1:
2: % aa.dem
3: % AA vers. 4.01, LaTeX class for Astronomy & Astrophysics
4: % demonstration file
5: % (c) Springer-Verlag HD
6: %---------------------------------------------------
7: %
8: %\documentclass[referee]{aa} % for a referee version
9: %
10: \documentclass[epsfig]{aa}
11: \usepackage{epsfig,deluxe}
12: % \textsc{}
13: \begin{document}
14:
15: \newcommand{\gsim}{\hbox{\rlap{$^>$}$_\sim$}}
16: \thesaurus{06; 19.63.1}
17: % A&A Section 6: Form. struct. and evolut. of stars}
18: % \thesaurus{06 % A&A Section 6: Form. struct. and evolut. of stars
19: % (03.11.1; % Cosmogony,
20: % 16.06.1; % Planets and satellites: general,
21: % 19.37.1; % Stars: formation of,
22: % 19.53.1; % Stars: oscillations of,
23: % 19.63.1)} % Stars: structure of.
24: %
25: \authorrunning{A. Dar \& A. De R\'ujula}
26: \titlerunning{A CB model of GRBs: properties of the $\gamma$-rays}
27: \title{A Cannonball model of $\gamma$-ray bursts: spectral and temporal
28: properties of the $\gamma$-rays}
29:
30: \author{Arnon Dar$^{1,2}$ and A. De R\'ujula$^1$}
31: \institute{1. Theory Division, CERN, CH-1211 Geneva 23, Switzerland\\
32: 2. Physics Department and Space Research Institute, Technion,
33: Haifa 32000, Israel }
34: \maketitle
35:
36: \maketitle
37:
38: \begin{abstract}
39:
40: Recent observations suggest that gamma ray bursts (GRBs)
41: and their afterglows are produced by highly relativistic jets emitted in
42: supernova explosions. We have proposed that the result of the event is not
43: just a compact object plus the ejecta: within days, a fraction of the
44: parent star falls back to produce a thick accretion disk. The subsequent
45: accretion generates jets and constitutes the GRB ``engine'', as in the
46: observed ejection of relativistic ``cannonballs'' of plasma by
47: microquasars and active galactic nuclei. Here we investigate the
48: production of a GRB as the jetted cannonballs exit the supernova shell
49: reheated by their collision with it, emitting highly forward-collimated
50: radiation. Each cannonball corresponds to an individual pulse in a GRB. We
51: cannot predict the timing sequence of these pulses, but the Cannonball
52: Model fares very well in describing the total energy, energy spectrum, and
53: time-dependence of the individual pulses.
54:
55:
56: \end{abstract}
57:
58: \keywords{gamma rays bursts, supernovae, black holes}
59:
60:
61: \section{Introduction}
62:
63:
64: Once upon a time, Gamma Ray Bursts (GRBs) constituted a sheer mystery,
65: whose unassailability was reflected in the scores of extremely different
66: ideas proposed to explain them. In spite of giant strides in the recent
67: observations ---the discovery of GRB afterglows (Costa et al. 1997; van
68: Paradijs et al. 1997), the discovery of the association of GRBs with
69: supernovae (Galama et al. 1998), and the measurements of the reedshifts
70: of their host galaxies (Metzger et al. 1997)--- the origin of GRBs is still an
71: unresolved enigma. In the recent past, the generally accepted view has
72: been that GRBs are generated by synchrotron emission from fireballs, or
73: firecones, produced by collapses or mergers of compact stars (Paczynski
74: 1986; Goodman et al. 1987; Meszaros and Rees 1992) by failed supernovae or
75: collapsars (Woosley 1993; Woosley and MacFadyen 1999; MacFadyen and
76: Woosley 1999, Woosley 1999) or by hypernova explosions (Paczynski 1998).
77: But various observations suggest that most GRBs are produced in supernova
78: events by highly collimated ultrarelativistic jets (Shaviv and Dar 1995;
79: Dar 1998; Dar and Plaga 1999; Cen 1999; Dar and De R\'ujula 2000a and
80: references therein).
81:
82: In a previous paper (Dar and De R\'ujula 2000a) we introduced a
83: relativistic-cannonball
84: model in which GRBs are produced by ``cannonballs'' (CBs)
85: of baryonic plasma emitted subsequently to a core-collapse supernova (SN)
86: explosion, and are observable when they happen to point
87: close to our direction. There, we concentrated on GRB afterglows
88: ---due to bremsstrahlung and
89: synchrotron emission from the CBs after they become transparent to their
90: own enclosed
91: radiation--- to emphasize how, in the case of GRB 980425,
92: it might be possible to observe the CBs' ``superluminal'' motion.
93: In this paper we briefly review the CB model and we derive its predictions
94: for the properties of the $\gamma$-rays in a GRB, generated as the
95: forward-collimated and blue-shifted
96: thermal radiation from a succession of
97: fast-moving, cooling and expanding CBs ---previously heated
98: by their collision with the SN shell---
99: escapes from the transparent outer regions of the shell.
100: We study the $\gamma$-ray distributions
101: in time, their energy-spectrum and the correlations between these two
102: observables, showing that the CB model explains the main observed features
103: of GRBs.
104:
105:
106:
107:
108: \section{Jets in astrophysics}
109:
110: Relativistic jets seem to be emitted by all astrophysical systems wherein
111: mass is accreted at a high rate from a disk onto a central compact object
112: (for a review, see Mirabel and Rodriguez 1999a).
113: Highly relativistic jets have been observed in galactic
114: sources, such as the microquasars GRS 1915+105 (Mirabel and
115: Rodriguez 1994, 1999a,b; Rodriguez and Mirabel 1999)
116: and GRO J165-40 (Tingay et al. 1995) where mass is accreted onto a stellar
117: black hole, and in many active galactic nuclei hosting a massive
118: black hole. These jets are not continuous streams: they consist of
119: individual blobs of plasma (plasmoids or cannonballs), and their firing
120: coincides with a sudden removal of the
121: accretion-disk material (Belloni 1997; Mirabel and Rodriguez 1999b).
122: Cannonballs in microquasars --and presumably also in quasars--
123: are emitted in pairs, moving in opposite directions.
124:
125: As they travel, microquasar CBs are observed to expand at a
126: speed comparable to, or smaller than, the sound speed of a relativistic
127: plasma (c/$\sqrt{3}$ in their rest system) probably because the
128: energy density of their enclosed radiation is comparable to that of their
129: matter constituency. As they become transparent
130: and cool down, the CBs' lateral size stabilizes to a roughly constant
131: value, presumably constrained by magnetic
132: self-containment and/or by the ram pressure of the ambient material. Quasar
133: CBs show no measurable expansion as they travel, sometimes for as long as a
134: million light years (see, e.g., Bridle 2000; Wilson et al. 2000).
135: Galactic and quasar CBs expand explosively when finally stopped by the
136: material they traverse.
137:
138:
139:
140: \section{The cannonball model of GRBs}
141:
142:
143: The ejection of matter in a supernova (SN) explosion is not fully understood. The
144: known mechanisms for imparting the required kinetic energy to the ejecta
145: are inefficient: the theoretical understanding of core-collapse SN events is
146: still unsatisfying. It has been proposed (De R\'ujula 1987; Woosley 1993,
147: Dar and De R\'ujula 2000a and references therein)
148: that the result of a SN event is not just a compact object plus
149: the ejecta: a fraction of the parent star may be ejected, but
150: another fraction of its mass may fall back onto the newly born
151: compact object. For vanishing angular momentum,
152: the free-fall time of a test-particle from a parent
153: stellar radius ${\rm R_\star}$ onto an object of mass ${\rm M_c}$ is:
154: \begin{eqnarray}
155: {\rm t_{fall}}&&{\rm ={\pi\,\left[{R_\star^3\over 8\,G\,M_c}\right]^{1/2} }}
156: \nonumber \\
157: &&{\rm \sim 1\; day\; \left[{R_\star\over 10^{12}\;cm}\right]^{3/2}\;
158: \left[{1.4\;M_\odot\over M_c}\right]^{1/2}}\, .
159: \label{tfall}
160: \end{eqnarray}
161: The free-fall time is shorter if the mass of the falling material
162: is not small relative to that of the compact object.
163: The fall-time is longer (except for material falling from the polar directions)
164: if the specific angular momentum is considerably
165: large, as it is in most stars. The estimate of Eq.(\ref{tfall})
166: is therefore a rough one.
167:
168:
169: It is quite natural to suppose that infalling material with non-vanishing
170: angular momentum settles
171: into an orbiting disk, or a thick torus if its mass is comparable
172: to ${\rm M_c}$. We assume that, as observed
173: in other cases of significant accretion
174: onto a compact object (microquasars and active galactic nuclei)
175: in which the infalling material is processed in a series of
176: ``catastrophic'' accretions, jets of
177: relativistic CBs of plasma are ejected. We presume their
178: composition to be ``baryonic'', as it is in the jets of
179: SS 433, from which Ly$_\alpha$ and Fe K$_\alpha$ lines have been detected
180: (Margon 1984), although the violence of the relativistic jetting-process
181: should in our case break most nuclei into their constituents.
182:
183: The mechanism producing relativistic jets in accretion
184: processes and its timing-sequence are not understood
185: (for suggested possibilities see, e.g., Blandford and Znajek 1977;
186: Meszaros and Rees 1997).
187: In our model we assume that a series
188: of CBs is ejected, each one giving rise to one of the
189: ``pulses'' of a specific GRB. After a few pulses the engine
190: runs out of fuel, and the $\gamma$-ray activity ceases.
191: The timing sequence of the successive pulses
192: we are unable to predict, but, as we shall show,
193: the CB model is quite successful in describing the time-dependence
194: of the $\gamma$-ray flux {\it within single GRB pulses}.
195:
196: In brief, the CB model is the following. A sequence of oppositely-directed pairs
197: of cannonballs is emitted at a time $\rm t_{fall}$ of ${\cal{O}}(1)$ day after
198: a SN core-collapse. By this time the SN outer shell, traveling at
199: a velocity $\rm v_S \sim c/10$ (see, e.g., Nakamura et al. 2000)
200: has moved to a distance:
201: \begin{equation}
202: \rm R_S=2.6 \times 10^{14} \;cm\;\left({t_{fall}\over 1\;d}\right)\;
203: \left({10\,v_S\over c}\right) .
204: \label{Rs}
205: \end{equation}
206: We adopt $\rm R_S=2.6 \times 10^{14}$ cm as a ``reference'' value,
207: to which our results will be scaled. The reference values of various
208: relevant parameters ---that serve as bench-marks to which to scale our results
209: and imply no strong commitment to their particular choices---
210: are listed in Table I, for quick reference. We denote with a
211: bar the actual value of a parameter in the units of its reference value
212: so that $\rm \overline{R}_S$, for instance, means a given SN-shell radius
213: divided by $2.6 \times 10^{14}$ cm.
214:
215:
216: Only if traveling at a small
217: angle $\theta$ relative to the line of sight, will a CB be
218: visible. As it hits the SN shell, the CB slows down and heats up.
219: Its radiation is obscured by the shell
220: up to a distance of order one radiation length from the shell's
221: outer surface. As this point is reached, the GRB
222: is emitted by a CB that
223: continues to travel, expand and cool down, its radiation being
224: boosted and collimated by the CB's ultrarelativistic motion.
225: We do not discuss in this paper the GRB
226: afterglows (Dar and De R\'ujula 2000a), the flash of X-ray lines
227: and the achromatic flare in the afterglow as the electrons and
228: protons in the GRB recombine (Dar and De R\'ujula 2000b),
229: nor the flux of high energy neutrinos and $\gamma$-rays
230: produced by the decays of pions made in the CB's collision
231: with the SN shell (Dar and De R\'ujula 2000c).
232:
233: There are other events in which a variety of GRBs
234: could be produced by mechanisms similar to the ones we have
235: discussed: large mass accretion episodes in binaries including
236: a compact object, mergers of neutron stars with neutron stars
237: or black holes (Paczynski 1986, Goodman et al. 1987),
238: transitions of neutron stars
239: to hyperon- or quark-stars (Dar 1999; Dar and De R\'ujula, 2000d), etc.
240: In each case, the ejected cannonballs would make GRBs by
241: hitting stellar winds or envelopes, circumstellar mass or light.
242: We discuss only core-collapse SN explosions, as the GRBs
243: they would produce
244: by our mechanism, although relatively ``standard'', satisfactorily
245: reproduce the general properties of the heterogeneous
246: ensemble of GRBs.
247:
248: \section{Four ``clocks'' and three energy scales}
249:
250: Let $\rm \gamma=1/\sqrt{1-\beta^2}={E_{CB}/(M_{CB}c^2)}$ be
251: the Lorentz factor
252: of a CB, that diminishes with time as the CB hits the SN shell
253: and as it subsequently plows through the interstellar medium. Four
254: clocks ticking at different paces are relevant to a CB's history.
255: Let $\rm t_{SN}$ be the
256: local time in the SN rest system, $\rm t_{CB}$ the time in the CB's
257: rest system, $\rm t_{Ob}$ the time measured by
258: a nearby observer viewing the CB at an angle $\theta$
259: away from its direction of motion and $\rm t$ the time
260: measured by an earthly observer viewing the CB at
261: the same angle, but from a ``cosmological'' distance
262: (redshift $\rm z\neq 0$).
263: Let x be the distance traveled by the CB in the SN rest system.
264: The relations between the above quantities are:
265: \begin{eqnarray}
266: &&\rm
267: dt_{SN}=\gamma\,dt_{CB}=\rm{dx\over\beta\, c}\, ;\;\;
268: \nonumber \\
269: &&\rm
270: dt_{CB}\equiv \delta\,dt_{Ob}\, ;\;\; dt=(1+z)\,dt_{Ob}\;,
271: \label{times}
272: \end{eqnarray}
273: where the Doppler factor $\delta$ is:
274: \begin{equation}
275: \rm
276: \delta\equiv\rm{1\over\gamma\,(1-\beta\cos\theta)}
277: \simeq\rm {2\,\gamma\over (1+\theta^2\gamma^2)}\; ,
278: \label{doppler}
279: \end{equation}
280: and its approximate expression is valid for $\theta\ll 1$ and $\gamma\gg 1$,
281: the domain of interest here.
282: Notice that for large $\gamma$ and $\theta\gamma$
283: not large, there is an enormous ``relativistic aberration'':
284: $\rm dt\sim dt_{SN}/\gamma^2$ and the observer sees
285: a long CB story as a film in extremely fast motion.
286:
287: The energy of the photons radiated by a CB
288: in its rest system, $\rm E^\gamma_{CB}$, their energy
289: in the direction $\theta$
290: in the local SN system, $\rm E^\gamma_{SN}$, and the photon
291: energy, E, measured by a cosmologically distant observer,
292: are related by:
293: \begin{equation}
294: \rm E^\gamma_{CB}= {E^\gamma_{SN}\over \delta}
295: \, ;\;\;E^\gamma_{SN}=(1+z)\,E\; ,
296: \label{energies}
297: \end{equation}
298: with $\delta$ as in Eq.(\ref{doppler}).
299:
300:
301: \section{The making of a GRB}
302: \subsection{Jet energy and CB mass}
303:
304: Let ``jet'' stand for the ensemble of CBs emitted in one direction in a SN
305: event. If a momentum imbalance between the opposite-direction jets is
306: responsible for the large peculiar velocities ${\rm v_{NS}\approx 450\pm
307: 90~ km~s^{-1}}$ (Lyne and Lorimer 1994) of neutron stars born in SNe, the
308: jet kinetic energy $\rm E_{jet}$ must be, as we shall assume for our GRB
309: engine, larger than $\sim 10^{52}$ erg (e.g. Dar and Plaga 1999). The
310: jet-emitting process may be ``up-down'' symmetric to a very good
311: approximation, in which case the jet energies may be much bigger. There is
312: evidence that in the accretion of matter by black holes in quasars
313: (Celotti at al. 1997; Ghisellini 2000) and microquasars (Mirabel and
314: Rodriguez 1999a,b) the
315: efficiency for the conversion of gravitational binding energy into jet
316: energy is surprisingly large.
317: If in the production of CBs the central compact object
318: in a SN ingurgitates several solar masses, it is not
319: out of the question that $\rm E_{jet}$ be as large as
320: $\rm M_\odot c^2\sim 1.8\times 10^{54}$ erg. We shall
321: adopt here a compromise value, $10^{53}$ ergs, as the reference
322: jet energy.
323:
324: Let $\rm \gamma_{in}$ be the Lorentz factor of a cannonball
325: as it is fired. Let $\rm E_{CB}=f\,E_{jet}$ be the energy of a CB;
326: on average GRBs have some five to ten significant pulses, so that the fraction
327: f may typically be 1/5 or 1/10. We shall adopt $\rm E_{CB}=10^{52}$ erg
328: as our reference value. For this value, the CB's mass is
329: comparable to an Earth mass:
330: %${\rm M_{jet}\sim 1.5 \times 10^{-6}\,M_{NS}\,(10^3/\gamma)}$,
331: ${\rm M_{CB}\sim 1.8\, M_\otimes (10^3/\gamma_{in})}$, for a Lorentz factor
332: of $\rm\gamma_{in}={\cal{O}}(10^3)$, that we shall find to be ``typical''.
333:
334: \subsection{CB deceleration by the SN shell}
335:
336: Let $\rm \beta_{in}\, c$ be the expansion velocity of a CB,
337: in its rest system, as it travels from the point of emission to
338: the point at which it reaches the SN shell. For the reference
339: value of $\rm \beta_{in}$, as reported in Table I, we use $1/(10\,\sqrt{3})$:
340: one tenth of the sound velocity of a relativistic plasma.
341: The CB reaches the shell with a radius
342: \begin{equation}
343: \rm R_{CB}\sim R_S\,{\beta_{in}\over \gamma_{in}}
344: \label{radiusCB}
345: \end{equation}
346: and sweeps
347: up a ``target'' mass
348: $\rm M_T\sim\pi\,R_{CB}^2\,X_S=M_S\,\beta_{in}^2/(4\,\gamma_{in}^2)$,
349: or some $\rm \sim 2.8\times 10^{-3}\,M_\otimes$, for our reference
350: parameter values and $\rm\gamma_{in}=10^3$.
351: The CB and the SN shell are ``thick'' in the sense of extending over
352: many radiation lengths and many nucleon-nucleon interaction lengths.
353: A high-energy nucleon suffering successive interactions in a dilute
354: gas or plasma loses roughly 2/3 of its energy to $\pi^\pm$
355: production, with most of the pion energy being
356: carried away by the neutrinos in $\pi\to \mu\,\nu$ decays
357: and the subsequent $\mu$ decays. The electrons
358: from $\mu$ decay and the photons from $\pi^0$ decay locally
359: deposit roughly 1/3 of the original nucleon energy.
360:
361: The Lorentz factor of the CB after it has swept the SN shell,
362: $\rm \gamma_{out}$, is simply the
363: ratio of the total energy to the invariant mass
364: ($\rm \sqrt{s}=M\,c^2$) of the outgoing object:
365: \begin{equation}
366: \rm \gamma_{out}\simeq {E_{CB}/3\over \sqrt{s}}
367: \simeq{E_{CB}/3\over
368: \sqrt{2\,M_T\,c^2\,E_{CB}/3+M^2_{CB}\,c^4}}\; ,
369: \label{gammaout1}
370: \end{equation}
371: where we have used $\rm E_{CB}\gg M_Tc^2$. Substituting for
372: $\rm M_T$ and $\rm M_{CB}$ as functions of $\rm\gamma_{in}$
373: and $\rm\beta_{in}$, one obtains:
374: \begin{equation}
375: \rm \gamma_{out} \simeq \gamma_{in}
376: \;\sqrt{2\,E_{CB}\over 3\,\beta_{in}^2\,M_S\, c^2+18\,E_{CB}}
377: \label{gammaout}
378: \end{equation}
379: whose limiting values are:
380: \begin{eqnarray}
381: && \rm \gamma_{out}\sim {\gamma_{in}\over 3}\;\;\;\;
382: (for\; 6\,E_{CB}\gg \beta_{in}^2\,M_S\,c^2)\, , \nonumber\\
383: && \rm \gamma_{out}
384: \sim {\gamma_{in}\over 3\,\overline{\beta}_{in}}\,
385: \left[{\overline{E}_{CB}\over \overline{M}_S}\right]^{1\over 2}\;\;\;\;
386: (for\; 6\,E_{CB}\ll \beta_{in}^2\,M_S\,c^2)\; .
387: \label{gammaout2}
388: \end{eqnarray}
389: For our reference $\rm \gamma_{out}\sim 10^3$,
390: the values of $\rm \gamma_{in}$ implied by Eqs.(\ref{gammaout2})
391: may look surprisingly large. But Eqs.(\ref{gammaout2}) do not
392: depend on $\rm R_S$: any relativistic jet exiting from the core of a SN
393: encounters the same amount of non-collapsed material, and must have a
394: $\rm \gamma_{in}$ considerably larger than $\rm\gamma_{out}$.
395:
396: The very large value of $\rm\gamma_{in}$ ($\sim 3 \times 10^3$ for our reference
397: parameters) implies that the fractional solid angle covered by a CB as it hits the
398: SN shell is tiny: $\rm \beta_{in}^2/(4\,\gamma_{in}^2)\sim 10^{-10}$,
399: again for our reference parameters. This presumably makes it unlikely for
400: successive CBs to hit precisely the same spot in the SN shell:
401: CB-CB collisions and mergers may be the exception, rather than the rule.
402:
403: \subsection{Attenuation of the $\gamma$ rays}
404:
405: The density profile of the outer layers of a SN shell as a function
406: of the distance x to the SN center can be measured from the photometry,
407: spectroscopy and evolution of the SN emissions (see e.g. Nakamura et al. 2000
408: and references
409: therein). The observations can be fit by a power law,
410: $\rm x^{-n}$, with $\rm n \sim 4\; to\, 8$. Our results are sensitive to
411: this density profile only in the outer region where the SN shell
412: becomes transparent (and the measurements are made), so that we
413: can adopt the same profile at all $\rm x>R_S$:
414: \begin{equation}
415: \rm \rho(x)=\rm\rho(R_S)\,\Theta(x-R_S)\,\left[{R_S\over x}\right]^n\, .
416: \label{profile}
417: \end{equation}
418: The SN-shell grammage still in front of a CB located at x is:
419: \begin{equation}
420: \rm X_S(x)=\int_x^\infty \, \rho(y)\,dy=
421: {M_S\over 4\,\pi\, R_S^2}\; \left[{R_S\over x}\right]^{n-1}\, .
422: \label{SNgram}
423: \end{equation}
424:
425: For photons in the MeV domain the attenuation length is similar, within
426: a factor 2, in all elements from H to Fe (Groom et al., 2000), and can be
427: roughly approximated by:
428: \begin{equation}
429: \rm X_\gamma(E)\sim 1.0\,(E/keV)^{0.33}\; g\, cm^{-2}\; .
430: \label{Xgamma}
431: \end{equation}
432: The value of $\rm X_\gamma(E)$ in the $\rm E=10$ keV to 1 MeV domain
433: (2.1 to 9.8 gr/cm$^2$) is close to the attenuation length in a hydrogenic
434: plasma ($\rm X_\gamma^{ion}\simeq m_p/\sigma_{_T}\simeq 2.6$ gr/cm$^2$,
435: with $\rm m_p$ the proton's mass and
436: $\rm \sigma_{_T}\simeq 0.65\times 10^{-24}$ cm$^2$ the Thomson
437: cross-section). Therefore, it makes little difference in practice whether
438: or not we take into account that the SN-shell material reached
439: by the CB may be ionized by its previously emitted radiation.
440: Equating $\rm X_S(x)=X_\gamma(E)$ and solving for x, we define a useful
441: quantity: $\rm x_{tp}(E)$, the position at which the SN shell becomes
442: (one-radiation-length) transparent:
443: \begin{equation}
444: \rm x_{tp}(E) = R_S\;\left[{M_S\over 4\,\pi\, R_S^2}\;
445: {1\over X_\gamma(E)}\right]^{1\over n-1}\propto E^{-0.33/(n-1)}\; ,
446: \label{SNtransparent}
447: \end{equation}
448: whose energy dependence is extremely weak.
449: Blue-shifted to the SN rest-system, as in Eq.(\ref{energies}),
450: GRB photons have energies
451: in the MeV range. Let $\rm \tilde x_{tp}\equiv x_{tp}(1\; MeV)$.
452: For our reference parameters, some representative results are:
453: $\rm \tilde x_{tp}\simeq 2.9\,R_S$ for $\rm n=8$,
454: $\rm \tilde x_{tp}\simeq 4.5\,R_S$ for $\rm n=6$.
455: At $\rm \tilde x_{tp}$, $\rm \rho(x)$
456: is orders of magnitude smaller than at $\rm x\sim R_S$, where most of
457: the SN-shell's mass is steeply concentrated. This will simplify our
458: discussion, for it is a fair approximation to have the CB
459: slow down at heat up close to $\rm x=R_S$, and proceed thereafter unperturbed
460: by the SN-shell material, except inasmuch as little of its radiation
461: can escape before it reaches $\rm x=\tilde x_{tp}$. At that point, the CB
462: has expanded from the radius $\rm R_{CB}$ of Eq.(\ref{radiusCB}) to
463: a radius at transparency:
464: \begin{equation}
465: \rm R_{CB}^{tp}\simeq R_{CB}+
466: {\tilde x_{tp}-R_S\over \gamma_{out}}\;\beta_{out}
467: \simeq
468: {\tilde x_{tp}-R_S\over \gamma_{out}}\;\beta_{out}\; ,
469: \label{Rtrans}
470: \end{equation}
471: some $2.9\times 10^{11}$ cm, for our reference parameters.
472: The CB itself becomes transparent to the radiation it encloses later,
473: when it reaches a radius
474: $\rm \widetilde R^{tp}_{CB}\simeq [3\,
475: M_{CB}\,\sigma_T/(4\,\pi\,m_p)]^{1\over 2}$.
476: We expect the CB to stop expanding at a proper
477: quasi-relativistic rate $\rm\beta_{out}$ soon after it becomes
478: transparent and its inner radiation pressure drops abruptly: the inertial
479: mildly relativistic
480: transverse motion of its matter constituents is slowed-down by
481: interstellar material and, perhaps, by self-confining magnetic fields.
482:
483:
484:
485: \subsection{Total energy of a GRB pulse}
486:
487: A CB expanding as a quasi-relativistic plasma ought to reach
488: the SN shell with a shape (in its rest system) very close to spherical.
489: The microscopic description of what happens as the CB and the
490: material of the SN shell collide and coalesce is elaborate
491: (Dar and De R\'ujula, 2000c).
492: Much of the available energy is deposited at the CB's front surface
493: by nucleons sharing their energy and $\gamma$'s from
494: $\pi^0$-decay depositing theirs. Electrons from $\mu$ decay deposit
495: their energy much deeper into the CB. The ionized CB's material
496: is hot and dense enough for the deposited energy to thermalize very fast.
497: As it impinges the SN shell, a CB may have a tendency to get flattened,
498: but the velocities, $\beta\approx 1-1/(2\gamma^2)$,
499: corresponding to $\rm\gamma_{in}$ and $\rm\gamma_{out}$
500: are so similar that no significant flattening occurs between
501: the time the CB hits the shell and the time it reaches
502: the point at which the shell becomes
503: transparent to the CB's radiation: flattening is subdominant relative
504: to the CB's expansion. For the subsequent estimates we approximate
505: the CB as a spherical body with a uniform internal temperature.
506:
507: The proper temperature $\rm T_0$ acquired by the CB as it hits the SN shell
508: is high enough for the CB's internal-radiation
509: energy-density to be much larger
510: than the mass-energy density of its matter constituents. Consequently,
511: $\rm T_0$ can be estimated by equating the total internal radiation
512: energy to the invariant mass in the CB-SN shell collision:
513: \begin{equation}
514: \rm T_0\simeq
515: \left[{3\over 8\,\pi\, a}\,\sqrt{3\over 2}\;
516: {(3\,\beta_{in}^2\,M_S\,c^2+18\,E_{CB})^{3/ 2}
517: \over \sqrt{E_{CB}}}\;{\gamma_{out}^2\over R_S^3}\right]^{1\over 4}
518: ,
519: \label{T0}
520: \end{equation}
521: where $\rm a\simeq 1.37 \times 10^{14}$ erg cm$^{-3}$ keV$^{-4}$
522: is the radiation-density constant.
523: The result, $\rm T_0\sim 3.4$ keV for our reference parameters,
524: is very insensitive to their values,
525: but for the $\rm R_S^{-3/4}$ dependence on the SN-shell's radius.
526:
527:
528: Since the SN-shell's material is highly concentrated close to $\rm x=R_S$,
529: as in Eq.(\ref{SNgram}), we can take $\rm T_0$ to be the temperature
530: at that point. A rough estimate of the total energy in a GRB pulse can
531: be obtained as follows.
532: While on the part of the shell that is not transparent,
533: the CB does not lose much energy via surface radiation, so that it
534: expands quasi-adiabatically
535: at roughly constant $\rm R_{CB}(t)\,T(t)$. At the point
536: at which the shell becomes transparent, the internal-radiation energy in the
537: CB is reduced, from the value $\rm \sqrt{s}$ of Eq.(\ref{gammaout1}), to
538: $\rm E_{tp}\simeq \sqrt{s}\; R_{CB}/R_{CB}^{tp}$.
539: Approximately 1/e of this energy is emitted thereafter, its value in the
540: CB's rest system is:
541: \begin{equation}
542: \rm E_{pulse}^{rest}\simeq {1\over 3\,e}\;{E_{CB}\over \gamma_{in}}\;
543: {R_{S}\over \tilde x_{tp}-R_S}\;
544: {\beta_{in}\,\gamma_{out}\over \beta_{out}\,\gamma_{in}}\; ,
545: \label{EGRB}
546: \end{equation}
547: whose limiting values are:
548: \begin{eqnarray}
549: &&\rm E_{pulse}^{rest}\simeq
550: 4.5\times 10^{45}\; erg\,\left[{3\,R_S\over \tilde x_{tp}-R_S}\right]\,
551: {1\over \overline{\gamma}_{out}}\,
552: {\overline{\beta}_{in}\over\overline{\beta}_{out}}\nonumber \\
553: &&\rm \;\;\;\;\;\;\;\;\;\;\;
554: (for\; 6\,E_{CB}\gg \beta_{in}^2\,M_S\,c^2)\, ,
555: \nonumber\\
556: &&\rm E_{pulse}^{rest}\simeq
557: 4.5\times 10^{45}\; erg\,\left[{3\,R_S\over \tilde x_{tp}-R_S}\right]\,
558: {1\over \overline{\gamma}_{out}\,\overline{\beta}_{in}\,\overline{\beta}_{out}}\,
559: {\overline{E}_{CB}^2\over \overline{M}_S}
560: \nonumber\\
561: &&\rm \;\;\;\;\;\;\;\;\;\;\;
562: (for\; 6\,E_{CB}\ll \beta_{in}^2\,M_S\,c^2)\; .
563: \label{EGRB1}
564: \end{eqnarray}
565:
566: An observer at rest,
567: located at a known luminosity distance $\rm D_L(z)$ from the CB and
568: viewing it at an angle $\theta$ from its direction of motion would measure
569: a ``total'' (time- and energy-integrated) fluence per unit area:
570: \begin{equation}
571: \rm {dF\over d\Omega}\simeq {1+z\over 4\,\pi\,D_L^2}
572: \,{E_{pulse}^{rest}}\;\delta^3\; ,
573: \label{dfdomega}
574: \end{equation}
575: where $\delta=\delta[\gamma,\theta]$ is given, here and in what follows,
576: by Eq.(\ref{doppler})
577: with $\rm \gamma=\gamma_{out}$.
578: The ``spherical'' energy deduced from this result would be an overestimate
579: of the true energy $\rm E_{pulse}^{rest}$
580: by the last factor in Eq.(\ref{dfdomega}), which, for $\rm \gamma_{out}=10^3$
581: and $\theta\gamma\sim {\cal{O}}(1)$, can be as large as $\sim 10^9$.
582: Enhanced by a factor ranging up to this large number, the GRB-pulse energies of
583: Eq.(\ref{EGRB}) can easily reproduce the observations, as discussed
584: in detail in Section 7.
585:
586: Armed with an expression such as Eq.(\ref{dfdomega}) one can embark in the
587: exercise of studying the extent to which GRBs are standard candles, by
588: checking whether the observations at fixed redshift are statistically compatible
589: with this expression for a uniformly distributed $\cos\theta$ distribution.
590: But the current number of GRBs with measured redshifts is only fifteen,
591: and their deduced total energies are affected by absorption, by experimental
592: efficiency and threshold effects, etc. We do not, in this paper, attempt such
593: an analysis, that has been initiated, with encouraging results, by Plaga (2000),
594: who uses ---to extract redshifts from a large collection of GRBs--- the
595: ``Cepheid-like'' relationship between variability and luminosity proposed
596: by Fenimore and Ramirez-Ruiz (2000).
597:
598:
599:
600: \subsection{Energy and time dependence of a $\gamma$-ray pulse}
601:
602: A CB, as it reaches the transparent outskirts of a SN shell, is
603: expanding and cooling and its radiation is becoming visible to
604: the observer. In what follows it is convenient to measure the
605: GRB observer's time,
606: t, setting $\rm t=0$ at the moment of the encounter of the CB and the
607: SN shell. The time of (one-radiation-length) transparency is then:
608: \begin{equation}
609: \rm t_{tp}\simeq {1+z\over \gamma_{out}\,\delta}\;
610: {\tilde x_{tp} - R_S\over c}\; .
611: \label{tttp}
612: \end{equation}
613: The CB temperature at $\rm t= t_{tp}$ is:
614: \begin{equation}
615: \rm T_{tp}\sim\rm\left[{3\over 4\,\pi\,a}\;
616: {E_{pulse}^{rest}\over (R_{CB}^{tp})^3}
617: \right]^{1\over 4}\; ,
618: \label{Ttrans}
619: \end{equation}
620: whose limiting values are:
621: \begin{eqnarray}
622: &&\rm T_{tp}\sim 0.1\;keV \left[{3\,R_S\over \tilde x_{tp}-R_S}\right]\,
623: {\overline{\gamma}_{out}^{1\over 2}\,
624: \overline{\beta}_{in}^{1\over 4}\over\overline{\beta}_{out}}\nonumber\\
625: &&\rm \;\;\;\;\;\;\;\;\;\;\;
626: (for\; 6\,E_{CB}\gg \beta_{in}^2\,M_S\,c^2)\, ,
627: \nonumber\\
628: &&\rm T_{tp}\sim 0.1\;keV \left[{3\,R_S\over \tilde x_{tp}-R_S}\right]\,
629: {\overline{\gamma}_{out}^{1\over 2}
630: \over \overline{\beta}_{in}^{1\over 4}\,\overline{\beta}_{out}}\;
631: {\overline{E}_{CB}^{1\over 2}\over \overline{M}_S^{1\over 4}}
632: \nonumber\\
633: &&\rm \;\;\;\;\;\;\;\;\;\;\;
634: (for\; 6\,E_{CB}\ll \beta_{in}^2\,M_S\,c^2)\; .
635: \label{Ttrans1}
636: \end{eqnarray}
637:
638: The time-dependences
639: of the CB's radius, its temperature, and the distance x of the CB from the
640: SN's center are, for $\rm t>0$:
641: \begin{eqnarray}
642: &&\rm R_{CB}[t]\simeq R_{CB} + R_{CB}^{tp}\,{t\over t_{tp}}
643: \sim R_{CB}^{tp}\,{t\over t_{tp}}
644: \, ,\nonumber\\
645: &&\rm T[t]\simeq \rm T_{tp}\;{R_{CB}^{tp}\over R_{CB}[t]}\, ,\nonumber\\
646: &&\rm x[t]\simeq R_S+{\delta \,\gamma_{out}\over 1+z}\;c\, t\;.
647: \label{timedeps}
648: \end{eqnarray}
649:
650: Let the number of photons per unit time and energy, assumed to be
651: isotropically emitted by the CB in its rest system, be:
652: \begin{equation}
653: \rm {dn_\gamma\over dE_\gamma\;dt_{CB}}\equiv F(E_{CB}^\gamma,T)\, .
654: \label{restemission}
655: \end{equation}
656: Using Eqs.(\ref{times}-\ref{energies}) to change variables to
657: $\rm E$ and t (the $\gamma$-ray energy and time in the
658: observer's frame), we obtain:
659: \begin{equation}
660: \rm {dn_\gamma\over dE\,dt}\simeq
661: F\left(E\,{1+z\over\delta},T[t]\right)\; .
662: \label{boostany}
663: \end{equation}
664: In the approximation in which the CB's emission in its rest system
665: is a thermal distribution from its surface, the function F is:
666: \begin{equation}
667: \rm F(E_{CB}^\gamma,T) \simeq
668: {2\,\pi\,\sigma\over \zeta(3)}\;(R_{CB}[t])^2\;
669: {(E_{CB}^\gamma)^2\over Exp\{{E_{CB}^\gamma/T}\}-1}\; ,
670: \label{thermal}
671: \end{equation}
672: where $\rm \sigma=c\,a/4$ is the Stefan-Boltzmann constant.
673:
674: The observed energy and time dependence of the photon intensity
675: (photon number per unit area, N)
676: of a single pulse in a GRB at an angle $\theta$ relative to the CB's
677: motion is then predicted to be:
678: \begin{eqnarray}
679: &&\rm {dN\over dE\,dt}\equiv {1+z\over 4\,\pi\,D_L^2}\;
680: \delta^2\, {dn_\gamma\over dE\,dt}\, ,\\
681: &&
682: \rm {dn_\gamma\over dE\,dt}\simeq {2\,\pi\,\sigma\over \zeta(3)}\;
683: {\left[R_{CB}[t]\;E\,(1+z)/\delta\right]^2\; Abs(E,t)\over
684: Exp\left\{E\,(1+z)/(\delta\, T[t])\right\}-1}\; ,
685: \label{boostthermal}
686: \end{eqnarray}
687: with $\rm R_{CB}[t]$ and $\rm T[t]$ as in Eqs.(\ref{timedeps}),
688: and where
689: \begin{equation}
690: \rm Abs(E,t)=
691: Exp\left[-{X_S(x[t])\over X_\gamma(E\,(1+z))}\right]
692: \label{attenuation}
693: \end{equation}
694: is the attenuation of the flux in the SN shell.
695:
696: For n in Eq.(\ref{SNtransparent}) as large as the observations
697: indicate ($\rm n\sim 6$), the absorption factor $\rm Abs(E^\gamma,t)$
698: rises very sharply from 0 to 1 around $\rm t=t_{tp}$, in which case
699: the width of a GRB pulse in energy and time is governed by
700: the exponential in the denominator of
701: Eq.(\ref{boostthermal}). The argument of that exponential
702: can be simply rewritten as $\rm E\,t/H$, with:
703: \begin{equation}
704: \rm H\equiv\rm {\tilde x_{tp}-R_S\over c\,\gamma_{out}}\, T_{tp}\; ,
705: \label{h}
706: \end{equation}
707: whose limiting values are:
708: \begin{eqnarray}
709: &&\rm H\sim 2.5\; keV\, s\;
710: { \overline{\beta}_{in}^{1\over 4}\over
711: \overline{\gamma}_{out}^{1\over 2}\,\overline{\beta}_{out}}
712: \nonumber\\
713: &&\rm \;\;\;\;\;\;\;\;\;\;\;
714: (for\; 6\,E_{CB}\gg \beta_{in}^2\,M_S\,c^2)\, ,
715: \nonumber\\
716: &&\rm H\sim 2.5\; keV\, s\;
717: { 1 \over
718: \overline{\gamma}_{out}^{1\over 2}\,\overline{\beta}_{out}\,
719: \overline{\beta}_{in}^{1\over 4}}\;
720: {\overline{E}_{CB}^{1\over 2}\over\overline{M}_S^{1\over 4}}
721: \nonumber\\
722: &&\rm \;\;\;\;\;\;\;\;\;\;\;
723: (for\; 6\,E_{CB}\ll \beta_{in}^2\,M_S\,c^2)\; .
724: \label{h1}
725: \end{eqnarray}
726:
727: \section{Some simplifications and approximate correlations}
728:
729: To guide the eye, we give a simplified
730: approximate form of Eq.(\ref{SNtransparent}), which we do not use
731: in our explicit calculations:
732: \begin{equation}
733: \rm {dN\over dE\,dt}\propto
734: {(E\,t)^2\over Exp\{E\,t/H\}-1}\,
735: Exp \left\{-\left[ {t_{tp}/ t}\right]^{n-1}\right\}
736: \;\Theta[t]\; .
737: \label{simple}
738: \end{equation}
739: The total photon intensity and energy flux are, in this
740: approximation:
741: \begin{equation}
742: \rm {dN\over\,dt}\propto
743: %{dI_\gamma\over dt}(t_{tp}) \,
744: \Theta[t]\; {t_{tp}\over t}\, Exp
745: \left\{-\left[ {t_{tp}/ t}\right]^{n-1}\right\}\,,
746: \label{simple2}
747: \end{equation}
748: %%
749: %%
750: \begin{equation}
751: \rm {F_E(t)}\propto
752: %F_\E(t_{tp})\,
753: \Theta[t]\; \left[{t_{tp}\over t}\right]^2\,
754: Exp \left\{-\left[ {t_{tp}/ t}\right]^{n-1}\right\}\,.
755: \label{simple3}
756: \end{equation}
757: Let the peak $\gamma$-ray
758: energy at a fixed time during a GRB pulse be defined as
759: $\rm E^\gamma_p(t) \equiv max\,[ E^2\,dI_\gamma/dE\, dt]$.
760: Its value is $\rm E^\gamma_p(t)\simeq 3.92\,\delta\,T[t]/(1+z) $, so that,
761: for t near or after $\rm t_{tp}$:
762: \begin{equation}
763: \rm E^\gamma_p(t) \simeq E^\gamma_p(t_{tp})
764: \;\Theta[t]\; {t_{tp}\over t}\,.
765: \label{simple4}
766: \end{equation}
767:
768: The total ``isotropic'' energy of a GRB pulse ---deduced from its observed
769: fluence assuming an isotropic emission--- can be deduced from
770: Eq.~(\ref{dfdomega}), to be:
771: \begin{equation}
772: \rm E_{iso}=
773: {4\,\pi\,D_L^2\, F \over 1+z}\simeq E_{pulse}^{rest}\, \delta^3\,.
774: \label{eisot}
775: \end{equation}
776:
777: If CBs were ``standard candles'' with fixed mass, energy
778: and velocity of expansion,
779: and if all SN shells had the same
780: mass, radius and density distribution, all differences between
781: GRB pulses would result from their different distances and angles
782: of observation. For such standard candles
783: it follows from Eqs.(\ref{times}-\ref{energies},\ref{eisot}) that the observed
784: durations (half widths at half maximum) of the photon intensity and
785: of the energy flux density ($\rm \Delta t_I$ and
786: $\rm \Delta t_F$),
787: their peak values ($\rm N_p$ and $\rm F_p)$, and the peak energy
788: ($\rm E^\gamma_p$) in a single GRB pulse are
789: roughly correlated to the total ``observed'' isotropic energy
790: ($\rm E_{iso}$) as follows:
791: \begin{equation}
792: \rm \Delta t_I\propto (1+z)\, [E_{iso}]^{-1/3}\,,
793: \label{twidthi}
794: \end{equation}
795: \begin{equation}
796: \rm \Delta t_F\propto (1+z)\, [E_{iso}]^{-1/3},
797: \label{twidthf}
798: \end{equation}
799: \begin{equation}
800: \rm N_p\propto E_{iso},
801: \label{Ipeak}
802: \end{equation}
803: \begin{equation}
804: \rm F_p\propto[ E_{iso}]^{4/3}\, (1+z)^{-1}\, ,
805: \label{Lpeak}
806: \end{equation}
807: \begin{equation}
808: \rm E^\gamma_p\propto [E_{iso}]^{1/3}\,(1+z)^{-1}\, .
809: \label{Epeak}
810: \end{equation}
811: These approximate correlations can be tested using the sample of 15 GRBs with
812: known redshifts.
813: Because of the strong dependence of the CB pulses
814: on the Doppler factor and their much weaker dependence on the
815: other parameters, they may be approximately satisfied
816: (see, e.g. Plaga 2000) in spite of the fact
817: that CBs and SN shells are likely to be sufficiently varied
818: not to result in standard candles.
819:
820: Within the standard-candle approximation there is also a simple
821: correlation between the rate and the fluence of GRBs. For the
822: region of the universe that is close enough to us to be approximately
823: homogeneous and Euclidean, Eq.(\ref{dfdomega}) implies that
824: $ \rm F=E_{rest}^{pulse}\, \delta^3 /(4\, \pi\, D^2)$.
825: If CBs were {\it stationary}, the
826: corresponding rate of GRB pulses with fluence larger than
827: a given $\rm F_0$
828: would satisfy the well known relation:
829: \begin{equation}
830: \rm \dot N(>F_0)\simeq \dot n_{CB} {4\,\pi\over 3}\,
831: \left[{E_{rest}^{pulse}\over 4\,\pi\,F_0}\right]^{3\over 2}\propto
832: F_0^{-3/2},
833: \label{euclid1}
834: \end{equation}
835: where $\rm \dot n_{CB}$ is the mean production rate of
836: CBs per unit volume. For our
837: highly relativistic CBs, whose ``isotropic'' energy
838: is multiplied by the factor $\rm \delta^3$, Eq.(\ref{euclid1})
839: is modified to:
840: \begin{equation}
841: \rm \dot N(>F_0)\simeq {3\over 7}\, {2^{7\over 2}\over\gamma^2}
842: \,\dot n_{CB}\,
843: {4\,\pi\over 3}\,\left[{\gamma^3\, E_{rest}^{pulse}\over
844: 4\,\pi\,F_0}\right]^{3\over 2}
845: \propto F_0^{-3/2},
846: \label{euclid2}
847: \end{equation}
848: yielding the same rate-to-fluence relation. The same power-law scaling
849: is obtained for the relation between the rate and the
850: peak-energy density-flux from CBs. Both relations
851: should be approximately satisfied by very bright (relatively nearby) GRBs.
852:
853: For distant GRBs the above relations are sensitive to the cosmological
854: model, to the not-well-determined SN- (or star-formation) rate and to
855: the strong selection effect favouring observations of distant GRBs with
856: large $\gamma$ and a small viewing angle $\theta$, and a correspondingly
857: large Doppler-enhancement $\delta$. Because of this, we do
858: not discuss here the rate-to-fluence relation for
859: faint GRBs (Yi, 1994; Plaga 2000).
860:
861: \section{Predictions of the Cannonball Model}
862:
863: Some common properties of GRB pulses
864: (for detailed light curves see Kippen 2000; Mallozzi 2000) are observed
865: to be:
866: \begin{itemize}
867: \item{(a)} The GRB fluences, integrated in energy and time,
868: lie within one or two orders of magnitude above or below
869: 10$^{-5}$ erg/cm$^2$ (see, e.g., Paciesas et al. 1999).
870: \item{(b)} Individual pulses are narrower in time, the higher the
871: energy interval of their individual photons
872: (see, e.g., Fenimore et al. 1995).
873: \item{(c)} Individual pulses rise and peak at earlier time, the higher the
874: energy interval of their individual photons
875: (see, e.g., Norris et al. 1999; Wu and Fenimore 2000)
876: \item{(d)} Individual pulses have smaller photon energies, the
877: later the time-interval of observation (see, e.g., Preece et al. 1998) .
878: \item{(e)} The energy spectrum of GRBs, or of their individual
879: pulses, if plotted as $\rm E^2\,dN/dE$, rises with energy as $\rm E^\alpha$,
880: with $\alpha \sim 1$, has a broad peak at $\rm E\sim 0.1$ to 1 MeV, and
881: decreases thereafter (see, e.g., Preece 2000).
882: \item{(f)} Most GRBs consist of pulses whose time-behaviour is
883: a fast rise followed
884: by an approximately exponential decay: a ``FRED'' shape. Some
885: GRBs have non-FRED, roughly time-symmetric pulses
886: (see e.g., Fenimore et al. 1995 and references therein)
887: The overwhelming majority of GRBs are either made of FRED or non-FRED
888: pulses: there are no GRBs with mixed pulse-shapes.
889: \end{itemize}
890:
891: All of the above items are properties of the CB model.
892:
893: In Fig.(\ref{flu}) we illustrate item (a) by plotting the total
894: fluence, estimated with use of Eq.(\ref{dfdomega}) and
895: varying one parameter at a time. Naturally, the highest sensitivity
896: is that to the viewing angle $\theta$, followed by that to
897: $\rm E_{CB}$ and z.
898: The remaining itemized GRB properties
899: all follow from Eq.(\ref{boostthermal});
900: items (b,c,d) are even apparent in the simplified Eq.(\ref{simple})
901: for the time and energy dependence of the $\gamma$-ray flux.
902: In Fig.(\ref{t3Es}) we illustrate items (b) and (c) by plotting
903: Eq.(\ref{boostthermal}) at three fixed $\gamma$-ray energies, for all parameters
904: fixed at the reference values of Table I. In Fig.(\ref{E3ts}) we
905: similarly illustrate item (d) in a plot at three different times, multiples
906: of the time of shell transparency. Item (e) is illustrated in Fig.(\ref{E})
907: where we plot $\rm E^2\, dN/dE$, obtained by integrating
908: Eq.(\ref{boostthermal}) over all times; the figure also reports the
909: sensitivity
910: to various parameters, by modifying them, one at a time, relative
911: to the reference parameters. In Fig.(\ref{t}) we
912: illustrate item (f) by plotting $\rm dN/dt$, obtained by integrating
913: Eq.(\ref{boostthermal}) over all energies above 30 keV. Once
914: again, we vary reference parameters as in Fig.(\ref{E}).
915: Redshift not being a free parameter specific to our model,
916: we separately illustrate in Fig.(\ref{z}) the z-dependence of the time-integrated
917: and energy-integrated versions of Eq.(\ref{boostthermal}).
918:
919:
920:
921: A look at Figs.(\ref{t}) and (\ref{z}) reveals that, for the parameter
922: ranges explored
923: therein, all the predicted GRB-pulse shapes are FREDs
924: and are relatively short in time (fractions of a second). Yet, these are not
925: general predictions of Eq.(\ref{boostthermal}). It is, for instance, quite
926: conceivable that the ejection of a shell in a SN explosion be due to
927: one or various CBs emitted immediately after core implosion: the shock
928: wave induced by their passage through the outer shells of the star
929: would trigger their ejection. In that case the outgoing shell would be
930: quite disrupted in the ``polar'' directions in which later CBs would
931: result in a GRB. It is also possible that a GRB be due to the passage
932: of CBs through material expelled by a parent-star's wind, as opposed
933: to the SN shell. In both cases, the density profile of the matter traversed
934: by a GRB may be very different from that described by a large index
935: $\rm n\sim 4$ to 8 in Eq.(\ref{profile}), indicated by observations of
936: complete SN shells, not of their small polar regions.
937: In Fig.(\ref{t2}) we illustrate these points by plotting
938: $\rm dN/dt$ for $\rm n=2,\,3$, with the rest of the parameters
939: at their reference values, and we also give an example with
940: $\rm n=6$ and a very large viewing angle $\rm \theta=20/\gamma_{out}$.
941: All three of these time-profiles are quite symmetrical non-FREDs
942: and have durations in the few-second range (it is also possible
943: to generate long-duration FREDs, as we shall see below in the
944: specific case of GRB 980425).
945:
946: In Fig.(\ref{6CB}) we plot a GRB with 6 CBs, shot at random times
947: in a 1.5 s interval and with random values of $\rm E_{CB}$ within
948: a factor of three of our reference value. All other parameters
949: in this figure, but the SN-shell density-profile index n,
950: have their reference values: the only difference
951: between Fig.(\ref{6CB}a) and Fig.(\ref{6CB}b) is that
952: $\rm n=8$ in the former, $\rm n=4$ in the latter.
953: These figures illustrate the obvious
954: fact that the correspondence between CBs and observed pulses
955: need not be biunivocal: a CB produces a GRB pulse, but an observed
956: pulse can be due to a superposition of CB subpulses. Notice that this is also
957: a way to obtain pulses that are very wide, or do not have FRED-
958: or symmetrical shapes.
959:
960:
961:
962: \section{Brief comparison to some data}
963:
964: Comparing a GRB theory with specific GRBs is a tricky task, for an
965: obvious reason: GRBs being all different, one may
966: be tempted to choose GRBs that fit the theory,
967: rather than doing the opposite. In this Section we investigate three
968: GRBs with measured redshifts. Of this ensemble, we use
969: the highest fluence event (Briggs et al 1999)
970: GRB 990123 ($\rm z=1.6$), to analize the energy
971: distribution; we use GRB 980425 (Kippen 2000),
972: whose redshift (Galama et al. 1998) is by far the smallest
973: ($\rm z=0.0085$) and yet has a conventional fluence
974: (Kippen et al. 1998), to study
975: the time-dependence of its single pulse; finally, we use GRB 990712
976: ($\rm z=0.4315$) to study the correlation between the
977: $\gamma$-ray energy- and time-distributions (Mallozzi 2000)
978: and to expose the limitations of the
979: CB model in its present simple form.
980:
981: \subsection{Energy dependence}
982: The predicted energy spectrum of a GRB is obtained by integrating
983: Eq.(\ref{boostthermal}) over all times. The resulting flux
984: distribution and the same result weighed with $\rm E^2$ are compared with
985: the GRB 990123 data in Fig.(\ref{123}). The parameters used
986: are $\rm\overline\beta_{in}=1/4$, $\rm\overline\beta_{out}=1$,
987: $\rm\overline{M}_S=1/5$, $\rm\overline{R}_S=1/2$,
988: $\rm\overline{E}_{CB}=20$, $\rm\overline{\gamma}_{out}=1.5$,
989: $\rm\overline{\theta}_S=1.46$, $\rm n=6$ and $\rm z=1.6$.
990: The value of $\rm E_{CB}$ may look large, but this is a
991: multiple-pulse GRB and the energy distribution is integrated
992: over all pulses: $\rm\overline{E}_{CB}=20$ corresponds only
993: to twice our reference value for $\rm E_{jet}$. For these parameters
994: the GRB fluence, as estimated via Eq.(\ref{dfdomega}), is the
995: observed 26.5 10$^{-5}$ erg cm$^{-2}$. Since the shape of
996: the energy distribution is insensitive to the various parameters,
997: as seen in Fig.(\ref{E}), it is easy to find many parameter ensembles
998: that result in the same prediction: the energy distribution by itself
999: is not a good observable to constrain the input, but is, on the other
1000: hand, a solid test of the model.
1001:
1002: The shape of the energy spectrum $\rm dN/dE$ of Fig.({\ref{123}a)
1003: can be easily understood. At the lower energies, the $\sim$1/E behaviour
1004: is the result of integration over thermal spectra with temperatures
1005: that decrease with time, see Fig.(\ref{E3ts}). The abrupt decrease
1006: of $\rm dN/dE$
1007: at the higher energies reflects the input thermal spectrum at
1008: the time and
1009: temperature at which the SN shell starts to become transparent.
1010:
1011: The comparison made in Fig.(\ref{123}) is quite satisfactory, particularly
1012: if one realises that many of the higher-energy data are but upper limits.
1013: In making this figure we used the thermal distribution of
1014: Eq.(\ref{thermal}), and the fact that at the higher energies the theory
1015: may undershoot relative to the data is to be expected. Indeed, the CB, in
1016: its rest system, is subject to a flux of high energy nuclei and electrons.
1017: While the electrons are being thermalized, they should contribute a
1018: nonthermal high-energy tail of photons emitted via the ``free-free''
1019: process. Such a power-law tail in an otherwise
1020: approximately-thermal emission is observed from
1021: young supernova remnants (see, e.g., Dyer et al. 2000) and clusters of
1022: galaxies (e.g., Fusco-Femiano et al. 1999; Rephaeli et al., 1999;
1023: Fusco-Femiano et al., 2000), both of which are systems
1024: wherein a dilute plasma at a temperature of
1025: $\cal{O}$(1 keV) is exposed to a flux of high energy cosmic rays.
1026:
1027:
1028: \subsection{Time dependence}
1029:
1030: In Fig.(\ref{425}) we compare the single-pulse
1031: light curve of GRB 980425 (Kippen 2000) with the CB theory, obtained by
1032: integrating Eq.(\ref{boostthermal}) over energy, in the
1033: 50-300 keV domain. The parameters used
1034: are $\rm\overline\beta_{in}=1/3$, $\rm\overline\beta_{out}=1/2$,
1035: $\rm\overline{M}_S=1$, $\rm\overline{R}_S=2$,
1036: $\rm\overline{E}_{CB}=10$ (corresponding to our reference
1037: jet energy in a single pulse), $\rm\overline{\gamma}_{out}=1/3$,
1038: $\rm\overline{\theta}_S=60$, $\rm n=8$ and $\rm z=0.0085$.
1039: For these parameters the GRB fluence, as estimated via
1040: Eq.(\ref{dfdomega}), is the observed $0.44\times 10^{-5}$ erg cm$^{-2}$.
1041: Notice that the value used for the viewing angle $\theta$ is very large:
1042: this is the explanation (Dar and De R\'ujula 2000a)
1043: why this particular GRB has a normal fluence, in spite of how close its
1044: progenitor (SN 1998bw) is to us.
1045:
1046: The comparison made in Fig.(\ref{425}) is entirely satisfactory. The parameter
1047: domain giving rise to a light curve with a particular shape, height and width
1048: is much more restricted than the corresponding domain for an energy
1049: distribution. Yet, we cannot entirely trust the approximate parameter
1050: values thus extracted, for the reasons to be discussed in the next two
1051: subsections.
1052:
1053: \subsection{The time-energy correlation}
1054:
1055: In Fig.(\ref{712}) we compare the single-pulse
1056: light curves of GRB 990712 (Mallozzi 2000) with the CB theory
1057: (the continuous red curves), obtained by
1058: integrating Eq.(\ref{boostthermal}) over energy, in the
1059: same domains as the data: 20-50 keV (BATSE channel 1.1), 50-100 keV (2.2),
1060: 100-300 keV (3.3) and $> 300$ keV (4.4). The parameters used
1061: are $\rm\overline\beta_{in}=1$, $\rm\overline\beta_{out}=1/3$,
1062: $\rm\overline{M}_S=1/4$, $\rm\overline{R}_S=3$,
1063: $\rm\overline{E}_{CB}=50$ (corresponding to five times our reference
1064: jet energy in a single pulse), $\rm\overline{\gamma}_{out}=1/5$,
1065: $\rm\overline{\theta}_S=1/2$, $\rm n=3$, and $\rm z=0.4315$.
1066:
1067: The comparison made in Fig.(\ref{712}) is rather unsatisfactory, in that
1068: the correlation between energy-interval and pulse-width is weaker
1069: in the observations than it is in the predictions. The theoretical
1070: correlation, for a thermal input spectrum, is roughly that implied
1071: by the simplified expression Eq.(\ref{simple}), that is
1072: $\rm \Delta t\;\Delta E\sim H$, in an obvious notation.
1073: The dashed blue curves in Fig.(\ref{712}) correspond to a modified input
1074: in which we have assumed that the CB cooling
1075: (as discussed in Section 8.5) may be not be linear in time,
1076: but closer to quadratic, so that
1077: $\rm \Delta t\propto 1/\sqrt{\Delta E}$. This modification
1078: goes in the right direction, but it is still not entirely
1079: satisfactory. We
1080: have not yet investigated in detail how a deviation from
1081: an input thermal spectrum at high energies --that we discussed
1082: in Section 8.1 in commenting Fig.(\ref{123})-- affects the
1083: time-energy correlation. But, since a non-thermal
1084: high-energy tail broadens the energy-distribution
1085: at all times, it ought to weaken
1086: even further the time-energy correlation, as required.
1087:
1088: We have studied the time-energy correlation for other single-pulse GRBs,
1089: such as 981022, 981221 and 990102. They all have a weaker
1090: energy-interval to pulse-width correlation than our model
1091: predicts, though the problem is most acute for GRB 990712,
1092: that we have thus chosen to expose the limitations of the CB model
1093: in its current formulation.
1094:
1095:
1096:
1097:
1098:
1099: \subsection{Lessons from the comparison with data}
1100:
1101: We conclude from our study of the general properties of GRBs in Section 7,
1102: and from the three comparisons with data in Section 8, that we may
1103: have deliniated the correct overall energetics of the collision
1104: of the CB with the SN shell, but our treatment of the time evolution
1105: of the processes of heating and cooling is oversimplified.
1106: A posteriori, there are many obvious reasons why this ought
1107: to be the case: the front of the CB is no doubt at a higher temperature
1108: than its bulk, since the CB is many collision-lengths long and
1109: is dominantly heated at the front: only muons and their decay
1110: electrons --but not photons from $\pi^0$ decay-- heat the bulk.
1111: The process is not a sudden heating followed by continuous
1112: cooling, as we assumed. We have included cooling by expansion,
1113: but not by emission from the CB's surface. We have assumed
1114: a constant expansion velocity, and not attempted to compute
1115: an actual expansion history from plasma dynamics. The CB
1116: may not have a constant density, it may even be a discontinuous
1117: ball of ``shrapnel''. Etc. etc.
1118:
1119: \subsection{An alternative simplified model}
1120:
1121: The six general properties of GRBs discussed in Section 7
1122: ought to be quite independent of the complex details
1123: of the CB's collision with the SN shell, since they only
1124: capitalize on the overall energetics and on the fact that,
1125: as it reaches the transparent outskirts of the SN shell,
1126: the CB is cooling by radiation and expansion.
1127: We illustrate this point by sketching an alternative model of
1128: CB heating and cooling, a simplified ``surface'' model that is in
1129: some sense the extreme opposite to the simplified ``volume''
1130: model we have discussed in detail.
1131: To lighten the discussion, in all numerical results
1132: in this chapter we fix the parameters to their reference values of Table I.
1133:
1134: In its rest frame, the front surface of the CB is bombarded by the nuclei
1135: of the SN shell, which have an
1136: energy $\rm m_p \,c^2\,\gamma\sim$ 1 TeV per nucleon,
1137: roughly 1/3 of which (from $\pi^0\to\gamma\gamma$ decays)
1138: is converted into these
1139: $\gamma$-rays within $\rm X_p\approx m_p/\sigma_{in}(pp)\approx 50\, g\,
1140: cm^{-2}$, where $\rm \sigma_{in}(pp)$ is the nucleon-nucleon
1141: inelastic cross section.
1142: These high energy photons initiate electromagnetic cascades that
1143: eventually convert their energy to thermal energy within the CB.
1144: The radiation length of high energy $\gamma$'s in hydrogenic plasma,
1145: dominated by $\rm e^+\,e^-$ pair production, is $\rm X_{\gamma e}
1146: \simeq 63$ g cm$^{-2}$,
1147: comparable to $\rm X_p$. The radiation length of thermalized
1148: photons in a hydrogenic plasma is
1149: $\rm X_\gamma^{ion}\approx m_p/\sigma_{_T}\approx 2.6$ g cm$^{-2}$.
1150:
1151: Assume that the quasi-thermal emission rate from
1152: the CB, within $\rm X_\gamma^{ion}$ from its
1153: surface, is in dynamical equilibrium with the fraction of energy deposited
1154: by the CB's collision with the SN shell in that outer layer.
1155: The temperature of the CB's front is then
1156: roughly given by:
1157: \begin{equation}
1158: \rm T(x)\simeq \left[{(n\! -\! 1)\,X_\gamma\, m_p\, c^3 [\gamma(x)]^2\,
1159: \sigma_{in}(pp) \over
1160: 6\,\sigma\, x_{tp}\,X_{\gamma e}
1161: \, \sigma_{_T}^2}\right ]^{1\over 4}
1162: \left[ {x\over x_{tp}}\right]^{-{n\over 4}}\!\!\! ,
1163: \end{equation}
1164: where $\rm\gamma(x)$ is a function that decreases
1165: monotonically from $\rm \gamma_{in}$
1166: to $\rm\gamma_{out}$.
1167: Remarkably, only the Lorentz factor of the CBs, but neither their mass
1168: nor their energy, appear in the above expression,
1169: except for the fact that, for the result to be correct, they must be
1170: large enough for the CB to pierce the SN shell and remain relativistic.
1171:
1172:
1173: For $\rm n=8$ the value of
1174: $\rm x_{tp}$ is $\rm \approx 3 \,R_S$, and, for t close
1175: to $\rm t_{tp}$ or later:
1176: \begin{equation}
1177: \rm T(t)\simeq 0.16\, keV\,
1178: \left[ {t_{tp}\over t}\right]^{2}\,\left[ {\gamma(t)\over
1179: 10^3}\right]^{1\over 2}\; .
1180: \label{newtemp}
1181: \end{equation}
1182: At the time of transparency this is quite comparable to the result of combining
1183: Eqs.(\ref{Ttrans1}) and (\ref{timedeps}). However, only for $\rm n=4$
1184: does the temperature decrease approximately
1185: as 1/t. For n $>4$ it diminishes faster than 1/t and for $\rm n=8$
1186: it decreases faster than
1187: $\rm 1/t^2$, the ``faster'' being due, in both cases, to the effect
1188: of a decreasing $\rm \gamma(t)$. For an exact $\rm 1/t^2$ behaviour
1189: the pulse width narrows with time as $\rm \Delta t \propto E^{-0.5}$ and,
1190: as we have also seen in
1191: Section 8.3, this goes in the direction of improving the
1192: predicted time-energy correlation. In fact, Fenimore et al.~(1995)
1193: found, from a large sample of GRB pulses, that
1194: $\rm \Delta t \propto E^{-0.46}$.
1195:
1196:
1197: The total radiated energy, in the CB rest frame, is roughly the thermal
1198: energy deposition within one radiation length from its
1199: front surface. After attenuation in the SN shell, it reduces to:
1200: \begin{equation}
1201: \rm E_{pulse}^{rest}\approx {\sigma_{in}(pp)\, \pi\,
1202: [R_{CB}^{tp}]^2\,\bar X_\gamma\, m_p\,c^2\, \gamma(t)
1203: \over 3\,X_{\gamma e}\, \sigma_{_T}^2}\,,
1204: \label{newenergy}
1205: \end{equation}
1206: where $\rm \bar X_\gamma$ is the radiation length
1207: in the obscuring shell averaged over the black body spectrum.
1208: For a typical $\gamma$-ray
1209: peak energy of $\rm E_p\sim 1\, MeV$ in the SN rest frame,
1210: $\rm \bar X_\gamma\simeq 10\ g\, cm^{-2}\,. $ Consequently,
1211: the CB's radius at transparency is
1212: $\rm R_{CB}^{tp}=4\times 10^{11}$ cm and
1213: $\rm E_{pulse}^{rest}\sim 3\times 10^{45}\, erg$,
1214: for $\rm\gamma(t)\sim 10^3$. This is consistent with the results in
1215: Eqs.(\ref{EGRB1}), implying that the predicted fluences in the
1216: surface-heating and volume-heating models are quite similar.
1217: The fact that the characteristic
1218: temperatures of the volume-heating and the surface-heating
1219: models around the time of transparency
1220: are also similar means that their predicted GRB
1221: individual-photon energies are comparable and both in agreement
1222: with the GRB observations.
1223:
1224: \section{Conclusions}
1225:
1226: In a previous paper (Dar and De R\'ujula 2000a) we have argued
1227: that the CB model provides a very good description of GRB
1228: afterglows, including those whose light curve is seen to
1229: rise before it drops, as is the case for GRB 970508. There
1230: we also contended that, in the case of GRB 980425, the model
1231: provides a strong motivation for the search of the superluminal
1232: motion of the afterglow-emitting CB, relative to the associated
1233: supernova: SN1998bw. This would be a decisive signature
1234: for highly relativistic cannonballs, as opposed to conically
1235: spreading jets. We plan to discuss in future work other important signatures
1236: of the CB model: high energy neutrinos and photons during the GRB,
1237: flare up and X-ray lines in its early afterglow.
1238:
1239: In this paper we have demonstrated that the CB model
1240: explains the fluence and energy spectrum of GRBs,
1241: as well as the characteristic properties of their light curves.
1242: The detailed heating, expanding and
1243: cooling of the CB ---as it hits and sweeps up the SN shell---
1244: we have treated only in a simplified fashion. As a consequence,
1245: the model in its present form does not provide a completely
1246: satisfactory quantitative description of the time-energy correlation.
1247: We may not have completely untied the perduring Gordian
1248: knot of the GRB conundrum, but we have argued that
1249: we have sliced it open.
1250: \vspace{.5cm}
1251:
1252: \noindent
1253: {\bf Aknowledgements}
1254: We are indebted to Rainer Plaga for interesting discussions.
1255: This work was supported in part by the Fund for Promotion Of Research At
1256: The Technion.
1257:
1258: \vskip 1 true cm
1259: %\vskip 0.2 true cm
1260: \begin{table}[h]
1261: %\vskip 0.1 true cm
1262: %\hspace{-.5cm} %if you want to center your table act on this argument
1263: \hspace{0.4 cm}
1264: \begin{tabular}{|l|c|c|c|}
1265: \hline
1266: \hline
1267: $\;\;\;\;\;\;\;\;\;$Parameter &Symbol &Value \\
1268: \hline
1269: SN-shell's mass & $\rm M_S$ & $\rm 10\; M_\odot$ \\
1270: SN-shell's radius & $\rm R_S$ & $2.6\times 10^{14}$ cm \\
1271: SN-shell's density index &n &8 \\
1272: Outgoing Lorentz factor & $\rm\gamma_{out}$ & $10^3$ \\
1273: CB's viewing angle & $\theta$ & $\rm 10^{-3}$ \\
1274: CB's energy & $\rm E_{CB}$ & $10^{52}$ erg \\
1275: Initial $\rm v_{_T}/c$ of expansion & $\rm\beta_{in}$ & $1/(10\,\sqrt{3})$ \\
1276: Final $\rm v_{_T}/c$ of expansion & $\rm\beta_{out}$ & $1/\sqrt{3}$ \\
1277: \hline
1278: Redshift &z &1 \\
1279: \hline
1280: \hline
1281: \end{tabular}
1282: \end{table}
1283: \vskip -0.3 true cm
1284: \noindent
1285: {\bf Table I.}
1286: List of the ``reference'' values of various parameters. In the text a barred
1287: parameter means its actual value divided by its reference value, so that,
1288: for instance, $\rm \overline M_S=1/2$ means the actual mass of the SN shell
1289: is taken to be $\rm 5\; M_\odot$.
1290:
1291: \begin{thebibliography}{}
1292:
1293:
1294:
1295:
1296:
1297: \bibitem{}
1298: Belloni T., et al., 1997, ApJ 479, 145
1299: \bibitem{}
1300: Blandford R.D., Znajek R.L., 1977, MNRAS 179, 433
1301: \bibitem{}
1302: Bridle A. 2000, http://www.cv.nrao.edu/$\sim$abridle/
1303: \newline radiogal.htm
1304: \bibitem{}
1305: Briggs M.S., et al., 1999, ApJ 524, 82
1306: \bibitem{}
1307: Cen R., 1999, ApJ 524, 51
1308: \bibitem{}
1309: Celotti A., et al. 1997 MNRAS 286, 415
1310: \bibitem{}
1311: Costa E., et al., 1997, Nature 387, 783
1312: \bibitem{}
1313: Dar A., 1998, ApJ 500, L93
1314: \bibitem {}
1315: Dar A., 1999, A\&A 138S, 505
1316: \bibitem{}
1317: Dar A., De R\'ujula, A., 2000a, A\&A accepted (astro-ph/0008474)
1318: \bibitem{}
1319: Dar A., De R\'ujula, A., 2000b, to be submitted
1320: \bibitem{}
1321: Dar A., De R\'ujula, A., 2000c, to be submitted
1322: \bibitem{}
1323: Dar A., De R\'ujula, A., 2000d, astro-ph/0002014, submitted to MNRAS
1324: \bibitem{}
1325: Dar A., Plaga R., 1999, A\&A 349, 259
1326: \bibitem{}
1327: De R\'ujula A., 1987, Phys. Lett. 193, 514
1328: \bibitem{}
1329: Dyer K.K., et al., 2000, astro-ph/0011578
1330: \bibitem{}
1331: Fenimore E.E., Ramirez-Ruiz E., 2000, astro-ph/0004176
1332: \bibitem{}
1333: Fenimore E.E., et al., 1995, ApJ 448, L101
1334: \bibitem{}
1335: Fusco-Femiano R., et al., 1999, ApJ, 513, L21
1336: \bibitem{}
1337: Fusco-Femiano R., et al., 2000, astro-ph/0003141
1338: \bibitem{}
1339: Galama T.J., et al., 1998, Nature 395, 670
1340: \bibitem{}
1341: Ghisellini G., 2000, astro-ph/0012125
1342: \bibitem{}
1343: Goodman J., Dar, A., Nussinov, S., 1987, ApJ 314, L7
1344: \bibitem{}
1345: Groom D.E., et al., 2000, {\it Review of Particle Physics}, Eur. Phys. J.
1346: C15, 1
1347: \bibitem{}
1348: Kippen R.M., et al., 1998, GCN 67
1349: \bibitem{}
1350: Kippen R.M., 2000,
1351: http://www.batse.msfc.nasa.gov/
1352: \newline $\sim$kippen/batserbr/brbr$_{-}$obs.html
1353: \bibitem{}
1354: Lyne A.G., Lorimer, D.R., 1994, Nature 369, 127
1355: \bibitem{}
1356: Metzger M.R., et al., 1997, Nature 387, 878
1357: \bibitem{}
1358: MacFadyen A.I., Woosley S.E., 1999, ApJ 524, 168
1359: \bibitem{}
1360: Mallozzi, R.S., 2000,
1361: http://www.batse.msfc.nasa.gov/batse/
1362: \bibitem{}
1363: Margon B.A., 1984, ARA\&A 22, 507
1364: \bibitem{}
1365: Meszaros P., Rees M.J., 1992, MNRAS 257, 29
1366: \bibitem{}
1367: Meszaros P., Rees M.J., 1997, ApJ 482, L29
1368: \bibitem{}
1369: Mirabel I.F., Rodriguez, L.F., 1994, Nature 371, 46
1370: \bibitem{}
1371: Mirabel, I.F., Rodriguez, L.F., 1999a, ARA\&A 37, 409
1372: \bibitem{}
1373: Mirabel, I.F., Rodriguez, L.F. 1999b, astro-ph/9902062
1374: \bibitem{}
1375: Norris J.P., et al., 1999, astro-ph/9903233
1376: \bibitem{}
1377: Nakamura T. et al., 2000, astro-ph/0007010
1378: \bibitem{}
1379: Paciesas W.S., et al., 1999, ApJS 122, 465)
1380: \bibitem{}
1381: Paczynski B., 1986, ApJ 308, L43
1382: \bibitem{}
1383: Paczynski B., 1998, ApJ 494, L45
1384: \bibitem{}
1385: Plaga R., 2000, astro-ph/001206
1386: \bibitem{}
1387: Preece R.D., 2000, ApJS 126, 19
1388: \bibitem{}
1389: Preece R.D., et al., 1998, ApJ 496, 849
1390: \bibitem{}
1391: Rephaeli Y., et al., 1999, ApJ 511 L21
1392: \bibitem{}
1393: Rodriguez L.F., Mirabel, I.F., 1999, ApJ 511, 398
1394: \bibitem{}
1395: Shaviv N.J., Dar A., 1995, ApJ 447, 863
1396: \bibitem{}
1397: Tingay S.J., et al., 1995, Nature 374, 141
1398: \bibitem{}
1399: van Paradijs J., et al., 1997, Nature 386, 686
1400: \bibitem{}
1401: Wilson A.S., et al., 2000, astro-ph/0008467
1402: \bibitem{}
1403: Woosley S.E., 1993, ApJ, 405, 273
1404: \bibitem{}
1405: Woosley S.E., 1999, in the {\it
1406: Fifth Huntsville Conference on Gamma-Ray
1407: Bursts}, eds. R. M. Kippen, R.S. Mallozzi, \& V. Connaughton, AIP
1408: \bibitem{}
1409: Woosley S.E., MacFadyen, A.I., 1999, A\&AS 138, 499
1410: \bibitem{}
1411: Wu B., Fenimore L., 2000, ApJ 535, L29
1412: \bibitem{}
1413: Yi I., 1994, ApJ 431, 543
1414:
1415: \end{thebibliography}{}
1416:
1417:
1418:
1419: %%%%%%%%%%%%%%%%%%%%%%%%%%%%1
1420: \begin{figure}
1421: \begin{center}
1422: \vspace*{1.0cm}
1423: \hspace*{-.5cm}
1424: \epsfig{file=fluenceall.ps,width=9cm}
1425: %\vspace*{-14.6cm}
1426: \caption{Fluence as a function of various parameters,
1427: in units of $10^{-5}$ erg/cm$^2$. The parameters
1428: $\theta$ through $\rm\gamma_{out}$ are in units
1429: of their reference values of Table I, thus the barred notation.
1430: The index n and the redshift z are not rescaled.
1431: All parameters not being varied are fixed at their
1432: reference values of Table I, but for $\theta$, fixed
1433: at a ``typical'' $\rm \overline{\theta}=3.$}
1434: \vspace*{-0.5cm}
1435: \label{flu}
1436: \end{center}
1437: \end{figure}
1438: %%%%%%%%%%%%%%%%%%%%%%%%%%%%2
1439: \begin{figure}
1440: \begin{center}
1441: \vspace*{1.0cm}
1442: \hspace*{-1cm}
1443: \epsfig{file=dndEdtfixedE.ps,width=7cm}
1444: %\vspace*{-14.6cm}
1445: \caption{GRB-pulse shape as a function of time, at various
1446: fixed $\gamma$-energies.}
1447: \vspace*{-0.5cm}
1448: \label{t3Es}
1449: \end{center}
1450: \end{figure}
1451: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1452:
1453: %%%%%%%%%%%%%%%%%%%%%%%%%%%%3
1454: \begin{figure}
1455: \begin{center}
1456: \vspace*{1.0cm}
1457: \hspace*{-1cm}
1458: \epsfig{file=abc.ps,width=7cm}
1459: %\vspace*{-14.6cm}
1460: \caption{GRB $\gamma$-energy distributions,
1461: at various fixed times, multiples of the (observer's)
1462: time at which the SN-shell becomes transparent.}
1463: \vspace*{-0.5cm}
1464: \label{E3ts}
1465: \end{center}
1466: \end{figure}
1467: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1468:
1469: %%%%%%%%%%%%%%%%%%%%%%%%%%%%4
1470: \begin{figure}
1471: \begin{center}
1472: \vspace*{1.0cm}
1473: \hspace*{-1cm}
1474: \epsfig{file=dnde.ps,width=7cm}
1475: %\vspace*{-14.6cm}
1476: \caption{Time-integrated $\rm E^2\,dN/dE$ distributions,
1477: illustrating the sensitivity to one parameter at a time.
1478: The absolute vertical scale is arbitrary, but the relative scales
1479: are not. All parameters not mentioned in each subfigure are
1480: kept at the reference values of Table I
1481: (but for $\rm\beta_{in}$, which is fixed at $1/\sqrt{3}$).
1482: Notice that the shape of the energy
1483: distribution is always the same, irrespective of the
1484: parameter values.}
1485: \vspace*{-0.5cm}
1486: \label{E}
1487: \end{center}
1488: \end{figure}
1489: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1490: %%%%%%%%%%%%%%%%%%%%%%%%%%%%5
1491: \newpage
1492: \begin{figure}
1493: \begin{center}
1494: \vspace*{1.0cm}
1495: \hspace*{-.5cm}
1496: \epsfig{file=dndtallnew.ps,width=9cm}
1497: %\vspace*{-14.6cm}
1498: \caption{Time distribution $\rm dN/dt$ of a single GRB pulse,
1499: integrated for all $\rm E>30$ keV. The absolute vertical scale is
1500: arbitrary, but the relative scales are not. All parameters not mentioned
1501: in each subfigure are kept at the reference values of Table I
1502: (but for $\rm\beta_{in}$, which is fixed at $1/\sqrt{3}$).
1503: For ease of comparison, in some subfigures, a curve has been rescaled,
1504: e.g. in (b) the result for an input $\rm E_{CB}/10$ has been multiplied by 20.}
1505: \vspace*{-0.5cm}
1506: \label{t}
1507: \end{center}
1508: \end{figure}
1509: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1510:
1511: %%%%%%%%%%%%%%%%%%%%%%%%%%%%6
1512: \newpage
1513: \begin{figure}
1514: \begin{center}
1515: \vspace*{1.0cm}
1516: \hspace*{-1cm}
1517: \epsfig{file=dndtdndEz.ps,width=7cm}
1518: %\vspace*{-14.6cm}
1519: \caption{Illustration of the sensitivity to redshift, z, of the
1520: energy-and time-distributions in a GRB pulse.}
1521: \vspace*{-0.5cm}
1522: \label{z}
1523: \end{center}
1524: \end{figure}
1525: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1526: %%%%%%%%%%%%%%%%%%%%%%%%%%%%7
1527: \begin{figure}
1528: \begin{center}
1529: \vspace*{1.0cm}
1530: \hspace*{-1cm}
1531: \epsfig{file=dndt2.ps,width=7cm}
1532: %\vspace*{-14.6cm}
1533: \caption{Examples of parameter values that give rise to
1534: non-FRED pulse shapes. The individual vertical
1535: scales are chosen for ease on comparison.
1536: All parameters not mentioned are
1537: as in Table I.}
1538: \vspace*{-0.5cm}
1539: \label{t2}
1540: \end{center}
1541: \end{figure}
1542: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1543: %%%%%%%%%%%%%%%%%%%%%%%%%%%%8
1544: \newpage
1545: \begin{figure}
1546: \begin{center}
1547: \vspace*{1.0cm}
1548: \hspace*{-1cm}
1549: \epsfig{file=dndt6CBall.ps,width=9cm}
1550: %\vspace*{-14.6cm}
1551: \caption{ ``Synthetic'' GRB light curves, generated by shooting
1552: six CBs at random in a 1.5 s time-interval, and with random values
1553: of $\rm E_{CB}$ within a factor of three of our reference value.
1554: The only difference between (a) and (b) is that $\rm n = 8$ in (a),
1555: while $\rm n = 4$ in (b). All other parameters in this figure have their
1556: reference values. The figure
1557: illustrates how a CB produces a GRB pulse, but a
1558: GRB-pulse may not correspond to a single CB.}
1559: \vspace*{-0.5cm}
1560: \label{6CB}
1561: \end{center}
1562: \end{figure}
1563: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1564:
1565: %%%%%%%%%%%%%%%%%%%%%%%%%%%%9
1566: \begin{figure}
1567: \begin{center}
1568: \vspace*{1.5cm}
1569: \hspace*{-1cm}
1570: \epsfig{file=990123Efit.ps,width=9cm}
1571: %\vspace*{-14.6cm}
1572: \caption{Comparison of theory and observation
1573: for the time-integrated energy distributions $\rm dN/dE$
1574: and $\rm E^2\,dN/dE$, in the case of GRB 990123.
1575: Notice that many experimental points at the
1576: higher energies are only upper limits.}
1577: \vspace*{-0.5cm}
1578: \label{123}
1579: \end{center}
1580: \end{figure}
1581: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1582: %%%%%%%%%%%%%%%%%%%%%%%%%%%%10
1583: \begin{figure}
1584: \begin{center}
1585: \vspace*{1.0cm}
1586: \hspace*{-1cm}
1587: \epsfig{file=grb980425lcfit.ps,width=8cm}
1588: %\vspace*{-14.6cm}
1589: \caption{Comparison of theory and observation for the
1590: light curve of GRB 980425, in the 50-300 keV energy interval. }
1591: \vspace*{-0.5cm}
1592: \label{425}
1593: \end{center}
1594: \end{figure}
1595: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1596: %%%%%%%%%%%%%%%%%%%%%%%%%%%%11
1597: \begin{figure}
1598: \begin{center}
1599: \vspace*{1.0cm}
1600: \hspace*{-.5cm}
1601: \epsfig{file=7648.ps,width=9cm}
1602: %\vspace*{-14.6cm}
1603: \caption{Comparison of theory and observations for the
1604: light curve of GRB 980712, in various BATSE energy intervals. }
1605: \vspace*{-0.5cm}
1606: \label{712}
1607: \end{center}
1608: \end{figure}
1609: %%%%%%%%%%%%%%%%%%%%%%%%%%%%
1610:
1611: \end{document}
1612:
1613:
1614:
1615:
1616:
1617:
1618:
1619:
1620:
1621:
1622:
1623:
1624:
1625:
1626:
1627: