astro-ph0307058/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: \newcommand{\vdag}{(v)^\dagger}
3: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
4: \slugcomment{To appear in {\em ApJ Lett.}}
5: \begin{document}
6: %\title{A beaming factor of 450 in GRBs with standard energies}
7: \title{Observational evidence for a correlation between 
8: peak-luminosities and beaming in GRBs}
9: \author{Maurice H.P.M. van Putten\altaffilmark{1} and Tania Regimbau\altaffilmark{1}}
10: \affil{LIGO Laboratory, NW 17-161, 175 Albany Street, Cambridge, MA 02139}
11: \begin{abstract}
12: We calculate the unseen-but-true GRB-event rate from a current flux-limited sample of 33 
13: GRBs with individually measured redshifts. We consider a GRB-event rate which is proportional
14: to the star-formation rate, in view of the GRB-SNe association in SN1998bw and GRB030329.
15: By fitting a log-normal distribution of GRB peak-luminosities, 
16: we find a ratio $1/f_r\simeq450$ 
17: for the true-to-observed GRB-event rates. This provides an independent derivation 
18: of the GRB-beaming factor $1/f_b\simeq500$ obtained by Frail et al. (2001) from 
19: sources with standard GRB-energies. We discuss applications to GRB980425.
20: \end{abstract}
21: \keywords{gamma-rays: bursts, beaming}
22: \section{Introduction}
23: New clues are emerging that long GRB are associated with supernovae 
24: \citep{gal98,blo99,kul00,rei01}. This is illustrated most recently 
25: by optical emissions lines in the late-time light-curve of the HETE-II burst 
26: GRB 030329 \citep{sta03}, which are remarkably similar to those observed
27: in GRB980425/SN1998bw \citep{gal98}. A GRB-supernova assocation provides important
28: support for the collapsar model of GRBs, representing a violent death
29: of evolved massive stars \citep{woo93,pac98}. The short lifespan of tens of Myrs
30: of massive stars implies that GRBs take place in star-forming regions
31: \citep{pac98,fru99}, and hence more broadly points towards an association
32: to molecular clouds. The event rate of GRBs per unit cosmological volume 
33: is hereby expected to be correlated to the cosmic star-formation rate
34: (e.g., \citet{bla00,ber03,cho02}). Because 
35: observations by past and current experiments (see Table I) 
36: are flux-limited, the observed GRB-event rate is strongly biased towards 
37: events at lower redshifts.
38: 
39: In this letter, we calculate the ratio $1/f_r$ of unseen-to-observed GRBs.
40: Our analysis is based on the bias in the observed GRB-event rate towards 
41: low redshift events in a flux-limited sample of 33 GRBs with individually
42: measured redshifts. We propose to use a linear relationship for the correlation
43: between the true-but-unseen GRB event rate and the cosmic SFR, to derive from
44: this bias a best-fit log-normal GRB peak-luminosity function. While it is an 
45: open question to what the degree cosmological evolution of metallicity 
46: affects the SFR independently of the GRB event rate (see, e.g., \cite{heg02}),
47: linear relationship between the two will serve as a leading-order approximation.
48: 
49: In \S2, we discuss the cosmological model. \S3 tabulates the current sample
50: of 33 GRBs with individually measured redshifts. In \S4, we present the
51: best fit of the data for a log-normal distribution function, estimate
52: the fraction of unseen-to-observed GRBs, and compare our results with
53: the beaming factor $1/f_b=500$ of Frail et al.(2001), 
54: based on a sample of GRBs with achromatic breaks
55: in their light-curves. We summarize our results and draw our main conclusions 
56: in \S5. 
57: 
58: \section{The cosmological model}
59: 
60: We model the intrinsic GRB-event rate in a flat $\Lambda$-dominated cold
61: dark matter cosmology with closure energy densities $\Omega_\Lambda$ = 0.70
62: and $\Omega_m$ = 0.30. These values are suggested by BOOMERANG and MAXIMA 
63: \citep{deb00,han00} on the power spectra of the CMB and distant Type Ia SNe
64: \citep{per99,sch98}. The Hubble parameter H$_o$ is taken to be 
65: $73\,\mathrm{km\,s}^{-1}\mathrm{Mpc}^{-1}$ (Freedman et al., 2001).
66: 
67: \citet{por01} provide three models of the cosmic SFR up to redshifts $z\sim5$,
68: reflecting some uncertainties in SFR estimates.
69: No significant changes have been noticed in application of these three models,
70: and we present only results derived from the second model, SFR2. We have 
71: \begin{equation}
72: R_{c}(z) = 
73: R_{SF2}(z)
74: \frac{E(\Omega_i,z)}{(1+z)^{3/2}}\; 
75: \mathrm{M}_{\odot}\,\mathrm{yr}^{-1}\,\mathrm{Mpc}^{-3}
76: \end{equation}
77: by transformation of the $R_{SF2}$ in a matter-dominated
78: unverse ($\Omega_{m}=1$), 
79: \begin{equation}
80: R_{SF2}(z) = \frac{0.15\,e^{3.4z}}{(22 + e^{3.4z})} \; \mathrm{M}_{\odot}\,\mathrm{yr}^{-1}\,\mathrm{Mpc}^{-3},
81: \end{equation}
82: where $E(\Omega_i,z) =[(1+z)^2(1+z\Omega_m)-z(2+z)\Omega_v]^{1/2}$.
83: 
84: For a GRB-event rate locked to the SFR, 
85: the true event rate between $z$ and $z+dz$, as observed in the observer's frame 
86: of reference, satisfies
87: \begin{equation}
88: dR_{GRB}(z)=\lambda _{GRB}\frac{R_{c}(z)}{1+z}{\frac{{dV}}{{dz}}}\,dz
89: \end{equation}
90: where $\lambda_{GRB}$ is the formation mass fraction of the source
91: progenitors. Here, the division by $1+z$ 
92: accounts for time-dilatation by cosmic expansion.
93: The element of comoving volume is
94: \begin{equation}
95: dV = \frac{4\pi r^{2}c}{H_0E(\Omega_i,z)}\,dz,~~~
96: r (z) = \int_{0}^{z} \frac{c}{H_0E(\Omega_i,z')}\,dz'
97: \end{equation}
98: The GRB redshift probability density can be written as 
99: \citep{cow02}
100: %\begin{equation}
101: $p (z) =  {dR_{GRB}/dz}\left({\int_{0}^{5}dR_{GRB}/dz\,dz}\right)^{-1}.$
102: %\end{equation}
103: Assuming that the mass fraction of GRBs progenitors is redshift independent, 
104: the scaling factor $\lambda _{GRB}$ is the only free parameter of our model.
105: For the flux-limited experiments listed in Table I
106: we define a probability-density function of detection as a function of
107: redshift as
108: %\begin{equation}
109: $p_{detect} (z) ={dR_{detect}/dz}\left({\int_{0}^{5}dR_{detect}/dz\,dz}\right)^{-1},$
110: %\end{equation}
111: where the dependence on the luminosity has been integrated out in the
112: detected event rate \citep{bro02}
113: \begin {equation}
114: dR_{detect} = dR_{GRB}(z) \int_{L_{lim}(z)} p(L)\,dL
115: \end {equation}
116: Here, $p(L)$ refers to the intrinsic GRBs luminosity function
117: in the BATSE energy range $50 - 300$ keV. 
118: 
119: The luminosity threshold as a function of redshift is given by
120: %\begin {equation}
121: $L_{lim} (z) = 4\pi d_{L}^{2}(z) S_{lim}$,
122: %\end {equation}
123: where $d_{L}$ is  the luminosity distance to a source at redshift $z$ and
124: where $S_{lim}$ denotes the sensitivity threshold of the
125: instrument.  Following \citep{bro02}, we take a flux-density
126: threshold of BASTE of 0.2 $\mathrm{photon}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}$.
127: 
128: \section{A redshift sample of 33 GRBs}
129: 
130: As of the time of this writing, we have 33 GRBs with
131: individually measured redshifts (Table I). They collectively
132: represent a variety of past and current experiments.
133: Due to a flux threshold in each of the instruments used, 
134: the observed GRB redshift distribution is strongly biased
135: towards low redshifts. This introduces a quantitative
136: selection effect, relative to the redshift distribution 
137: predicted by the SFR model (Fig. 1) -- the true distribution 
138: of GRB-redshifts representing what would be observed in the 
139: ideal case of a zero-flux threshold in the instruments.
140: In view of the various instruments involved, the dependency of
141: the observed GRB redhift-distribution function on the flux-limit
142: is simulated in Fig. 2.
143: 
144: %\section{The GRB-luminosity function}
145: %We set out to derive the intrinsic GRBs luminosity function $p(L)$
146: %by comparing the true GRB-redshift distribution locked to the SFR 
147: %and the observed GRB-redshift distribution, listed in Table I.
148: 
149: \section{A fit to a log-normal distribution function}
150: 
151: We shall assume that the GRB-luminosity function is redshift independent,
152: i.e., without cosmological evolution of the nature of its progenitors.
153: We take a log-normal probability density for the luminosity shape-function,
154: with mean $\mu$ and width $\sigma$ given by
155: \begin{equation}
156: p(L) = \frac{1}{(2\pi)^{1/2} \sigma L}\exp
157: \left(\frac{-(\ln L-\mu)^{2}}{2\sigma^{2}}\right),
158: \label{EQN_LNP}
159: \end{equation}
160: where $L$ is normalized with respect to  1cm$^{-2}$s$^{-1}$.
161: The optimized parameters of our model are
162: \begin{eqnarray}
163: (\mu , \sigma) = (124 , 3)
164: \pm (2 , -0.4).
165: \label{EQN_PAR}
166: \end{eqnarray}
167: This notation means that the estimated parameters can be either 
168: (122,3.4), (123,3.2), (125,2.8), or (126,2.6), but not (122,2.6)
169: for instance. Our results are compatible to the expectations
170: of Sethi et Bhargavi (2001) who derive a log-normal luminosity
171: function with $\mu = 129$ and $\sigma = 2$ from a different flux-limited
172: sample. They mentioned that because of selection effects, the
173: inferred average luminosity over-estimates the true mean luminosity 
174: by a factor of 2 or 3, and the variance is within 40-50\% of the true
175: variance. The observed and predicted redshift distribution, based on the SFR,
176: are shown in Fig. 1 in case of optimal parameters (\ref{EQN_PAR}).
177: The results indicate a good fit to the data, suggesting that selection effects 
178: are adequaly modeled. The fraction of 
179: {\em detectable} GRBs as a function of redshift
180: \begin{equation}
181: F(z) = \frac{dR_{detect}}{dR_{GRB}(z)} =  \int_{L_{lim}(z)} p(L)\,dl .
182: \end{equation}
183: shows a the steep decrease as the luminosity threshold
184: increases, making high-redshift GRBs less likely to be detected.  
185: 
186: A satisfactory test for our model is provided by 
187: comparing the predicted flux distribution
188: of GRBs above the sensitivity threshold with the peak flux distribution of
189: a sample of 67 GRBs observed with IPN (Fig. 3).
190: The fluxes derived from our luminosity function in 50-300 keV have been 
191: extrapolated to the IPN range of 25-100 keV, assuming an $\mathrm{E}^{-2}$ 
192: energy spectrum and using the formula given in Appendix B of
193: Sethi et Bhargavi (2001). 
194: The conversion factor from $\mathrm{erg}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}$ to
195: $\mathrm{photon}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}$ has been taken
196: to be $0.87 \times 10^{-7}$, and the sensitivity
197: threshold equal to 
198: 5 $\mathrm{photon}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}$ (Hurley, private
199: communication).
200: 
201: A quantitative result provided by our model is the ratio $1/f_r$
202: of the true GRB-event rate (what would be seen in case of a zero-flux limit
203: in the detector) and the observed GRB-event rate. For the optimal
204: parameters (\ref{EQN_PAR}), we have
205: \begin{eqnarray}
206: 1/f_r = 450.
207: \label{EQN_R}
208: \end{eqnarray}
209: The factor $1/f_r$ is between 200-1200 in the error box of (\ref{EQN_PAR}).
210: 
211: \section{A correlation between $1/f_r$ and $1/f_b$}
212: 
213: Our estimate of the true-to-observed GRB-event rate $1/f_r$ is strikingly 
214: similar to the GRB-beaming factor $1/f_b$ of about 500 derived by Frail et al. 
215: (2001).
216: Our analysis is independent of the mechanism providing a broad distribution 
217: in GRB luminosities. Without further input, our results may reflect 
218: (a) isotropic sources with greatly varying energy output, 
219: (b) beamed sources with standard energy output and varying opening angles, or 
220: (c) anisotropic but geometrically standard sources \citep{ros02,zha02}.
221: 
222: We find that the GRB peak-luminosities and beaming are correlated. 
223: To see this, we simply note that the case of no correlation between 
224: peak-luminosities and beaming give rise to an unseen-but-true GRB event rate 
225: which is $1/f_r\times 1/f_b\simeq 2.5\times10^5$ times the observed rate. The 
226: true GRB-event rate hereby approaches that of Type II supernovae -- we discard 
227: this possibility.  A correlation between peak-luminosities and beaming is 
228: naturally expected sources (b) and (c) with standard energy output -- the 
229: picture that bears out of \citet{fra01}. This introduces a correlation 
230: $L\propto 1/f_b$ between peak-luminosity and beaming factor in (b) and, for 
231: a flux-limited sample, also in (c). For a flux-limited sample, both (b) and
232: (c) give rise to an anticorrelation between inferred beaming and distance 
233: such that to leading-order $\theta_jz\sim$const. Fig. 4 shows that this 
234: anticorrelation holds in the sample of \citet{fra01}. 
235: For a related discussion on estimating the GRB beaming factor from
236: flux limited surveys, see \citet{lev02}.
237: 
238: \section{Conclusions}
239: 
240: A table of 33 GRBs with individually determined
241: redshifts allows an estimate of the GRB-luminosity function, based
242: on constant of proportionality between the GRB event rate and the 
243: cosmic star-formation rate. We have tested our fit by reproducing the 
244: distribution of peak luminosities in the IPN sample of 67 GRBs.
245: 
246: A flux-limited sample introduces a ratio $1/f_r$ of unseen GRBs,
247: whose emissions fall below the detector threshold, to observed GRBs. 
248: A best fit analysis of the luminosity function gives $1/f_r\simeq450$. 
249: This number is both large and close to the GRB beaming factor of 500. 
250: The possibility that $1/f_r$ and $1/f_b$ are uncorrelated 
251: gives rise to prohibitively large true GRB event rates, and is discarded.
252: The results support a relation $L\propto 1/f_b$ between the peak-luminosity 
253: and the beaming factor in the context of conical or anisotropic GRB-emissions 
254: with standard GRB-energies. This points towards an anticorrelation between 
255: distance and half-opening angle, which is approximately supported by the 
256: sample of \citet{fra01}.
257: 
258: The results presented here and those of \citet{fra01} indicate a true
259: GRB event rate of 1 per year within $D=100$Mpc. 
260: GRB980425/SN1998bw ($z=0.0085$) is consistent with this true
261: event rate. For anisotropic but geometrically standard sources, 
262: GRB980425 is hereby {\em not} anomalous, but 
263: consistent with the trend shown in Fig. 4.
264: 
265: \acknowledgments
266: 
267: The authors greatfully acknowledge input from Kevin Hurley. 
268: MVP thanks R. Mochkovitch for discussions on GRB980425.
269: This research is
270: supported by the LIGO Observatories, constructed by Caltech and MIT
271: with funding from NSF under cooperative agreement PHY 9210038.
272: The LIGO Laboratory operates under cooperative agreement
273: PHY-0107417. This paper has been assigned LIGO document number
274: LIGO-P030022-00-D.
275: 
276: \begin{thebibliography}{}
277: \bibitem[Berger et al.(2003)]{ber03} 
278: Berger, E., et al., 2003, ApJ, 588, 99
279: \bibitem[Blain \& Natarajan(2000)]{bla00} 
280: Blain, A.W., \& Natarajan, P., 2000, MNRAS, 312, L39
281: \bibitem[Bloom et al.(1999)]{blo99} 
282: Bloom, J.S., et al, 1999, Nature, 401, 453
283: \bibitem[Bromm \& Loeb(2002)]{bro02}
284: Bromm, J.S \& Loeb, A., 2002, ApJ, 575, 111
285: \bibitem[Choudhury \& Srianand(2002)]{cho02} 
286: Choudhury, T.R., \& Srianand, R., 2002, MNRAS, 336, L27
287: \bibitem[Coward et al.(2002)]{cow02}
288: Coward, D.M., Burman, R.R., Blair, D.G., 2002, MNRAS, 329, 411 
289: \bibitem[de Bernardis et al.(2000)]{deb00}
290: de Bernardis, P., et al, 2000, Nature 404, 995
291: \bibitem[Frail et al.(2001)]{fra01} 
292: Frail, D.A., et al., 2001, ApJ, 562, L55
293: \bibitem[Freedman et al. (2001)]{free01}
294: Freedman W., et al., 2001,  Astrophys. J. 553, 47 
295: \bibitem[Fruchter et al.(1999)]{fru99}
296: Fruchter, A.S., et al., 1999, ApJ, 519, L13
297: \bibitem[Galama et al.(1998)]{gal98}
298: Galama, T.J, et al, 1998, Nature, 395, 670
299: \bibitem[Hanany et al.(2000)]{han00}
300: Hanany, S., et al, 2000, ApJ, L545, 5
301: \bibitem[Heger et al.(2002)]{heg02}
302: Heger, A., Fryer, C.L., Woosley, S.E., Langer, N., \& Hartmann, D.H., 2002,
303:  submitted (astro-ph/0212469)
304: \bibitem[Kulkarni et al.(2000)]{kul00}
305: Kulkarni, S.R., et al, 2000, Proc.SPIE, 4005, 9
306: \bibitem[Levinson et al.(2002)]{lev02} 
307: Levinson, A., Ofek, E., Waxman, E., Gal-Yam, A., 2002, ApJ, 576, 923  
308: \bibitem[Perlmutter et al.(1999)]{per99}  
309: Perlmutter, S., et al, 1999, ApJ, 517, 565
310: \bibitem[Porciani \& Madau(2001)]{por01}
311: Porciani, C., \& Madau, P., 2001, ApJ, 548, 522 
312: \bibitem[Paccynski(1998)]{pac98}
313: Paczynsky B., 1998, ApJ, 494, L45
314: \bibitem[Reichart(2001)]{rei01}
315: Reichart, D.E., 2001, ApJ, 554, 643
316: \bibitem[Rossi et al.(2002)]{ros02}
317: Rossi, E., Lazzati, D., \& Rees, M.J., 2002a, MNRAS, 332, 945
318: \bibitem[Schmidt et al.(1998)]{sch98}
319: Schmidt, B., et al, 1998, ApJ, 507, 46
320: \bibitem[Sethi\&Bhargavi(2001)]{set01}
321: Sethi, \& Bhargavi, 2001, A\&A, 376, 10
322: \bibitem[Stanek et al.(2003)]{sta03}
323: Stanek, K.Z., et al., 2003, astro-ph/0304173
324: \bibitem[Woosley(1993)]{woo93} 
325: Woosley, S., 1993, ApJ, 405, 273
326: \bibitem[Zhang \& Meszaros(2002)]{zha02} 
327: Zhang, B., \& Meszaros, P., 2002, ApJ, 571, 876
328: \end{thebibliography}
329: 
330: \begin{deluxetable}{crrrrr}
331: \tabletypesize{\scriptsize}
332: \tablecaption{A sample of 33 GRBs with individually determined redshifts
333: \tablenotemark{a}\label{tbl-1}}
334: \tablewidth{0pt}
335: \tablehead{
336: %\colhead{GRB} 
337: GRB & Redshift $z$ & Photon flux\tablenotemark{b} & 
338: Luminosity \tablenotemark{c}& $\theta_j$\tablenotemark{d} & Instrument}
339: \startdata
340: 970228  & 0.695 & 10      & $2.13 \times 10^{58}$ &         & SAX/WFC\\
341: 
342: 970508  & 0.835 & 0.97    & $3.24 \times 10^{57}$ & 0.293   & SAX/WFC\\
343: 
344: 970828  & 0.9578& 1.5     & $7.04 \times 10^{57}$ & 0.072   & RXTE/ASM\\
345: 
346: 971214  & 3.42  & 1.96    & $2.08 \times 10^{59}$ & $>0.056$& SAX/WFC\\
347: 
348: 980425  & 0.0085& 0.96    & $1.54 \times 10^{53}$ &         & SAX/WFC\\
349: 
350: 980613  & 1.096 & 0.5     & $3.28 \times 10^{57}$ & $>0.127$& SAX/WFC\\
351: 
352: 980703  & 0.966 & 2.40    & $1.15 \times 10^{58}$ & 0.135   & RXTE/ASM\\
353: 
354: 990123  & 1.6   & 16.41   & $2.74 \times 10^{59}$ & 0.050   & SAX/WFC\\
355: 
356: 990506  & 1.3   & 18.56   & $1.85 \times 10^{59}$ &         & BAT/PCA\\
357: 
358: 990510  & 1.619 & 8.16    & $1.40 \times 10^{59}$ & 0.053   & SAX/WFC\\
359: 
360: 990705  & 0.86  &         &                     & 0.054   & SAX/WFC\\
361: 
362: 990712  & 0.434 & 11.64   & $7.97 \times 10^{57}$ & $>0.411$& SAX/WFC\\
363: 
364: 991208  & 0.706 & 11.2*   & $2.48 \times 10^{58}$ & $<0.079$& Uly/KO/NE\\
365: 
366: 991216  & 1.02  & 67.5    & $3.70 \times 10^{59}$ & 0.051   & BAT/PCA\\
367: 
368: 000131  & 4.5   & 1.5*    & $3.05 \times 10^{59}$ &$<0.047$ & Uly/KO/NE\\
369: 
370: 000210  & 0.846 & 29.9    & $1.03 \times 10^{59}$ &         & SAX/WFC\\
371: 
372: 000301C & 0.42  & 1.32*   & $8.37 \times 10^{56}$ & 0.105   & ASM/Uly\\
373: 
374: 000214  & 2.03  &         &                     &         & SAX/WFC\\
375: 
376: 000418  & 1.118 & 3.3*    & $2.27 \times 10^{58}$ & 0.198   & Uly/KO/NE \\
377: 
378: 000911  & 1.058 & 2.86    & $1.72 \times 10^{58}$ &         & Uly/KO/NE\\
379: 
380: 000926  & 2.066 & 10*     & $3.13 \times 10^{59}$ & 0.051   & Uly/KO/NE\\
381: 
382: 010222  & 1.477 &         &                     &         & SAX/WFC\\
383: 
384: 010921  & 0.45  &         &                     &         & HE/Uly/SAX\\
385: 
386: 011121  & 0.36  & 15.04*  & $6.63 \times 10^{57}$ &         & SAX/WFC\\
387: 
388: 011211  & 2.14  &         &                     &         & SAX/WFC\\
389: 
390: 020405  & 0.69  & 7.52*   & $1.58 \times 10^{58}$ &         & Uly/MO/SAX\\
391: 
392: 020813  & 1.25  & 9.02*   & $8.19 \times 10^{58}$ &         & HETE\\
393: 
394: 021004  & 2.3   &         &                     &         & HETE\\
395: 
396: 021211  & 1.01  &         &                     &         & HETE\\
397: 
398: 030226  & 1.98  & 0.48*   & $1.35 \times 10^{58}$ &         & HETE\\
399: 
400: 030323  & 3.37  & 0.0048* & $4.91 \times 10^{56}$ &         & HETE\\
401: 
402: 030328  & 1.52  & 2.93*   & $4.31 \times 10^{58}$ &         & HETE\\
403: 
404: 030329  & 0.168 & 0.0009* & $7.03 \times 10^{52}$ &         & HETE
405: \enddata
406: \tablenotetext{a}{Compiled from S. Barthelmy's IPN redshifts and fluxes
407: (http://gcn.gsfc.nasa.gov/gcn/) and J.C. Greiner's catalogue on GRBs
408: localized with WFC (BeppoSax), BATSE/RXTE or ASM/RXTE, IPN, HETE-II
409: or INTEGRAL (http://www.mpe.mpg.de/~jcg/grbgeb.html)}
410: \tablenotetext{b}{in cm$^{-2}$s$^{-1}$}
411: \tablenotetext{c}{Photon luminosities in s$^{-1}$ derived from the measured redshifts and observed gamma-ray fluxes for the cosmological model described in \S2}
412: \tablenotetext{d}{Opening angles $\theta_j$ in the GRB-emissions refer to the 
413: sample listed
414: in Table I of Frail et al.(2001).}
415: \tablenotetext{*}{Extrapolated to the BATSE energy range 50 - 300 keV using the formula given in 
416: Appendix B of Sethi et Bhargavi (2001)}
417: \end{deluxetable}
418: 
419: \newpage
420: \centerline{Figure Captions}
421: %\mbox{}\\
422: %\mbox{}\\
423: %{\bf Figure 1.} Shown is the luminosity threshold as a function of redshift for the 
424: %BATSE flux-threshold of 0.2 $\mathrm{photon}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}$,
425: %where $L_{\mbox{\tiny min}}$ is in units of cm$^{-2}$s$^{-1}$.
426: \mbox{}\\
427: \mbox{}\\
428: {\bf Figure 1.} Shown are three redshift distributions: the observed sample derived
429: from Table 1 (white), the true sample assuming the GRB event rate is 
430: locked to the star-formation rate (hachured), and the sample of detectable 
431: GRBs predicted by our model according to a log-normal peak-luminosity distribution
432: function (grey). The continuous line represents the cosmic star formation rate
433: according to a $\Lambda-$dominated cold dark matter universe.
434: \mbox{}\\
435: \mbox{}\\
436: {\bf Figure 2.} Shown is a simulation of the redshift distribution of the
437: observed GRBs as a function of flux-limit, corresponding to various
438: instruments including the upcoming SWIFT mission. The results are  
439: derived assuming the GRB event rate to be locked to the star-formation
440: rate, using the best fit log-normal peak-luminosity distribution function
441: used in Fig. 2. HETE-II tresholds are 0.21 (SXC), 0.07 (WXM) and 0.3 (FREGATE)
442: in units of cm$^{-2}$s$^{-1}$.
443: \mbox{}\\
444: \mbox{}\\
445: {\bf Figure 3.} Shown is a comparison between the flux distributions
446: derived from our model (in grey) and from a sample of 67 GRBs
447: observed with IPN (in wite). The fit serves as a test for
448: our model assumptions, namely a log-normal GRB-luminosity 
449: function and a GRB event rate locked to the star-formation rate.
450: \mbox{}\\
451: \mbox{}\\
452: %{\bf Figure 5.} Shown is the fraction of detectable GRBs as a function of redshift 
453: %for the optimal parameters in fitting a log-normal GRB-luminosity function.
454: %\mbox{}\\
455: %\mbox{}\\
456: {\bf Figure 4.} Shown is a plot of the opening angle $\theta_j$ of GRB-emissions
457: versus redshift $z$ in the sample of \citet{fra01}, as derived from achromatic breaks 
458: in the GRB light curves. These results indicate an anticorrelation
459: between $\theta_j$ and $z$. For standard GRB-energies, this introduces a peak-luminosity
460: function of GRBs which is correlated with the beaming factor $1/f_b$. This 
461: allows the beaming factor to be determined also in terms of the ratio of the
462: unseen-but-true GRB event rate to the observed GRB event rate, using the current 
463: flux-limited sample of 33 GRBs with individually measured redshifts.
464: 
465: \newpage
466: %\begin{figure}
467: %\plotone{fig1}
468: %\end{figure}
469: 
470: \begin{figure}
471: \plotone{f1}
472: \caption{}
473: \end{figure}
474: 
475: \begin{figure}
476: \plotone{f2}
477: \caption{}
478: \end{figure}
479: 
480: \begin{figure}
481: \plotone{f3}
482: \caption{}
483: \end{figure}
484: 
485: %\begin{figure}
486: %\plotone{fig5}
487: %\end{figure}
488: 
489: \begin{figure}
490: \plotone{f4}
491: \caption{}
492: \end{figure}
493: 
494: \end{document}
495: