 
3500: \label{fig214} 
3501: \end{figure} 
3502: 	 
3503: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3504:  
3505: \begin{figure}[t] 
3506: \begin{tabular}{cc} 
3507: \hskip 2truecm 
3508: \vspace*{2cm} 
3509: \hspace*{-1.7cm} 
3510: \epsfig{file=ag1n0613.eps, width=8cm} \\ 
3511: %\hskip 1truecm 
3512: \hspace*{.5cm} 
3513: \epsfig{file=ag2n0613.eps, width=8cm} 
3514: \end{tabular} 
3515: \caption{Comparisons between our fitted R-band afterglow 
3516: (upper curves) and the observations listed in Table II, 
3517: not corrected for  extinction, 
3518: for GRB 980613, at $\rm z=1.096$. 
3519: Upper panel: Without subtraction of the host 
3520: galaxy's contribution (the straight line). 
3521: Lower panel: With the host galaxy subtracted, and the 
3522: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3523: (\ref{cubic}). The contribution 
3524: from a 1998bw-like supernova placed at the GRB's 
3525: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3526: is indicated in both panels by a line of crosses. 
3527: A SN 1998bw-like contribution, though based on 
3528: only one significant point, appears to be required.} 
3529: \label{fig613} 
3530: \end{figure} 
3531:  
3532:  
3533: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3534:  
3535: \begin{figure}[t] 
3536: \begin{tabular}{cc} 
3537: \hskip 2truecm 
3538: \vspace*{2cm} 
3539: \hspace*{-1.7cm} 
3540: \epsfig{file=ag1n0703.eps, width=8cm} \\ 
3541: %\hskip 1truecm 
3542: \hspace*{.5cm} 
3543: \epsfig{file=ag2n0703.eps, width=8cm} 
3544: \end{tabular} 
3545: \caption{ Comparisons between our fitted R-band afterglow 
3546: (upper curves) and the observations listed in Table II, 
3547: not corrected for  extinction, 
3548: for GRB 980703, at $\rm z=0.966$. 
3549: Upper panel: Without subtraction of the host 
3550: galaxy's contribution (the straight line). 
3551: Lower panel: With the host galaxy subtracted, and the 
3552: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3553: (\ref{cubic}). The contribution 
3554: from a 1998bw-like supernova placed at the GRB's 
3555: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3556: is indicated  by a line of crosses. 
3557: A SN 1998bw-like contribution, though the errors 
3558: are large, appears to be required.} 
3559: \label{fig703} 
3560: \end{figure} 
3561:  
3562:  
3563: \clearpage 
3564: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3565:  
3566: \begin{figure}[t] 
3567: \begin{tabular}{cc} 
3568: \hskip 2truecm 
3569: \vspace*{2cm} 
3570: \hspace*{-1.7cm} 
3571: \epsfig{file=ag1n0123.eps, width=8cm} \\ 
3572: %\hskip 1truecm 
3573: \hspace*{.5cm} 
3574: \epsfig{file=ag2n0123.eps, width=8cm} 
3575: \end{tabular} 
3576: \caption{Comparisons between our fitted R-band afterglow 
3577: (upper curves) and the observations listed in Table II, 
3578: not corrected for  extinction, 
3579: for GRB 990123, at $\rm z=1.6$. 
3580: Upper panel: Without subtraction of the host 
3581: galaxy's contribution (the straight line). 
3582: Lower panel: With the host galaxy subtracted, and the 
3583: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3584: (\ref{cubic}). The contribution 
3585: from a 1998bw-like supernova placed at the GRB's 
3586: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3587: is indicated by a line of crosses. The fit is not very good. 
3588: There are no observations at the time when a SN1998bw-like 
3589: contribution may have been seen.} 
3590: \label{fig123} 
3591: \end{figure} 
3592:  
3593: % 
3594:  
3595: %\clearpage 
3596: %\goodbreak 
3597: %\filbreak 
3598: %\vfill\eject 
3599:  
3600: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3601:  
3602: \begin{figure}[t] 
3603: \begin{tabular}{cc} 
3604: \hskip 2truecm 
3605: \vspace*{2cm} 
3606: \hspace*{-1.7cm} 
3607: \epsfig{file=ag1n0510.eps, width=8cm} \\ 
3608: %\hskip 1truecm 
3609: \hspace*{.5cm} 
3610: \epsfig{file=ag2n0510.eps, width=8cm} 
3611: \end{tabular} 
3612: \caption{Comparisons between our fitted R-band afterglow 
3613: (upper curves) and the observations listed in Table II, 
3614: not corrected for  extinction, 
3615: for GRB 990510, at $\rm z=1.619$. 
3616: Upper panel: Without subtraction of the host 
3617: galaxy's contribution (the straight line). 
3618: Lower panel: With the host galaxy subtracted, and the 
3619: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3620: (\ref{cubic}). The contribution 
3621: from a 1998bw-like supernova placed at the GRB's 
3622: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3623: is indicated in both panels by a line of crosses. 
3624: The expected supernova contribution is too weak to be observed.} 
3625: \label{fig510} 
3626: \end{figure} 
3627: \clearpage 
3628: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3629:  
3630: \begin{figure}[t] 
3631: \begin{tabular}{cc} 
3632: \hskip 2truecm 
3633: \vspace*{2cm} 
3634: \hspace*{-1.7cm} 
3635: \epsfig{file=ag1nv0510.eps, width=8cm} \\ 
3636: %\hskip 1truecm 
3637: \hspace*{.5cm} 
3638: \epsfig{file=ag2nv0510.eps, width=8cm} 
3639: \end{tabular} 
3640: \caption{Comparisons between our fitted V-band afterglow 
3641: (upper curves) and the observations listed in Table II, 
3642: not corrected for  extinction, 
3643: for GRB 990510, at $\rm z=1.619$. 
3644: Upper panel: Without subtraction of the host 
3645: galaxy's contribution (the straight line). 
3646: Lower panel: With the host galaxy subtracted, and the 
3647: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3648: (\ref{cubic}). The contribution 
3649: from a 1998bw-like supernova placed at the GRB's 
3650: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3651: is indicated in both panels by a line of crosses. 
3652: The expected supernova contribution is too weak to be observed.} 
3653: \label{fig510V} 
3654: \end{figure} 
3655:  
3656: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3657:  
3658: %\clearpage 
3659:  
3660: \begin{figure}[t] 
3661: \begin{tabular}{cc} 
3662: \hskip 2truecm 
3663: \vspace*{2cm} 
3664: \hspace*{-1.7cm} 
3665: \epsfig{file=ag1n0712.eps, width=8cm} \\ 
3666: %\hskip 1truecm 
3667: \hspace*{.5cm} 
3668: \epsfig{file=ag2n0712.eps, width=8cm} 
3669: \end{tabular} 
3670: \caption{Comparisons between our fitted R-band afterglow 
3671: (upper curves) and the observations listed in Table II, 
3672: not corrected for  extinction, 
3673: for GRB 990712, at $\rm z=0.434$. 
3674: Upper panel: Without subtraction of the host 
3675: galaxy's contribution (the straight line). 
3676: Lower panel: With the host galaxy subtracted, and the 
3677: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3678: (\ref{cubic}). The contribution 
3679: from a 1998bw-like supernova placed at the GRB's 
3680: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3681: is indicated in both panels by a line of crosses. 
3682: The SN contribution is clearly discernible, but a bump at slightly 
3683: earlier times than that of our standard-candle 
3684: SN1998bw would provide a better description.} 
3685: \label{fig712} 
3686: \end{figure} 
3687:  
3688: \clearpage 
3689: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3690:  
3691: \begin{figure}[t] 
3692: \begin{tabular}{cc} 
3693: \hskip 2truecm 
3694: \vspace*{2cm} 
3695: \hspace*{-1.7cm} 
3696: \epsfig{file=ag1nv0712.eps, width=8cm} \\ 
3697: %\hskip 1truecm 
3698: \hspace*{.5cm} 
3699: \epsfig{file=ag2nv0712.eps, width=8cm} 
3700: \end{tabular} 
3701: \caption{Comparisons between our fitted V-band afterglow 
3702: (upper curves) and the observations listed in Table II, 
3703: not corrected for  extinction, 
3704: for GRB 990712, at $\rm z=0.434$. 
3705: Upper panel: Without subtraction of the host 
3706: galaxy's contribution (the straight line). 
3707: Lower panel: With the host galaxy subtracted, and the 
3708: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3709: (\ref{cubic}). The contribution 
3710: from a 1998bw-like supernova placed at the GRB's 
3711: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3712: is indicated in both panels by a line of crosses. 
3713: The SN  is clearly discernible, but a bump at slightly 
3714: earlier times than that of our standard-candle 
3715: SN1998bw would provide a better description.} 
3716: \label{fig712V} 
3717: \end{figure} 
3718: %\clearpage 
3719:  
3720: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3721:  
3722: \begin{figure}[t] 
3723: \begin{tabular}{cc} 
3724: \hskip 2truecm 
3725: \vspace*{2cm} 
3726: \hspace*{-1.7cm} 
3727: \epsfig{file=ag1n1208.eps, width=8cm} \\ 
3728: %\hskip 1truecm 
3729: \hspace*{.5cm} 
3730: \epsfig{file=ag2n1208.eps, width=8cm} 
3731: \end{tabular} 
3732: \caption{Comparisons between our fitted R-band afterglow 
3733: (upper curves) and the observations listed in Table II, 
3734: not corrected for  extinction, 
3735: for GRB 991208, at $\rm z=0.706$. 
3736: Upper panel: Without subtraction of the host 
3737: galaxy's contribution (the straight line). 
3738: Lower panel: With the host galaxy subtracted, and the 
3739: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3740: (\ref{cubic}). The contribution 
3741: from a 1998bw-like supernova placed at the GRB's 
3742: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3743: is indicated in both panels by a line of crosses. 
3744: The SN contribution is clearly discernible.} 
3745: \label{fig208} 
3746: \end{figure} 
3747: \clearpage 
3748: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3749:  
3750: \begin{figure}[t] 
3751: \begin{tabular}{cc} 
3752: \hskip 2truecm 
3753: \vspace*{2cm} 
3754: \hspace*{-1.7cm} 
3755: \epsfig{file=ag1n1216.eps, width=8cm} \\ 
3756: %\hskip 1truecm 
3757: \hspace*{.5cm} 
3758: \epsfig{file=ag2n1216.eps, width=8cm} 
3759: \end{tabular} 
3760: \caption{Comparisons between our fitted R-band afterglow 
3761: (upper curves) and the observations listed in Table II, 
3762: not corrected for  extinction, 
3763: for GRB 991216, at $\rm z=1.020$. 
3764: Upper panel: Without subtraction of the host 
3765: galaxy's contribution (the straight line). 
3766: Lower panel: With the host galaxy subtracted, and the 
3767: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3768: (\ref{cubic}). The contribution 
3769: from a 1998bw-like supernova placed at the GRB's 
3770: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3771: is indicated in both panels by a line of crosses. 
3772: The data show possible evidence for a SN bump.} 
3773: \label{fig216} 
3774: \end{figure} 
3775:  
3776:  
3777: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3778:  
3779: \begin{figure}[t] 
3780: \begin{tabular}{cc} 
3781: \hskip 2truecm 
3782: \vspace*{2cm} 
3783: \hspace*{-1.7cm} 
3784: \epsfig{file=ag1n0131.eps, width=8cm} \\ 
3785: %\hskip 1truecm 
3786: \hspace*{.5cm} 
3787: \epsfig{file=ag2n0131.eps, width=8cm} 
3788: \end{tabular} 
3789: \caption{Comparisons between our fitted R-band afterglow 
3790: (upper curves) and the observations listed in Table II, 
3791: not corrected for  extinction, 
3792: for GRB 000131, at $\rm z=4.5$. 
3793: Upper panel: Without subtraction of the host 
3794: galaxy's contribution (the straight line). 
3795: Lower panel: With the host galaxy subtracted, and the 
3796: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3797: (\ref{cubic}). The contribution 
3798: from a 1998bw-like supernova placed at the GRB's 
3799: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3800: is indicated by a line of crosses. 
3801: The data are too scarce for conclusions to be drawn.} 
3802: \end{figure} 
3803: \clearpage 
3804: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3805:  
3806: \begin{figure}[t] 
3807: \begin{tabular}{cc} 
3808: \hskip 2truecm 
3809: \vspace*{2cm} 
3810: \hspace*{-1.7cm} 
3811: \epsfig{file=ag1n0301.eps, width=8cm} \\ 
3812: %\hskip 1truecm 
3813: \hspace*{.5cm} 
3814: \epsfig{file=ag2n0301.eps, width=8cm} 
3815: \end{tabular} 
3816: \caption{Comparisons between our fitted R-band afterglow 
3817: (upper curves) and the observations listed in Table II, 
3818: not corrected for  extinction, 
3819: for GRB 000301c, at $\rm z=2.040$. 
3820: Upper panel: Without subtraction of the host 
3821: galaxy's contribution (the straight line). 
3822: Lower panel: With the host galaxy subtracted, and the 
3823: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3824: (\ref{cubic}). The contribution 
3825: from a 1998bw-like supernova placed at the GRB's 
3826: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3827: is indicated by a line of crosses. 
3828: The expected SN contribution is too weak to be 
3829: observable.} 
3830: \label{fig301} 
3831: \end{figure} 
3832:  
3833: %\clearpage 
3834:  
3835: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3836:  
3837: \begin{figure}[t] 
3838: \begin{tabular}{cc} 
3839: \hskip 2truecm 
3840: \vspace*{2cm} 
3841: \hspace*{-1.7cm} 
3842: \epsfig{file=ag1n0418.eps, width=8cm} \\ 
3843: %\hskip 1truecm 
3844: \hspace*{.5cm} 
3845: \epsfig{file=ag2n0418.eps, width=8cm} 
3846: \end{tabular} 
3847: \caption{Comparisons between our fitted R-band afterglow 
3848: (upper curves) and the observations listed in Table II, 
3849: not corrected for  extinction, 
3850: for GRB 000418, at $\rm z=1.119$. 
3851: Upper panel: Without subtraction of the host 
3852: galaxy's contribution (the straight line). 
3853: Lower panel: With the host galaxy subtracted, and the 
3854: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3855: (\ref{cubic}). The contribution 
3856: from a 1998bw-like supernova placed at the GRB's 
3857: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3858: is indicated in both panels by a line of crosses. 
3859: There is some indication of a SN1998bw-like 
3860: contribution.} 
3861: \label{fig418} 
3862: \end{figure} 
3863: \clearpage 
3864: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3865: \begin{figure}[t] 
3866: \begin{tabular}{cc} 
3867: \hskip 2truecm 
3868: \vspace*{2cm} 
3869: \hspace*{-1.7cm} 
3870: \epsfig{file=ag1n0926.eps, width=8cm} \\ 
3871: %\hskip 1truecm 
3872: \hspace*{.5cm} 
3873: \epsfig{file=ag2n0926.eps, width=8cm} 
3874: \end{tabular} 
3875: \caption{Comparisons between our fitted R-band afterglow 
3876: (upper curves) and the observations listed in Table II, 
3877: not corrected for  extinction, 
3878: for GRB 000926, at $\rm z=2.066$. 
3879: Upper panel: Without subtraction of the host 
3880: galaxy's contribution (the straight line). 
3881: Lower panel: With the host galaxy subtracted, and the 
3882: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3883: (\ref{cubic}). The contribution 
3884: from a 1998bw-like supernova placed at the GRB's 
3885: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3886: is indicated in both panels by a line of crosses. 
3887: A SN1998bw-like contribution could not have been seen.} 
3888: \label{fig926} 
3889: \end{figure} 
3890: %\clearpage 
3891: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3892:  
3893:  
3894: \begin{figure}[t] 
3895: \begin{tabular}{cc} 
3896: \hskip 2truecm 
3897: \vspace*{2cm} 
3898: \hspace*{-1.7cm} 
3899: \epsfig{file=ag1nt0222.eps, width=8cm} \\ 
3900: %\hskip 1truecm 
3901: \hspace*{.5cm} 
3902: \epsfig{file=ag2nt0222.eps, width=8cm} 
3903: \end{tabular} 
3904: \caption{Comparisons between our fitted R-band afterglow 
3905: (upper curves) and the observations listed in Table II, 
3906: not corrected for  extinction, 
3907: for GRB 010222, at $\rm z>1.474$. 
3908: Upper panel: Without subtraction of the host 
3909: galaxy's contribution (the straight line). 
3910: Lower panel: With the host galaxy subtracted, and the 
3911: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3912: (\ref{cubic}). The contribution 
3913: from a 1998bw-like supernova placed at the GRB's 
3914: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3915: is indicated in both panels by a line of crosses. 
3916: There are no observations at the time a 1998bw-like 
3917: supernova would have significantly contributed.} 
3918: \label{fig222} 
3919: \end{figure} 
3920:  
3921: \clearpage 
3922: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3923:  
3924: \begin{figure}[t] 
3925: \begin{tabular}{cc} 
3926: \hskip 2truecm 
3927: \vspace*{2cm} 
3928: \hspace*{-1.7cm} 
3929: \epsfig{file=ag1n1cb0508.eps, width=8cm} \\ 
3930: %\hskip 1truecm 
3931: \hspace*{.5cm} 
3932: \epsfig{file=ag2n1cb0508.eps, width=8cm} 
3933: \end{tabular} 
3934: \caption{Comparisons between our fitted R-band afterglow 
3935: (upper curves) and the observations listed in Table II, 
3936: not corrected for  extinction, 
3937: for GRB 970508, at $\rm z=0.835$. 
3938: Upper panel: Without subtraction of the host 
3939: galaxy's contribution (the straight line). 
3940: Lower panel: With the host galaxy subtracted, and the 
3941: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3942: (\ref{cubic}). The contribution 
3943: from a 1998bw-like supernova placed at the GRB's 
3944: redshift, Eq.~(\ref{bw}), corrected for  extinction, 
3945: is indicated in both panels by a line of crosses. 
3946: This is the best fit not including the possibility of lensing; 
3947: and it is terrible. } 
3948: \label{1CB508} 
3949: \end{figure} 
3950: %\clearpage 
3951: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3952:  
3953: \begin{figure}[t] 
3954: \begin{tabular}{cc} 
3955: \hskip 2truecm 
3956: \vspace*{2cm} 
3957: \hspace*{-1.7cm} 
3958: \epsfig{file=ag1varn0508.eps, width=8cm} \\ 
3959: %\hskip 1truecm 
3960: \hspace*{.5cm} 
3961: \epsfig{file=ag2varn0508.eps, width=8cm} 
3962: \end{tabular} 
3963: \caption{The same as in Fig.~(\ref{1CB508}), but with a density profile
3964: that jumps from one constant value to another, at a distance $\sim 0.24$ kpc
3965: from the progenitor.
3966: Upper panel: Without subtraction of the host 
3967: galaxy's contribution (the straight line). 
3968: Lower panel: With the host galaxy subtracted, and the 
3969: CB's AG (the line of squares) given by Eqs.~(\ref{fluxdensity}) and 
3970: (\ref{cubic}).
3971: This time the fit, unlike that of Fig.~(\ref{1CB508}), is quite satisfactory. 
3972: A SN1998bw-like contribution is necessary.} 
3973: \label{jump508} 
3974: \end{figure} 
3975: \clearpage 
3976: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3977:  
3978:  
3979: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
3980:  
3981: \begin{figure}[t] 
3982: \begin{tabular}{cc} 
3983: \hskip 2truecm 
3984: \vspace*{2cm} 
3985: \hspace*{-1.7cm} 
3986: \epsfig{file=ag1nt0123.eps, width=8cm} \\ 
3987: %\hskip 1truecm 
3988: \hspace*{.5cm} 
3989: \epsfig{file=ag2nt0123.eps, width=8cm} 
3990: \end{tabular} 
3991: \caption{Comparisons between our fitted R-band afterglow 
3992: (upper curves) and the observations, 
3993: not corrected for extinction, for GRB 990123. 
3994: Upper panel: Without subtraction of the host 
3995: galaxy's contribution (the straight line). 
3996: Lower panel: With the host galaxy subtracted. 
3997: Unlike in the previous ensemble of figures, these AG data 
3998: start from a short time after the GRB. The starting $\rm t^{-2}$ 
3999: behaviour is that expected if the CBs are moving trough a 
4000: density profile, $\rm n\propto r^{-2}$, 
4001: induced by the parent-star's pre-SN wind and 
4002: ejections.} 
4003: \label{early123} 
4004: \end{figure} 
4005:  
4006: %\clearpage 
4007: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4008:  
4009: \begin{figure} 
4010: \begin{center} 
4011: \vspace*{.003cm} 
4012: \hspace*{-0cm} 
4013: \epsfig{file=ag0nx0228.eps,width=8.5cm} 
4014: %\vspace*{-14.6cm} 
4015: \caption{The early-time and late-time X-ray AG of GRB 970228 in the 2--10 
4016: keV range 
4017: as measured by Costa et al. (1997), fitted with 
4018: Eq.~(\ref{Xdensity}) for a constant density along the 
4019: CB trajectory. This assumption 
4020: should be vastly inappropriate between 
4021: $ \rm \sim 2\times 10^{-3}$ and $\sim 0.2$ days, 
4022: during which $\rm n_e$ and the X-ray AG are 
4023: expected to diminish by two or three orders of magnitude. 
4024: Coincidentally, there are no data there.} 
4025: \vspace*{-0.5cm} 
4026: \label{X228} 
4027: \end{center} 
4028: \end{figure} 
4029:  
4030: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4031:  
4032: \begin{figure} 
4033: \begin{center} 
4034: \vspace*{.003cm} 
4035: \hspace*{-0cm} 
4036: \epsfig{file=ag0nx0828.eps,width=8.5cm} 
4037: %\vspace*{-14.6cm} 
4038: \caption{The early-time and late-time X-ray AG of GRB 970828 in the 2--10 
4039: keV range as measured by Smith et al. (2001) and Yoshida et al. (2001), 
4040: fitted with Eq.~(\ref{Xdensity}) for a constant density along the 
4041: CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4042: \vspace*{-0.5cm} 
4043: \label{X828} 
4044: \end{center} 
4045: \end{figure} 
4046: \clearpage 
4047: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4048:  
4049: \begin{figure} 
4050: \begin{center} 
4051: \vspace*{.003cm} 
4052: \hspace*{-0cm} 
4053: \epsfig{file=ag0nx1214.eps,width=8.5cm} 
4054: %\vspace*{-14.6cm} 
4055: \caption{The early-time and late-time X-ray AG of GRB 971214 in the 2--10 
4056: keV range as measured by Dal Fiume et al. (2000)  
4057: fitted with Eq.~(\ref{Xdensity}) for a constant density along the 
4058: CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4059: \vspace*{-0.5cm} 
4060: \label{X1214} 
4061: \end{center} 
4062: \end{figure} 
4063:  
4064: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4065:  
4066: \begin{figure} 
4067: \begin{center} 
4068: \vspace*{.003cm} 
4069: \hspace*{-0cm} 
4070: \epsfig{file=ag0nx0613.eps,width=8.5cm} 
4071: %\vspace*{-14.6cm} 
4072: \caption{The early-time and late-time X-ray AG of GRB 980613 in the 2--10 
4073: keV range (Piro 2000) 
4074: fitted with Eq.~(\ref{Xdensity}) for a constant density along the 
4075:     CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4076: \vspace*{-0.5cm} 
4077: \label{X613} 
4078: \end{center} 
4079: \end{figure} 
4080:  
4081:  
4082: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4083:  
4084: \begin{figure} 
4085: \begin{center} 
4086: \vspace*{.003cm} 
4087: \hspace*{-0cm} 
4088: \epsfig{file=ag0nx0123.eps,width=8.5cm} 
4089: %\vspace*{-14.6cm} 
4090: \caption{The early-time and late-time X-ray AG of GRB 990123 in the 2--10 
4091: keV range (Piro 2000)  
4092: fitted with Eq.~(\ref{Xdensity}) for a constant density along the 
4093:     CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4094: \vspace*{-0.5cm} 
4095: \label{X123} 
4096: \end{center} 
4097: \end{figure} 
4098:  
4099: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4100: \begin{figure} 
4101: \begin{center} 
4102: \vspace*{.003cm} 
4103: \hspace*{-0cm} 
4104: \epsfig{file=ag0nx0510.eps,width=8.5cm} 
4105: %\vspace*{-14.6cm} 
4106: \caption{The early-time and late-time X-ray AG of GRB 990510 in the 2--10 
4107: keV range as measured by Pian et al. (2001) 
4108: fitted with Eq.~(\ref{Xdensity}) for a constant density along the 
4109: CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4110: \vspace*{-0.5cm} 
4111: \label{X510} 
4112: \end{center} 
4113: \end{figure} 
4114:  
4115: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4116:  
4117: \begin{figure} 
4118: \begin{center} 
4119: \vspace*{.003cm} 
4120: \hspace*{-0cm} 
4121: \epsfig{file=ag0nx0926.eps,width=8.5cm} 
4122: %\vspace*{-14.6cm} 
4123: \caption{The early-time and late-time X-ray AG of GRB 000926 in the 2--10 
4124: keV range as measured by Piro et al. (2001), fitted with 
4125: Eq.~(\ref{Xdensity}) for a constant density along the 
4126: CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4127: \vspace*{-0.5cm} 
4128: \label{X926} 
4129: \end{center} 
4130: \end{figure} 
4131:  
4132:  
4133:  
4134:  
4135: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4136:  
4137:  
4138:  
4139: \begin{figure} 
4140: \begin{center} 
4141: \vspace*{.003cm} 
4142: \hspace*{-0cm} 
4143: \epsfig{file=ag0nx0222.eps,width=8.5cm} 
4144: %\vspace*{-14.6cm} 
4145: \caption{The early-time and late-time X-ray AG of GRB 010222 in the 2--10 
4146: keV range as measured by In `t Zand et al. (2001), fitted with 
4147: Eq.~(\ref{Xdensity}) for a constant density along the 
4148: CB trajectory. The same comments as in Fig~(\ref{X228}) apply here.} 
4149: \vspace*{-0.5cm} 
4150: \label{X222} 
4151: \end{center} 
4152: \end{figure} 
4153:  
4154: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4155:  
4156: \begin{figure} 
4157: \begin{center} 
4158: \vspace*{.2cm} 
4159: \hspace*{-0cm} 
4160: \epsfig{file=ag0varnx0508.eps,width=7.5cm} 
4161: %\vspace*{-14.6cm} 
4162: \caption{The early-time and late-time X-ray AG of GRB 970508 in the 2--10 
4163: keV range as measured by Piro et al. (1998), fitted with 
4164: Eq.~(\ref{Xdensity}) for a $\rm \sim 1/r^2$ plus constant density along the 
4165: CB trajectory. The contribution of the X-ray line observed before 0.8 d was 
4166: subtracted from the data. This reduces the two observed points between 
4167: 0.2 and 1 day by a factor $\sim 0.39$. 
4168:  The overall result is compatible with an effect that, at late times, 
4169: is achromatic, since the late optical and X-ray AGs are both
4170: proportional to $\rm n_e$; see Fig.~(\ref{jump508}) 
4171: for the optical counterpart.} 
4172: \vspace*{-0.5cm} 
4173: \label{X508} 
4174: \end{center} 
4175: \end{figure} 
4176:  
4177: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4178: \begin{figure} 
4179: \begin{center} 
4180: \vspace*{.003cm} 
4181: \hspace*{-0cm} 
4182: \epsfig{file=ag0nx1216.eps,width=8.5cm} 
4183: %\vspace*{-14.6cm} 
4184: \caption{The early-time and late-time X-ray AG of GRB 991216 in the 2--10 
4185: keV range as measured by Corbet and Smith (1999) and  Takeshima et al. 
4186: (1999) fitted with Eq.~(\ref{Xdensity}) for a density declining like 
4187: $\rm r^{-2}$ plus a constant  density along the CB trajectory.} 
4188: \vspace*{-0.5cm} 
4189: \label{X1216} 
4190: \end{center} 
4191: \end{figure} 
4192:  
4193:  
4194: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4195:  
4196: \begin{figure}[t] 
4197: \begin{tabular}{cc} 
4198: \hskip 2truecm 
4199: \vspace*{2cm} 
4200: \hspace*{-2.1cm} 
4201: \epsfig{file=ag0nx0425.eps, width=8cm} \\ 
4202: %\hskip 1truecm 
4203: \hspace*{-0.4cm} 
4204: \epsfig{file=425.ps,width=9cm} 
4205: \end{tabular} 
4206: \caption{Upper panel: A fit to the X-ray afterglow of the SN1998bw/GRB 980425 
4207: pair. We call ``plateau'' the slowly-declining late measurements. 
4208: Lower panel: The V-band light curve of 
4209: the same pair, with   
4210: the blue ``SN'' curve a fit to the SN by Sollerman et al.~(2000), 
4211: dominated after day $\sim\!40$ by $^{56}$Co decay. 
4212:  The red ``AG'' curve is our prediction for the CB-induced 
4213: AG component, as given by Eq.~(\ref{fluxdensity}), 
4214: with the parameters determined from the X-ray AG fit in the 
4215: upper panel. The SN contribution 
4216: dominates up to day $\sim\! 600$. The last point is an 
4217: HST measurement at day 778, that precisely agrees with the (dashed) 
4218: SN plus CB prediction for the total AG. For an earlier version of these 
4219: results, see DD2000a.} 
4220: \vspace*{-0.5cm} 
4221: \label{425} 
4222: \end{figure} 
4223:  
4224: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4225:  
4226: \begin{figure} 
4227: \begin{center} 
4228: \vspace*{.003cm} 
4229: \hspace*{-0cm} 
4230: \epsfig{file=alpha.eps,width=8.7cm} 
4231: %\vspace*{-14.6cm} 
4232: \caption{Distribution of $\alpha$ values, from the fits to the optical AGs. 
4233: The prediction is $\alpha \approx 1.1$. 
4234: Binning from $\alpha=1.06$ to 1.26 would have made this 
4235: distribution look even more impressively narrow. 
4236: The GRBs are labelled by the last three digits of their date.} 
4237: \vspace*{-0.5cm} 
4238: \label{alphadist} 
4239: \end{center}  
4240: \end{figure} 
4241:  
4242: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4243:  
4244: \begin{figure} 
4245: \begin{center} 
4246: \vspace*{.003cm} 
4247: \hspace*{-0cm} 
4248: \epsfig{file=gamma2.eps,width=8.7cm} 
4249: %\vspace*{-14.6cm} 
4250: \caption{Distribution of $\gamma_0$ values from the fits to the optical AGs. 
4251: The expectation is $\gamma_0\sim 10^3$. 
4252: The GRBs are labelled by their last three digits.} 
4253: \vspace*{-0.5cm} 
4254: \label{gammadist} 
4255: \end{center}  
4256: \end{figure} 
4257:  
4258:  
4259: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4260: \begin{figure} 
4261: \begin{center} 
4262: \vspace*{.003cm} 
4263: \hspace*{-0cm} 
4264: \epsfig{file=theta2.eps,width=8.7cm} 
4265: %\vspace*{-14.6cm} 
4266: \caption{Distribution of $\theta$ values, in milliradians, from the fits to the 
4267: optical AGs. 
4268: The GRBs are labelled by their last three digits.} 
4269: \vspace*{-0.5cm} 
4270: \label{thetadist} 
4271: \end{center}  
4272: \end{figure} 
4273:  
4274: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4275: \begin{figure} 
4276: \begin{center} 
4277: \vspace*{.003cm} 
4278: \hspace*{-0cm} 
4279: \epsfig{file=xinf2.eps,width=8.7cm} 
4280: %\vspace*{-14.6cm} 
4281: \caption{Distribution of $\rm 
4282: Log_{10}[x_\infty(Mpc)]$ values, from the fits to the 
4283: optical AGs. The expectation 
4284: from Eq.~(\ref{gamoft}) is 0.114. 
4285: The GRBs are labelled by their last three digits.} 
4286: \vspace*{-0.5cm} 
4287: \label{xdist} 
4288: \end{center}  
4289: \end{figure} 
4290:  
4291: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
4292: \begin{figure} 
4293: \begin{center} 
4294: \vspace*{.003cm} 
4295: \hspace*{-0cm} 
4296: \epsfig{file=ECB2.eps,width=8.7cm} 
4297: %\vspace*{-14.6cm} 
4298: \caption{Distribution on $\rm E_\gamma^{CB}$ values, 
4299: as given by Eq.~(\ref{ECBrest}), in units of $10^{44}$ erg. 
4300: The GRBs are labelled by their last three digits.} 
4301: \vspace*{-0.5cm} 
4302: \label{Edist} 
4303: \end{center}  
4304: \end{figure} 
4305:  
4306: \end{document} 
4307:  
4308:  
4309:  
4310:  
4311:  
4312:  
4313:  
4314:  
4315:  
4316:  
4317:  
4318:  
4319:  
4320:  
4321:  
4322:  
4323:  
4324:  
4325:  
4326:  
4327:  
4328:  
4329:  
4330:  
4331:  
4332:  
4333:  
4334:  
4335:  
4336:  
4337:  
4338:  
4339:  
4340:  
4341:  
4342:  
4343:  
4344:  
4345:  
4346:  
4347:  
4348:  
4349:  
4350:  
4351:  
4352:  
4353:  
4354:  
4355:  
4356:  
4357:  
4358:  
4359:  
4360:  
4361:  
4362:  
4363:  
4364:  
4365:  
4366:  
4367:  
4368:  
4369:  
4370:  
4371:  
4372:  
4373:  
4374:  
4375:  
4376:  
4377:  
4378:  
4379:  
4380:  
4381:  
4382:  
4383:  
4384:  
4385:  
4386:  
4387:  
4388:  
4389: