0808.3391/ms.tex
1: \documentclass{emulateapj}
2: \usepackage{times}
3: \usepackage{lscape}
4: %\documentstyle[12pt,aasms4]{article}
5: %\documentclass[12pt,preprint]{aastex}
6: 
7: \voffset=-.915truein
8: 
9: \def\la{\mathrel{\mathpalette\fun <}}
10: \def\ga{\mathrel{\mathpalette\fun >}}
11: 
12: %\def\space{\noalign{\medskip}}
13: \def\fun#1#2{\lower0.837ex\vbox{\baselineskip0ex\lineskip0.209ex
14:   \ialign{$\mathsurround=0ex#1\hfil##\hfil$\crcr#2\crcr\sim\crcr}}}
15: 
16: 
17: \def\rh{R_{\rm h}}
18: \def\trh{t_{\rm rh}}
19: \def\pc{{\rm ~pc}}
20: \def\yr{{\rm ~yr}}
21: \def\kms{{\rm ~km~s^{-1}}}
22: \def\sec{{\rm ~sec}}
23: \def\erg{{\rm ~ erg}}
24: \def\cm{{\rm ~ cm}}
25: \def\gr{{\rm ~ g}}
26: \def\Vlasov{collisionless Boltzmann equation}
27: \def\etal{{\it et al. }}
28: \def\MSUN{\rm M_{\odot}}
29: \newcommand{\et}{et al.\ }
30: \def\degr{^\circ}
31: 
32: \def\msun{M_\odot}
33: \def\msunyr{M_\odot \ {\rm yr}^{-1}}
34: \def\sles{\lower2pt\hbox{$\buildrel {\scriptstyle <}
35:    \over {\scriptstyle\sim}$}}
36: 
37: \def\sgreat{\lower2pt\hbox{$\buildrel {\scriptstyle >}
38:    \over {\scriptstyle\sim}$}}
39: \def\Rf{\parindent=0pt\medskip\hangindent=3pc\hangafter=1}
40: \def\la{\mathrel{\mathpalette\fun <}}
41: \def\ga{\mathrel{\mathpalette\fun >}}
42: \def\aprop{\mathrel{\mathpalette\fun \propto}}
43: \def\pgf{\noindent\hangindent = 0.5in\hangafter=1}
44: \def\gks{}
45: 
46: 
47: \begin{document}
48: 
49: 
50: \title{Correlations of Prompt and Afterglow 
51:     Emission in {\it Swift} Long and Short Gamma Ray Bursts}
52: \shortauthors{GEHRELS et al.}
53: \author{
54: N.~Gehrels\altaffilmark{1},
55: S.~D.~Barthelmy\altaffilmark{1},
56: D.~N.~Burrows\altaffilmark{2},
57: J.~K.~Cannizzo\altaffilmark{1,3},
58: G.~Chincarini\altaffilmark{4,5},
59: E.~Fenimore\altaffilmark{6},
60: C.~Kouveliotou\altaffilmark{7},
61: P.~O'Brien\altaffilmark{8},
62: D.~M.~Palmer\altaffilmark{6},
63: J.~Racusin\altaffilmark{2},
64: P.~W.~A.~Roming\altaffilmark{2},
65: T.~Sakamoto\altaffilmark{1,3},
66: J.~Tueller\altaffilmark{1},
67: R.~A.~M.~J.~Wijers\altaffilmark{9},
68: B.~Zhang\altaffilmark{10}
69: }
70: \altaffiltext{1}{NASA-Goddard Space Flight Center, Greenbelt, MD 20771, 
71:                  neil.gehrels@nasa.gov}
72: \altaffiltext{2}{Department of Astronomy \& Astrophysics,
73:                  Pennsylvania State University, State College, PA 16802}
74: \altaffiltext{3}{CRESST/Joint Center for Astrophysics,
75:                  University of Maryland, Baltimore County, Baltimore, MD 21250}
76: \altaffiltext{4}{INAF-Osservatorio Astronomico di Brera, 1-23807 Merate, Italy}
77: \altaffiltext{5}{Universita degli studi di Milano Bicocca, 1-20126, Milano, Italy}
78: \altaffiltext{6}{Los Alamos National Laboratory, P.O. Box 1663, Los Alamos, NM, 87545}
79: \altaffiltext{7}{NASA-Marshall Space Flight Center, NSSTC, VP-62, 
80:                  320 Sparkman Drive, Huntsville, AL 35805}
81: \altaffiltext{8}{Department of Physics \& Astronomy, University of Leicester, Leicester, LE1 7RH, UK}
82: \altaffiltext{9}{Faculty of Science, Astronomical Institute ``Anton Pannekoek'', 
83:               University of Amsterdam, Kruislaan 403, 1098 SJ Amsterdam, The Netherlands}
84: \altaffiltext{10}{Department of Physics and Astronomy, University of 
85:                   Nevada Las Vegas, Las Vegas, NV 89154}
86: 
87: 
88: 
89: %ABSTRACT
90: 
91: \begin{abstract}
92: Correlation studies of prompt and afterglow emissions
93: from gamma-ray bursts (GRBs) between different spectral
94: bands has been difficult to do in the past because 
95: few bursts had comprehensive and intercomparable afterglow 
96: measurements.  In this paper\footnote{to appear in The Astrophysical Journal, Dec 20, 2008, v. 689, no. 2}
97:   we present a large and 
98: uniform data set for correlation analysis based on bursts 
99: detected by the {\it Swift} mission.  For the first time, short
100: and long bursts can be analyzed and compared.  It is found
101: for both classes that the optical, X-ray and gamma-ray 
102: emissions are linearly correlated, but with a large spread 
103: about the correlation line; stronger bursts tend to have 
104: brighter afterglows, and bursts with brighter X-ray afterglow 
105: tend to have brighter optical afterglow.  Short bursts are, 
106: on average, weaker in both prompt and afterglow emissions.  
107: No short bursts are seen with extremely low optical 
108: to X-ray ratio as occurs for ``dark'' long bursts.  
109: Although statistics are still poor for short bursts, there is 
110: no evidence yet for a subgroup of short bursts with high 
111: extinction as there is for long bursts.  Long bursts are 
112: detected in the dark category at the same fraction as for 
113: pre-{\it Swift} bursts.  Interesting cases are discovered of 
114: long bursts that are detected in the optical, and yet have 
115: low enough optical to X-ray ratio to be classified as dark. 
116:  For the prompt emission, short and long bursts have different 
117: average tracks on flux {\it vs} fluence plots.  In {\it Swift}, GRB 
118: detections tend to be fluence limited for short bursts and 
119: flux limited for long events.
120: \end{abstract}
121: 
122: 
123: \keywords{
124:  gamma rays: bursts
125: }
126: 
127: 
128: 
129: 
130: 
131: \section{Introduction}
132: 
133: %1. INTRODUCTION
134: 
135: One of the longest enduring Gamma Ray Burst (GRB) classification 
136: schemes is based on their distributions in duration and spectral 
137: hardness.  Both quantities seem to cluster into two separate 
138: classes with the longer events (those above $\sim2$ s; 
139: Kouveliotou et al. 1993) being predominantly softer while the 
140: shorter ones are harder. The mechanism for the origin of the 
141: GRB explosions (the central engine) appears to be quite 
142: different for the two types. % (van Paradijs, Kouveliotou \& Wijers 2000, 
143:                              %and references therein).  
144: Long bursts are ascribed to the 
145: core collapse to a black hole of a massive, young, rapidly 
146: rotating star in the ``collapsar'' model (Woosley 1993; 
147: MacFadyen \& Woosley 1999; Woosley \& Bloom 2006) which is 
148: supported by observations such as the coincidence of SNe with 
149: well-observed nearby GRBs (Galama et al. 1998; 
150: Bloom et al. 1999; Staneck et al. 2003; Hjorth et al. 2003;
151: Pian et al. 2006). 
152: %The prevalent model for short bursts suggests they are caused 
153:  The prevalent model for short bursts  has  them        caused 
154: by the coalescence of a binary pair of compact old stars 
155: (Lattimer \& Schramm 1974; Paczy\'nski 1986; Eichler et al. 1989; 
156: Mochkovitch et al. 1993; Rosswog, Ramirez-Ruiz, \& Davies 2003;
157: Oechslin, Janka, \& Marek  2007) which is supported by recent observations 
158: of progenitor sites with low star formation activity 
159: (Gehrels et al. 2005; Bloom et al. 2006; Fox et al. 2005; 
160: Villasenor et al. 2005, Hjorth et al. 2005; 
161: Barthelmy et al. 2005b; Berger et al. 2005).  In both scenarios, 
162: a highly-relativistic collimated outflow of particles and 
163: %adiation (jet) occurs producing prompt gamma-ray emission 
164: radiation       occurs producing prompt gamma-ray emission 
165: from shock accelerated electrons, which evolves into a 
166: long-lasting afterglow from shock interactions with 
167: the circumburst medium (e.g., M\'esz\'aros \& Rees 1997).  
168: For short bursts there are also models for the afterglow in 
169: which a radioactive wind causes emission in the first 
170: day or so (Li \& Paczy\'nski 1998, Kulkarni 2005).
171: 
172: Correlation studies of prompt and afterglow emission 
173: are crucial for understanding their production mechanisms 
174: and environmental effects.  For example, Jakobsson 
175: et al. (2004) developed a criterion for determining which GRBs 
176: are ``dark'' bursts, by comparing the relative intensity of 
177: their X-ray and optical afterglows to find what fraction of 
178: bursts have high column densities. 
179: Stratta et al. (2004) studied the X-ray and optical absorption
180: properties of 13 GRBs studied by {\it BeppoSax}.
181:  Roming et al. (2006) and 
182: Fynbo et al. (2007) expanded on previous work to include (long) bursts 
183: from the {\it Swift} satellite.  A more detailed work on dark bursts 
184: using a broad-band spectral analysis is given by 
185: Rol et al. (2005, 2007).  Zhang et al. (2007) present a study 
186: comparing radiative efficiencies for short and long bursts as 
187: derived from a correlation analysis.  Using {\it Swift} short bursts,
188:  Berger (2007) compared their X-ray afterglow to their 
189: gamma-ray prompt emission, and found that 20\% have anomalously 
190: low X-ray to gamma ray ratios indicating very low density 
191: burst sites, possibly in globular clusters, for that 
192: subpopulation (see also Berger et al 2007).
193:  %
194:    Other correlation studies
195:    have been undertaken by
196:    Salmonson \& Galama (2002),
197:    Firmani et al. (2006),
198:    Nava et al. (2006),
199:    Butler (2007), and
200:    Nysewander, Fruchter, \& Pe'er (2008).
201: % are useful for 
202: %determining the properties of long and short bursts, and 
203: %particularly the properties of the circumburst environment.  
204: An early study of X-ray afterglow properties at $t=11$ hr
205: was carried out by Piran et al. (2001).
206: 
207: %To date there has not been an comprehensive analysis of the 
208: %correlations for both long and short GRBs.
209: In this study we perform correlation studies using the
210:  extensive data set from {\it Swift}.  Sections 2 and 3 cover
211:  observations and results, respectively, while in Section 4, 
212: we discuss the implications of the results and in Section 5 
213: the conclusions and future prospects.
214: 
215: \section{Observations}
216: %2. OBSERVATIONS
217: 
218: \subsection{{\it Swift} Studies}
219: %2.1 {\it Swift} Studies
220: 
221: The {\it Swift} mission (Gehrels et al. 2004) has so far 
222: provided {\it uniform observations 
223: of prompt and afterglow emission}
224: for hundreds of GRBs.  This sample is an order of magnitude 
225: larger than the one previously available with e.g., the 
226: {\it BeppoSAX} satellite
227: (de Pasquale et al. 2006).\footnote{see also website by J. Greiner: 
228:  http://www.mpe.mpg.de/$\sim$jcg/grbgen.html}
229: Further, {\it Swift} X-ray observations 
230: covering time-scales from 1 minute to several days 
231: after the burst are provided for the first time 
232: for   most every GRBs.  After three years of operations, 
233: our data set has now reached a critical size where 
234: statistically meaningful correlations can be studied.
235: 
236: 
237: 
238: We present here three correlation studies:  
239: (1) X-ray {\it vs} optical afterglow, 
240: (2) gamma-ray prompt 
241: {\it vs} X-ray afterglow, and 
242: (3) prompt gamma-ray peak 
243: flux {\it vs} fluence. 
244:                All the data used in this study are listed 
245: in Tables $1-4$ except that gamma-ray data are 
246: listed only for 
247: those bursts with, at least, an X-ray afterglow.  The full list 
248: of fluences and fluxes for the 193 bursts used for study (3) are 
249: directly from the Sakamoto et al. (2008) tables.  
250: We include all {\it Swift} bursts from January 2005 
251: through July 2007 for studies (1) and (2) and through 
252: February 2007 for study (3).
253: % add:
254:    We adopt $T90 = 2$ s for the dividing line
255:    between long and short GRBs,
256:    except for ones with soft extended emission.
257:    In those cases
258:    the duration of the initial hard pulse was required to be 
259:    $<2$ s, and only that emission was used in the analysis
260:    (GRB 050724, 051227, 061006, 061210, and 070714B).
261: %
262:    Including the extended emission
263:    in the fluence would increase it by a factor
264:    $\la 2$         and would not significantly
265:    change the correlations.
266: 
267: 
268: 
269: For the X-ray {\it vs} optical afterglow study, we use the 
270: methods developed by Jakobsson et al. (2004) in their 
271: comparison of X-ray and optical afterglow fluxes for 
272: pre-{\it Swift} bursts.  In order to compare to the 
273: Jakobsson et al. results, we use the same definition 
274: of quantities:  the X-ray flux density at 3 keV, the optical flux in the 
275: $R-$band, and sampling time at 11 hr after the burst. 
276:  The {\it Swift} X-ray lightcurves have been found typically 
277: to have complex shapes (Nousek et al. 2005; Zhang et al. 2005) 
278: often including a poorly understood ``plateau phase''; the use of 
279: flux at 11 hr in most cases avoids sampling during 
280: the plateau phase and gives a measure of the true burst afterglow.
281: 
282: \subsection{X-ray Fluxes}
283: %2.2 X-ray Fluxes
284: 
285: The X-ray fluxes are from measurements 
286: of the {\it Swift} X-Ray Telescope 
287: (XRT; Burrows et al. 2005).  Our primary data product 
288: for the XRT flux is the integral flux between 
289: 0.3 and 10 keV corrected for absorption at low 
290: energies (unabsorbed flux).  This is converted 
291: to the flux density at 3 keV using the measured 
292: spectral index. 
293: %
294: Given an integral $0.3-10$ keV X-ray flux $[I_x]=$ erg cm$^{-2}$ s$^{-1}$
295:     and a 
296: $0.3-10$ keV X-ray photon index $n$, 
297: the flux density at 3 keV, in $\mu$Jy, is given by
298: \begin{equation}
299: f_X(3 \ {\rm keV}) = 4.13\times 10^{11} {I_x (2-n) E_0^{1-n} \over{\left( { E_2^{2-n} - E_1^{2-n}  } \right) }},
300: \end{equation}
301: where $E_0=  3$ keV,
302:       $E_1=0.3$ keV, and
303:       $E_2=10$ keV.
304: %
305: %I used her data on flux and photon spectral index to calculate the microJy values.
306: %
307: %Here is the actual value from my spreadsheet:
308: %=N4*(1-X4)*3^(-1*X4)/(10^(1-X4) - 0.3^(1-X4))/0.00000000000242
309: %
310: %where N4 is the Jacusin flux and X4 is the energy index (photon index-1).
311: %
312: %
313:  The integral fluxes, photon spectral 
314: indices and flux densities are listed 
315: in Tables $1-3$.
316:     A 10\% systematic uncertainty was 
317: added in quadrature to the measured error to account 
318: for uncertainties in the shape and variability of the lightcurves.
319: 
320: 
321: The integral flux calculation was carried out as follows
322: (see J. Racusin et al. 2008, in preparation,
323: for a more detailed discussion of the method).
324:  Level 1 data products were downloaded from 
325: the NASA/GSFC {\it Swift} Data Center (SDC) and 
326: processed using XRTDAS software (v2.0.1).  The 
327: {\tt xrtpipeline} task was used to generate 
328: level 2 cleaned event files.  Only events with 
329: Windowed Timing (WT) mode grades $0-2$ and 
330: Photon Counting (PC) mode grades $0-12$ and energies 
331: between $0.3-10.0$ keV were used in subsequent 
332: temporal and spectral analysis.
333: 
334: The XRT light curves were created by extracting 
335: the counts in a circular region around 
336: the GRB afterglow with a variable source 
337: radius designed to optimize the $S/N$ depending 
338: on the count rate.  They were background subtracted, 
339: pile-up corrected where applicable, 
340: exposure map corrected, and corrected for
341: the fraction of the PSF excluded by the extraction region.  
342: The number of counts per bin is variable and dependent 
343: on the count rate.  Time intervals of significant 
344: flaring were removed from the light curves and 
345: they were fit to power-laws, broken power-laws, and 
346: multiply broken power-law.  
347: Using these temporal fits, we 
348: interpolated the count rate at 11 hr.
349: 
350: Spectra for the power-law segments of the light curves 
351: were extracted individually to limit contamination 
352: by potential spectral variability.  The segment used 
353: for the counts to flux conversion was that
354:  at  11 hr.  The spectra were created by extracting 
355: the counts in a 20 pixel radius extraction region and 
356: a 40 pixel radius background region.  
357: The Ancillary Response Files were made using the {\tt xrtmkarf}
358: task and grouped with 20 counts per bin using the 
359: {\tt grppha} task.  The spectra were fit in XSPEC to absorbed 
360: power-laws and used to measure the $0.3-10$ keV flux and 
361: count rate which was applied to the interpolated count 
362: rate to convert into flux units.
363: 
364: 
365: 
366: \subsection{Optical and Gamma Ray Fluxes}
367: %2.3 Optical and Gamma Ray Fluxes
368: 
369: The optical fluxes are from measurements by ground-based 
370: telescopes and from the {\it Swift} UV Optical Telescope 
371: (UVOT; Roming et al. 2005).  An extensive literature search was done 
372: to find the best optical data for each burst.  Bursts were included 
373: in the study if measurements were available within a factor of 2 of 
374: 11 hr (i.e., at $>5.5$ hr or $<22$ hr).  The value at 11 hr was 
375: estimated by interpolations and extrapolations when measurements 
376: were not available exactly at 11 hr. The correction applied to the $R$ 
377: data for $t_{\rm obs}\ne$ 11 hr was 
378:   $\Delta m_R = -2.5\log_{10}(t_{\rm obs}/11. \ {\rm hr})$.
379:  The one exception to the 
380: factor of 2 criterion was GRB 070508 with measurements to only 4 hr, 
381: which was included because it appears to be an interesting 
382: dark burst candidate.  A few bursts are listed with optical 
383: flux upper limits at the bottom of Table 2.  This is not an 
384: exhaustive list of optical limits, but only those with low optical 
385: to X-ray ratio limits.  A 10\% systematic uncertainty was added in 
386: quadrature to the measured error to account for uncertainties in 
387: the shape and variability of the lightcurves. 
388: 
389: 
390: Galactic extinction was taken into account
391: using the study of
392:  Schlegel, Finkbeiner, \& Davis (1998).\footnote{http://nedwww.ipac.caltech.edu/forms/calculator.html} 
393: For the precise sky map positions
394:   we utilize  the XRT localizations.
395: For each data source reference, 
396:     a determination had to be made as to whether the
397: galactic correction had already been made. (For the GCN entries, it
398:   was always assumed the correction had not been made.)
399:  For most of the GRBs, the $R$ band correction is small (a few tenths
400: of a magnitude). The exceptions from Table 1 are 
401:        050724 ($\Delta m_R = 1.64$)
402:      % 051227 ($\Delta m_R = 3.66$),
403:                and
404:        061006 ($\Delta m_R = 0.85$);
405:                  the exceptions from Table 2 are 
406:        050713B ($\Delta m_R = 1.249$) and
407:        070704  ($\Delta m_R = 5.014$).
408:   Corrections this large are highly uncertain due to 
409: the patchiness in extinction in the Galactic plane.
410:  
411: 
412: 
413:    Our sample contains three  GRBs with redshift values
414: large enough ($z\simeq4$)  so that Lyman blanketing may affect the $R-$band fluxes.
415:  For these bursts $-$
416:                     050730, $z=3.97$; 
417:                     060206, $z=4.05$; and 
418:                     060210, $z=3.91$ $-$  %all have $z\la4$
419:     the expected redshifting of the Lyman series $(1+z)\simeq5$.
420: %  Thus  %, e.g.,
421:  For  Ly$\alpha$,  $1215.7  \AA \ \rightarrow \ \sim 6080 \AA$
422:    and  for
423:       Ly$\infty$,  $911.3 \AA \ \rightarrow \ \sim  4560 \AA$.
424: Thus the effect of the redshifted  absorption is to
425: % would  not be expected to significantly 
426: impact the blue edge of the  $R-$band filter
427:   $\lambda_R \approx 6600 \pm  800 \AA$.
428:           The $R-$band fluxes for the three highest $z$ bursts
429: (indicated by circles in Fig. 1)
430: scatter about the mean $R-$band flux line, however, 
431:      rather than being concentrated
432: at low $F_R$ values as would have been expected had blanketing been an  
433: issue.
434: Although for $z=4$ the Ly$\alpha$ feature will be shifted redward of the 
435: $\sim 6000 \AA$ (skew-symmetric)
436:    peak of the $R$ filter, the centroid and FWHM of the filter
437: still predict the bulk of the filter response
438:   to lie redward of most of the  Lyman series,
439:   which would be most dominant for  $z=4$ at  $\lambda < 6000 \AA$
440:    (e.g.,
441: for the Cousins [Bessell 90] $R_C$ filter,
442:     $\lambda_{\rm eff}=6588 \AA$ and $\Delta\lambda_{\rm FWHM} =1568 \AA$ $-$
443: Fukugita, Shimasaku, \& Ichikawa 1995, see their Table 9).
444: The absorptive effect would likely warrant a corrective multiplicative factor $\la1.5$, 
445:   which is small
446: given the $\sim 5$ decade spread in $F_R$ for Figure 1.  Therefore we do not  
447: attempt to correct $F_R$ for redshifted 
448:          Lyman absorption for these three high-z bursts.% indicated in Fig. 1.
449: 
450: 
451: 
452: \begin{figure}
453: \centering
454: \epsscale{1.205}
455: \plotone{f1.ps}
456: \vskip -2cm
457: \figcaption{
458: The
459: % X-ray afterglow 
460:  optical afterglow 
461: {\it vs} 
462: %optical afterglow 
463:   X-ray afterglow 
464: flux densities of {\it Swift}
465: short (shown in red) and long (shown in blue) GRBs
466:  at 11 hr after the burst.  The three circled bursts are those
467: for which $z>3.9$.
468: Also plotted are the pre-{\it Swift}
469:  GRBs (shown in green) taken from Jakobsson et al (2004).
470: For the Jakobsson et al subsample with upper limits,
471: we only plot those bursts for which the limiting magnitude
472: is fainter than $m_R=23$ (i.e., $\sim2$ $\mu$Jy).
473:   The XRT X-ray flux densities 
474: are at 3 keV and the optical flux densities are in the $R-$band 
475: (see Table 1 and 2).  Also shown is the ``dark'' burst 
476:     separation     line $\beta_{OX} = 0.5$ (Jakobsson et al 2004),
477:  and a line indicating  $\beta_{OX} = 1.0$.
478: \label{fig1}}
479: \smallskip
480: \end{figure}
481: 
482: 
483: 
484: The gamma-ray fluences and peak fluxes are in the $15-150$ keV band 
485: and are from the {\it Swift} Burst Alert Telescope 
486: (BAT; Barthelmy et al. 2005a) as compiled in the BAT GRB catalog 
487: (Sakamoto et al. 2008).  For the gamma-ray flux needed in 
488: study (3), we use 1 s binning as quoted by Sakamoto et al. (2008).
489: 
490: 
491: \subsection{Correlation Analysis}
492: %2.4 Correlation Analysis
493: 
494: For each study, we have performed fits to the two-parameter correlation 
495: data using the Spearman rank test (Spearman 1904; Press et al. 1986) and derived 
496: the correlation coefficient, $r$, to determine the degree and significance 
497: of the correlation.  Upper limits were not included in the fits.  
498: In the Spearman rank test, the probability 
499:   of a null correlation, $P_{\rm null}$, is given by
500: \begin{equation}
501: P_{\rm null} = {\rm erfc} [r (N/2)^{1/2}]
502: \end{equation}
503: where $N$ is the number of data points.  The significance of 
504: the correlation is $P_{\rm cor} = 1 - P_{\rm null}$.  
505: The fraction of the observed spread of the data that 
506: can be explained by the correlation is given by $r^2$.  
507: The fit parameters and correlation $r$ values are listed in Table 4.
508: %
509:  Equation (2)  only applies in the limit  of $N$ large ($\ga10-20$).
510:   For $N\simeq1-10$, the concept of applying a significance criterion
511: to a correlation study begins to lose its
512:   meaning.\footnote{This can be seen 
513: in the limit $N\rightarrow 2$ 
514:            where one considers two data points $(x_1,y_1)$ and
515:  $(x_2,y_2)$.  In this example
516:        $r\equiv 1$,
517:      so the statement ``$r=1$'' 
518:  carries no information and has no significance.}
519:  Therefore, although 
520: for completeness
521:       we list $r$ and $P_{\rm null}$ values for
522: cases with small $N$, we stress that %one cannot place any 
523: %significance nor draw any conclusions based on these values.
524: they are only indicative of trends in those cases. % and not accurate.
525: %
526: % One could in principle simulate the data using Monte Carlo techniques
527: % in the small $N$ limit, and assign significance values based on
528: % general, Bayesian considerations.
529: 
530:   
531: 
532: \section{Results}
533: %3. RESULTS
534: 
535: \subsection{X-ray and Optical Afterglow Correlations}
536: %3.1 X-ray and Optical Afterglow Correlations
537: 
538: Figure 1 shows the {\it Swift} X-ray afterglow average 
539: flux density 
540: at 3 keV as a function of the 
541: $R-$band optical flux density, both 
542: converted to $\mu$Jy at 11 hr after the burst, for short and long bursts. 
543: The pre-{\it Swift} data points are taken from 
544: Jakobsson et al. (2004) and are shown as filled green points.  
545: Also shown is the solid line of constant X-ray to optical spectral 
546: index that they propose separates the true ``dark'' bursts 
547: from the rest.  As listed in Table 4, the Spearman rank test for the 
548: two GRB populations in Figure 1 gives a null probability 
549: %f 0.13       or an 87\% correlation probability between the optical 
550: %nd X-ray flux of the long GRBs, and only 57\%       for the short population.  
551: of $\sim0.01$ or a  99\% correlation probability between the optical 
552: and X-ray flux densities of the long GRBs, 
553:           and only $\sim30$\% for the short population.  
554: 
555: The long {\it Swift} GRBs fall in the same general region of 
556: the plot as the pre-{\it Swift} ones.  As with the pre-{\it Swift} bursts, 
557: several {\it Swift} long bursts (detections and upper limits) 
558: also fall below the Jakobsson et al. dark line.  The brightest 
559: short GRBs fall     in the midst of the long GRB points, 
560: but in the region toward lower flux densities. % The dimmest short GRBs 
561: %extend significantly below the long burst population. 
562:   To 
563: date there are no short bursts that fall below the dark burst line; 
564: those with low optical flux densities or upper limits tend to also have 
565: weak X-ray flux densities that place them above the line. 
566: 
567: 
568: \subsection{Gamma Ray Prompt and X-ray Afterglow Correlations}
569: %3.2 Gamma Ray Prompt and X-ray Afterglow Correlations
570: 
571: 
572: \begin{figure}
573: \centering
574: \epsscale{1.205}
575: \plotone{f2.ps}
576: \vskip -2cm
577: \figcaption{
578:  The X-ray afterglow flux density {\it vs} gamma-ray prompt fluence of {\it Swift}
579:  short (shown in red) and long (shown in blue) GRBs at 11 hr after the burst.
580:  The XRT X-ray flux densities are at 3 keV and the BAT
581: gamma-ray fluences are between 15 and 150 keV (Sakamoto et al. 2008).
582:     The XRT and BAT
583: data are given in Table 1, 2 and 3.
584: \label{fig2}}
585: \smallskip
586: \end{figure}
587: 
588: 
589: 
590: 
591: We show in Figure 2 the average X-ray 
592:    afterglow flux density {\it vs} the gamma-ray 
593: fluence of the prompt emission for long and short {\it Swift} GRBs.  
594: %e find a highly significant correlation (99.9996\% probability) for 
595: We find a highly significant correlation (99.9999996\% probability) for 
596: the long GRBs, albeit with a wide spread in the data.  The correlation 
597: %f the short bursts is less significant (95\% probability) mostly due 
598: of the short bursts is less significant (69\% probability) mostly due 
599: to the smaller number of points.  There is an overlap between 
600: the brightest short bursts and the faintest long GRBs.  
601: The weakest short bursts are fainter than the weakest long bursts.
602: 
603: \subsection{Prompt Gamma Ray Fluence and Peak Flux Correlations}
604: %3.3 Prompt Gamma Ray Fluence and Peak Flux Correlations
605: 
606: 
607: \begin{figure}
608: \centering
609: \epsscale{1.205}
610: \plotone{f3.ps}
611: \vskip -2cm
612: \figcaption{
613: The prompt gamma-ray fluence {\it vs} peak flux measured by 
614: BAT in the 15 to 150 keV band for all bursts through 
615: February 2007 (Sakamoto et al. 2008). 
616:  Short bursts are shown by red symbols
617: and long bursts by blue.
618: \label{fig3}}
619: \smallskip
620: \end{figure}
621: 
622: 
623: 
624: 
625: 
626: Figure 3 shows the prompt emission fluence as a function of peak 
627: flux for GRBs detected by BAT.  We see a linear correlation for 
628: both short and long bursts with a significant spread in the 
629: correlation. The correlation probability is virtually 100\% 
630: %null probability $= 2\times 10^{-20}$) for long bursts and 96\% 
631: (null probability $= 2\times 10^{-29}$) for long bursts and 99.9998\% 
632: for short bursts.  The best fit lines are distinctly different 
633: for short and long bursts, with the long burst having a higher 
634: fluence on average for a given flux level than 
635: short bursts as expected from duration alone.
636: 
637: 
638: \section{Discussion}
639: %4. DISCUSSION
640: 
641: \subsection{Correlations \& Short/Long Distributions}
642: %4.1 Correlations \& Short/Long Distributions
643: 
644: 
645: We show in this work that correlations exist between 
646: prompt and afterglow fluxes of GRBs and between different 
647: wavelength bands in the afterglow.  The highest significance 
648: correlation is between the prompt emission gamma-ray fluence 
649: and the X-ray afterglow flux at a significance level 
650: %f 99.9996\% for long bursts and 95\% for short bursts.  
651: of 99.9999996\% for long bursts and 69\% for short bursts.  
652: The correlation between the optical afterglow and 
653: %-ray afterglow fluxes is less significant at 87\% 
654: %ignificance for long bursts and only 57\% for 
655: X-ray afterglow fluxes is less significant at 99\% 
656: significance for long bursts and only $\sim30$\% for 
657: short bursts (for a small sample, however).  %Due to the small sample size of the short GRBs,
658: %The specific significance values
659: %  for the short GRB correlations are only very  rough estimates
660: %since the assumption $N>>1$ underlying the 
661: %aforementioned $P_{\rm null}$ formalism is not satisfied 
662: %($N=8$, 10, and 17,
663: %respectively, for the three short GRB correlations discussed).
664: 
665: 
666: 
667: It is important to note that there is a 
668: wide spread in the data for all of the correlations.  
669: The correlations are real and significant, but the fraction 
670: of the observed variations due to the correlations 
671: between the above parameters accounts for only a portion 
672: of the data spread.  The correlation can only be used 
673: to predict a flux level to within approximately an 
674: order of magnitude.  The fraction of the variation due 
675: to the correlations is given by the square of the 
676: correlation parameter, $r$, which, as shown in Table 4, 
677: varies from a few percent to 50\%.  The rest of the data 
678: spread is due to other factors such as correlations with 
679: additional unknown parameters.  An example of an additional 
680: parameter is extinction in the optical afterglow.
681: 
682: 
683: Short bursts are weaker on average than long bursts in 
684: afterglow fluxes.  There is overlap with the dimmer 
685: long bursts, but the short bursts extend to lower intensities 
686: than seen for long bursts.  The average X-ray flux density at 3 keV 
687: at 11 hr for the short bursts is $<F_x({\rm short})> = 9.6\times10^{-3}\mu$Jy, 
688: which is more than an order of magnitude less than 
689: the average for long bursts of    $<F_x({\rm long})> = 0.10\mu$Jy.
690: 
691: 
692: %The correlation power-law fits in Table 4 show some differences 
693: %between long and short bursts in all 3 studies.  However, for the 
694: %first two studies (optical {\it vs} X-ray afterglow and gamma-ray prompt 
695: %{\it vs} X-ray afterglow) the differences between the fits are 
696: %statistically marginal of less than 3s significance.  
697: %%The following discussion also fits into the section, and can replace the
698: %%sentence "From the analysis of Zhang et al. (2007) ...."
699: %Similarly, Figure 2 also displays a positive correlation with a slope of
700: %roughly unity. This suggests that brighter bursts also have more kinetic
701: %energies in the afterglow phase to power the afterglow. This is a
702: %manifestation of similar radiative efficiency among different bursts,
703: %including both long and short GRBs. Such a point was made by Zhang et al.
704: %(2007) based on an analysis of a smaller sample of early {\it Swift} GRBs.
705: %Zhang et al.
706: %also performed a detailed analysis of both gamma-ray energy and afterglow
707: %kinetic energy based on the afterglow models. The more complete sample
708: %presented here strengthens this conclusion.
709: %%From the analysis of Zhang et al. (2007), the implication of 
710: %%this result is that the radiative efficiency is similar 
711: %%for long and short GRBs. 
712: % For the prompt emission fluence {\it vs} 
713: %flux study, the differences are highly statistically significant 
714: %as expected from the very fact that short and long bursts 
715: %have different durations as discussed below.
716: 
717: % Except for the bursts below the ``dark'' line, most bursts in Figure 1 are
718: %generally confined within a certain region defined by two lines with a
719: %slope $\beta_{OX} \sim 0.5$. This is consistent with that in general the
720: %optical and X-ray emission belong to the same spectral component with an
721: %index close to 0.5. For slow cooling which is generally relevant at $t=11$
722: %hr, one expects $\beta_{OX} \sim (p-1)/2$ for $\nu_m < \nu_O < \nu_X <
723: %\nu_c$, which has a typical value of 0.5 for $p=2$. This suggests that
724: %on average, the cooling frequency is above or not much below the
725: %X-ray band at 11 hr.  The lower boundary is defined by the detector
726: %sensitivity. The upper boundary is roughly 2 orders of magnitude above the
727: %lower boundary. This is defined by the upper limit of GRB energetics
728: %(highest luminosities) convolved by the redshift distribution effect. 
729: 
730: 
731: %- Sect 4.1: �Last 2 paragraphs. �I just wrote to Bing about these and suggested the following replacements:
732: 
733:  The X-ray to gamma-ray correlation in Figure 2 has a positive correlation with a slope of roughly unity. This
734: suggests that brighter bursts have more kinetic energy  in the afterglow phase to power the afterglow. This is a
735: manifestation of similar radiative efficiency among different bursts and between long and short GRBs. Such a point was
736: made by Zhang et al. (2007) based on an analysis of a smaller sample of early {\it Swift} GRBs.
737: 
738: Except for the bursts below the ``dark'' line, most bursts in Figure 1 are confined between lines with 
739: $\beta_{OX} =$
740: 0.5 and 1.0. This is consistent with a general interpretation that 
741: the optical and X-ray emission belong to the same
742: spectral component with an index close to 0.75.
743: Within the standard model for emission via synchrotron radiation,
744:  for slow cooling 
745: which is generally relevant at 
746:   $t = 11$ hr, one
747: expects $\beta_{OX} \sim (p-1)/2$ for $\nu_m < \nu_O < \nu_X < \nu_c$, 
748: which has a typical value of 0.75 for
749:     electron distribution power law $p = 2.5$. 
750: %
751:    (An equivalent statement is that 
752:      for this spectrum, the predicted ratio  
753:     $F_R/F_X \approx 350$ yields a
754:     line intermediate between %the 
755:     $\beta_{OX}=0.5$ and $1.0$ in % $1.0 in %a lines in 
756:     Fig. 1.) % $-$  broadly descriptive of our results. 
757:    This
758: suggests that on average, the cooling frequency is above or 
759: not much below the X-ray band at 11 hr.
760: 
761: 
762: 
763: \subsection{Dark GRBs}
764: %4.2 Dark GRBs
765: 
766: Another comparison of short and long GRBs relates to dark bursts.  
767: Jakobsson et al. (2004, see also De Pasquale et al. 2003)
768:        used the simple criterion to define dark bursts 
769: as those with extremely low optical to X-ray afterglow ratio, falling 
770: below the line of optical to X-ray spectral index, $\beta_{\rm OX}$, equal to 0.5. 
771:  It may seem counterintuitive that there can be dark bursts 
772: with optical detections and bursts not detected in the optical 
773: that are not ``dark'', but the important criterion is how 
774: optically faint the burst is relative to its X-ray flux.  
775: For the pre-{\it Swift} sample there were 5 bursts with 
776: upper limits below the dark-burst line
777: (restricting the Jakobsson et al. sample
778:   to include only those with 
779: upper limits fainter than
780: $m_R=23$, or $\sim2\mu$Jy),
781: compared to 24 bursts with 
782: actual measurements (not upper limits) above the line, giving a fraction 
783: %f $\sim17$\% in the dark category.  For {\it Swift} there are 4 bursts 
784: of $\sim17$\% in the dark category.  For {\it Swift} there are 2 bursts 
785: %ith upper limits (GRB 050713B, 061222A, 070621, and 070704) and 
786: with upper limits (GRB 050713B and  061222A) and 
787: %3 cases with measurements (GRB 060210, 070419B and 070508) below 
788:  3 cases with measurements (GRB 060210, 070419B and 070508) below 
789: the line compared with 34 long  bursts above the line for 
790: a fraction of $\sim17$\% in the dark category,     the same as 
791: the pre-{\it Swift} sample.  The conclusion is that {\it Swift}
792:  is sampling the same source environments as previous   instruments.
793: 
794: 
795: The discovery of 3 cases of dark bursts with optical detections is 
796: particularly interesting.  One possible concern with this finding is 
797: that {\it Swift} X-ray afterglows are contaminated in many bursts 
798: by emission components not from the external shocks, e.g. X-ray flares.
799:   In such instances, the Jakobsson et al. (2004) approach 
800: to define dark bursts is no longer relevant since it assumes that 
801: the X-ray and optical emission is from the same emission component, 
802: but separated by a cooling break.  However, the X-ray lightcurves 
803: for the {\it Swift} dark bursts are smooth around 11 hr
804: (and beyond the end of the X-ray plateau),
805:   with 
806: no significant contamination from other components.
807:  These are real ``dark'' bursts from both an observational 
808: and physics perspective.
809: 
810: 
811: 
812: %> Section 4.2
813: %> On dark GRBs, in paragraph 2 you make the (correct) point that the light curves are smooth. I wonder if you should
814: %> also say that the 11hr data, for the ones that are dark, are beyond the end of the X-ray plateau.
815: %> 
816: 
817: 
818: %In Section 4.2 (near the end), one can add the following discussion:
819: 
820: 
821: %The method applied here (as in Jakobsson et al 2004) 
822: %to investigate dark bursts
823: %is based on the hypothesis that both X-ray and optical emission are from
824: %the external forward shock. Multiwavelength observations in the {\it Swift} era
825: %reveal  puzzling chromatic features of 
826: %   afterglow breaks (e.g., Panaitescu
827: %et al. 2006; Liang et al. 2007, 2008) that are not consistent with the
828: %simplest forward shock model. Models invoking non-forward-shock origin of
829: %X-ray afterglows have been discussed in the literature (e.g., Genet et al.
830: %2007; Uhm \& Beloborodov 2007; Ghisellini et al. 2007; Shao \& Dai 2007;
831: %Panaitescu 2008). On the other hand, analyses suggest that the X-ray data
832: %are generally consistent with the temporal-index and spectral-index
833: %relations (e.g., Zhang \& M\'esz\'aros 2004) predicted by the forward shock
834: %models (Liang et al. 2007). The optical/X-ray data of some bursts (e.g.,
835: %Grupe et al. 2007; Mangano et al. 2007) are well consistent with the same
836: %forward shock model. The analysis presented in this paper suggests that
837: %generally optical/X-ray data are not inconsistent with belonging to the
838: %same afterglow component.
839: 
840: 
841: % Replace with stronger:
842: Correlation analyses between optical and X-ray can help answer 
843: the question of whether these two afterglow
844: components originate from the same physical processes. 
845:  It is assumed in the Jakobsson et al (2004) study
846: that both X-ray and optical emission arise from the external forward
847: shock. Multiwavelength observations in the {\it Swift} era reveal 
848: puzzling chromatic features of
849: afterglow breaks (e.g., Panaitescu et al. 2006; Liang et al. 2007, 2008) 
850: that are not consistent
851: with the simplest forward shock model. Models invoking 
852: non-forward-shock origin of X-ray
853: afterglows have been discussed in the literature 
854: (e.g., Genet et al. 2007; Uhm \& Beloborodov
855: 2007; Ghisellini et al. 2007; Shao \& Dai 2007; Panaitescu 2008). 
856: On the other hand, analyses
857: suggest that the X-ray data are generally consistent with 
858: the temporal index and spectral
859: index relations (e.g., Zhang \& M\'esz\'aros 2004) predicted by 
860: the forward shock models, although not in
861:  every case. (Liang
862: et al. 2007,
863: %I think you should also reference Dick's paper (Willingale,
864: %> R. et al. 2007, ApJ, 662, 1093) with Liang et al.
865: %,a 
866: Willingale, et al. 2007).
867:    The optical/X-ray data of some bursts 
868: (e.g., Grupe et al. 2007; Mangano et
869: al. 2007) are      consistent with the same forward shock model. 
870: %> The fourth sentence could then be
871: %"
872:    Regardless of the exact process, 
873:  the analysis presented in
874: this paper shows that generally optical/X-ray afterglow 
875: fluxes are correlated, which suggests that they are due to
876: the same emission process. The few cases well below the 
877: correlation line are found to be dark due to extinction
878: in the host galaxy.
879: 
880: 
881: For the first time we can search for dark short bursts.  
882: No short bursts are seen that fall below the dark-burst line.  
883: It is hard to find dark GRBs using this criterion since X-ray 
884: afterglow fluxes are also low for the short bursts.  However, there are 
885: some short bursts with bright X-ray afterglow, and, to date, none of 
886: those is seen to be highly deficient in optical afterglow.  Statistics 
887: are still small with only 5 optical detections, but if the observed 
888: trend continues we will be able to conclude that short bursts do not 
889: occur in regions with extremely high extinction as occurs for some 
890: long bursts.
891: 
892: %John,
893: %
894: %That is a nice find.  Can you add a short paragraph on this in section 4.2 and point out the importance?    Here is some
895: %for the text:
896: 
897:  We are beginning to have optical detections of bursts 
898: below the dark burst line.
899: In one of the three dark bursts with
900: detections (GRB 060210), the burst
901: is found to have high extinction associated with its host galaxy,
902: explaining the low optical flux (Curran et al. 2007b). 
903: %
904:  By modeling the differences between $\beta_{\rm opt}$, $\beta_X$, and $\beta_{OX}$, 
905:  and taking into
906:  account the Lyman$-\alpha$ absorption ($z=3.91$), 
907: the authors find the $R-$band source extinction could be $3.9\pm0.7$ mag
908:  ($\nu_c > \nu_O$) or $6.7\pm0.6$ mag   ($\nu_c < \nu_O$).
909: %
910:  This is an important development
911: in our understanding of dark GRBs.  
912:  (For two of the three dark bursts with
913: detections - 070419B and 070508 -
914: there has not yet been sufficiently detailed follow-up work
915: on the putative hosts for constraints to be placed on the host extinction.)
916: Assuming that the dark bursts can be largely explained by extinction, then the
917: optical - X-ray correlations, ignoring the dark bursts, would hold true.
918: % add:
919:     We note that new studies are being done to examine
920:   dark burst definitions (van der Horst et al. 2008).
921: 
922: 
923: 
924: 
925: %Please also add a note to the caption of Fig 1 about the beta_OX=1.0 line.
926: %Thanks,
927: %Neil
928: %
929: %
930: %On Jul 14, 2008, at 2:02 PM, John Cannizzo wrote:
931: %
932: %> 
933: %> 
934: %> Neil--
935: %> 
936: %> 
937: %55>  For two of them - 070419B and 070508 - within ADS there are no linked citations to the original GCN refs (6325 and
938: %> 6398), so it looks like there so far have not been detailed follow-up work. The GCN info is too sketchy to do much
939: %> with it.
940: %> 
941: %>  For 060210, there has been a journal follow-up with detailed constraints attempted on the host extinction (as
942: %> distinct from galactic extinction). By modeling the differences between beta_opt, beta_X, and beta_OX and taking into
943: %> account the abs. from Lyman alpha redshifting (z=3.91), the authors find the R extinction could be 3.9\pm0.7 mag
944: %> (cooling break \nu_c above optical) or 6.7\pm0.6 mag (cooling break
945: %> below optical). (Curran et al. 2007, A&A, 467, 1059).
946: %> 
947: %>  So yes - for the one GRB where one can make a statement, the host extinction does look very significant.
948: %> 
949: 
950: 
951: \subsection{Prompt Fluence and Flux Comparisons}
952: %4.3 Prompt Fluence and Flux Comparisons
953: 
954: 
955: The comparison of fluences and peak fluxes in the prompt emission as 
956: shown in Figure 3 is a different kind of study than in the other two 
957: above.  In this case, the strong observed correlation and high degree 
958: of separation of short and long bursts is expected; brighter bursts 
959: with higher peak fluxes naturally have higher fluences and short bursts 
960: tend to have lower fluence for a given flux by the very fact of their 
961: short duration.  Within the short and long classes, the spread in 
962: fluence that is seen for a given peak flux is due to the diversity 
963: of durations and spectral indices.  Bursts with longer duration and 
964: hard spectra have higher fluences for the same peak flux.
965: 
966: It is interesting to note in Figure 3 that the short bursts tend to be 
967: fluence limited in the BAT, while long bursts tend to be peak 
968: flux limited.  This is due to the way BAT operates.  A valid GRB trigger 
969: requires a statistically significant excess in both the rate 
970: and image domains (Fenimore et al. 2004).  The ability to form an 
971: image depends on the number of photons collected on various 
972: trigger timescales, which is related to the burst fluence.  Even for 
973: relatively high peak-fluxes, short bursts can have low fluence  values 
974: and be limited in the number of photons available for the image trigger.
975:   On the other hand, long bursts tend to have higher fluences for a 
976: given peak-flux and become rate limited before the image limit 
977: is reached.  BAT also has a pure-image mode for triggering where 
978: very long duration GRBs and other transients are found by comparing 
979: sky images instead of having a rate trigger.  The lowest long-burst 
980: point in Figure 3 at a peak flux of $\sim0.1$ was such an image-mode 
981: trigger for the very long ($T90 = 35$ min) and weak GRB 060218.  
982: A caveat on the above discussion is that the BAT trigger algorithm 
983: is complex with $\sim500$ different trigger criteria evaluated.  
984: There are many different thresholds and limits coming into play 
985: for short and long burst triggering, with some mix of flux and 
986: fluence limits for both types.
987: 
988: This study was based on a 1 s binning for the gamma-ray fluxes.  
989: We have also investigated the effect of using a smaller 
990: bin size of 64 ms.  Smaller bins pick out larger peak flux values 
991: when there is short time structure or when the burst has a duration 
992: shorter than the bin size.  The effect of the smaller bin size 
993: is to shift the short bursts to the right (higher peak flux) 
994: relative to the long bursts by about a factor of 5.  
995: The larger bin size that we use allows for
996:     better statistics and is more reliable for long bursts.  
997: In either case, the short bursts tend to cluster toward lower 
998: fluences than long bursts.
999: 
1000: \section{Future Prospects}
1001: %5. FUTURE PROSPECTS
1002: 
1003: The combined prompt and afterglow data set for {\it Swift} GRBs is the 
1004: largest available to date.  We have chosen a criterion on the 
1005: afterglow measurements for inclusion in this study of being a 
1006: solid measurement 11 hr after the burst.  Even with this 
1007: stringent definition, there are more than 100 long bursts with 
1008: X-ray afterglow data.  The optical detections at 11 hr are 
1009: less numerous with about 40 good measurements, but still enough 
1010: statistics for conclusions to be reached.
1011: 
1012: The short burst correlation  studies are possible now and key 
1013: results are beginning to emerge.  The {\it Swift} data base is 
1014: growing quickly with time.  In its expected lifetime of 
1015: $\sim10$ yr, the mission should provide a sample of 
1016: $>40$ short and $>400$ long GRBs with good afterglow and 
1017: prompt observations. 
1018: % The hints of a diverse population of 
1019: %short bursts in the current correlation plots and significant 
1020: %differences compared to long bursts predict an interesting 
1021: %future for this work.
1022: %- Sect 5: �replace last sentence with:
1023: That sized data set will allow more detailed correlations 
1024: studies to investigate the interesting trends found in
1025: the current analysis.
1026: 
1027: 
1028: 
1029: \acknowledgements
1030: %Acknowledgements:
1031: 
1032: The authors thank the following people for useful discussions: 
1033: A. Fruchter, J. Fynbo, P. Jakobsson, A. Loeb, M. Nysewander, and E. Rol.
1034: We also acknowledge our anonymous referee who
1035: provided excellent suggestions for improving the paper.
1036: 
1037: 
1038: %REFERENCES
1039: 
1040: \def\mnras{MNRAS}
1041: \def\apj{ApJ}
1042: \def\apjs{ApJS}
1043: \def\apjl{ApJL}
1044: \def\aj{AJ}
1045: \def\araa{ARA\&A}
1046: \def\aap{A\&A}
1047: 
1048: 
1049: \begin{thebibliography}{75}
1050: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
1051: 
1052: \bibitem[]{}
1053: Antonelli, L.~A., et al. 2006, A\&A, 456, 509
1054: 
1055: \bibitem[]{}
1056: Barthelmy, S.~D., et al. 2005a, Space Sci Rev, 120, 143
1057: 
1058: \bibitem[]{}
1059: Barthelmy, S.~D., et al. 2005b, Nature, 438, 994
1060: 
1061: \bibitem[]{}
1062: Berger, E.,  et al. 2005, Nature, 438, 988
1063: 
1064: \bibitem[]{}
1065: Berger, E.,  et al. 2006, GCN Circ. 5697, http://gcn.gsfc.nasa.gov/gcn/gcn3/5697.gcn3
1066: 
1067: %\bibitem[]{}
1068: %Berger, E., \& Soderberg, A.~M. 2005, 
1069: %                        GCN Circ. 4419, http://gcn.gsfc.nasa.gov/gcn/gcn3/4419.gcn3
1070: 
1071: \bibitem[]{}
1072: Berger, E. 2007, ApJ, 670, 1254
1073: %erger, E. 2007, ApJ, in press (astro-ph 0702694)
1074: 
1075: \bibitem[]{}
1076: %erger, E., et al. 2006, ApJ, in press (astro-ph/0611128)
1077: Berger, E., et al. 2007, ApJ, 664, 1000
1078: 
1079: \bibitem[]{}
1080: Bikmaev, I., et al. 2005, GCN Circ. 3797, http://gcn.gsfc.nasa.gov/gcn/gcn3/3797.gcn3
1081: 
1082: \bibitem[]{}
1083: Bloom, J.~S., et al. 1999, Nature, 401, 453
1084: 
1085: \bibitem[]{}
1086: Bloom, J.~S., et al. 2006, ApJ, 638, 354
1087: 
1088: \bibitem[]{}
1089: Burrows, D.~N., et al. 2005, Space Sci Rev, 120, 165
1090: 
1091: \bibitem[]{}
1092: Butler, N.~R. 2007, ApJ, 656, 1001
1093: 
1094: %\bibitem[]{}
1095: %Butler, N.~R., et al. 2006, ApJ, 652, 1390
1096: 
1097: %\bibitem[]{}
1098: %Cenko, S.~B., Fox, D.~B., \& Price, P.~A. 2006,
1099: %                         GCN Circ. 5912, http://gcn.gsfc.nasa.gov/gcn/gcn3/5912.gcn3
1100: 
1101: \bibitem[]{}
1102: Cenko, S.~B., et al. 2006, ApJ, 652, 490
1103: 
1104: \bibitem[]{}
1105: Chen, H.-W., et al. 2007, ApJ, 663, 420
1106: 
1107: \bibitem[]{}
1108: Cobb, B.~E., \& Bailyn, C.~D. 
1109:                    2005, GCN Circ. 3104, http://gcn.gsfc.nasa.gov/gcn/gcn3/3104.gcn3
1110: 
1111: \bibitem[]{}
1112: Curran, P.~A., et al. 2007a, MNRAS, 381, L65
1113: 
1114: \bibitem[]{}
1115: Curran, P.~A., et al. 2007b, A\&A, 467, 1049
1116: 
1117: %(Curran et al. 2007, A&A, 467, 1059).
1118: 
1119: %\bibitem[]{}
1120: %Dado, S., Dar, A., \& De Rujula, A. 2007, astro-ph/0706.0880v1
1121: 
1122: \bibitem[]{}
1123: Dai, X., et al. 2007, ApJ, 658, 509
1124: 
1125: \bibitem[]{}
1126: D'Avanzo, P., et al.
1127:                    2006, GCN Circ. 5884, http://gcn.gsfc.nasa.gov/gcn/gcn3/5884.gcn3
1128: 
1129: \bibitem[]{}
1130: Della Valle, M., et al. 2006, ApJ, 642, L103
1131: 
1132: \bibitem[]{}
1133: de Pasquale, M.,  et al.
1134: %e Pasquale, M., Piro, L., Perna, R., Costa , E., Feroci, M,
1135: %         Gandolfi, G., In't Zand, J., Nicastro, L., 
1136: %          Frontera, F., Antonelli, L.~A., Fiore, F.,
1137: %         \& Stratta, G. 
1138:           2003, ApJ, 592, 1018
1139: 
1140: \bibitem[]{}
1141: de Pasquale, M., et al.
1142: %e Pasquale, M., Piro, L., Gendre, B., Amati, L., Antonelli, L.~A., 
1143: %          Costa, E., Feroci, M., Frontera, F., Nicastro, L., Soffitta, P.,
1144: %         \& in't Zand, J. 
1145:           2006, A\&A, 455, 813
1146: 
1147: \bibitem[]{}
1148: Durig, D.~T., \& Price, A.
1149:                    2005, GCN Circ. 4023, http://gcn.gsfc.nasa.gov/gcn/gcn3/4023.gcn3
1150: 
1151: \bibitem[]{}
1152: Efimov, Yu., et al. 2006, GCN Circ. 5986, http://gcn.gsfc.nasa.gov/gcn/gcn3/5986.gcn3
1153: 
1154: \bibitem[]{}
1155: Eichler, D., Livio, M., Piran, T., \& Schramm, D.~N. 1989, Nature, 340, 126
1156: 
1157: \bibitem[]{}
1158: Fenimore, E., et al. 2004, AIP, 727, 667
1159: 
1160: \bibitem[]{}
1161: Firmani, C., Ghisellini, G., Avila-Reese, V., 
1162:           \& Ghirlanda, G. 2006,
1163:                 MNRAS, 370, 185
1164: 
1165: \bibitem[]{}
1166: Fox, D.~B., et al. 2005, Nature, 437, 845
1167: 
1168: \bibitem[]{}
1169: Fukugita, M., Shimasaku, K., \&  Ichikawa, T. 1995, PASP, 107, 945
1170: 
1171: \bibitem[]{}
1172: Fynbo, J.~P.~U., et al. 2007, astro-ph/0703458
1173: 
1174: \bibitem[]{}
1175: Fynbo, J.~P.~U., et al. 2006, Nature, 444, 1047
1176: 
1177: \bibitem[]{}
1178: Galama, T.~J., et al. 1998, Nature, 395, 670
1179: 
1180: \bibitem[]{}
1181: Garnavich, P., \& Karska, A. 2006,
1182:                         GCN Circ. 5253, http://gcn.gsfc.nasa.gov/gcn/gcn3/5253.gcn3
1183: 
1184: \bibitem[]{}
1185: Gehrels, N., et al. 2004, ApJ, 611, 1005
1186: 
1187: \bibitem[]{}
1188: Gehrels, N., et al. 2005, Nature, 437, 859
1189: 
1190: \bibitem[]{}
1191: Genet, F., Daigne, F., \& Mochkovitch, R. 2007, MNRAS, 381, 732
1192: 
1193: \bibitem[]{}
1194: George, K., Banerjee, D.~P.~K., Chandrasekhar, T., \& Ashok, N.~M. 2006, ApJ, 640, L13
1195: 
1196: \bibitem[]{}
1197: Ghirlanda, G., Nava, L., Ghisellini, G., \& Firmani, C. 2007, A\&A, 466, 127
1198: 
1199: \bibitem[]{}
1200: Ghisellini, G., Ghirlanda, G., Nava, L., \& Firmani, C. 2007, ApJ, 658, L75
1201: 
1202: \bibitem[]{}
1203: Grupe, D. et al. 2007, ApJ, 662, 443
1204: 
1205: %\bibitem[]{}
1206: %Guidorzi, C., et al. 2007, A\&A, 463, 539
1207: 
1208: %\bibitem[]{}
1209: %Hicken, M., \& Garnavich, P. 
1210: %                      2006, GCN Circ. 5070, http://gcn.gsfc.nasa.gov/gcn/gcn3/5070.gcn3
1211: 
1212: \bibitem[]{}
1213: Hjorth, J., et al. 2003, Nature, 423, 847
1214: 
1215: \bibitem[]{}
1216: Hjorth, J., et al. 2005, Nature, 437, 851
1217: 
1218: \bibitem[]{}
1219: Huang, K.~Y., et al. 2007, ApJ, 654, L25
1220: 
1221: \bibitem[]{}
1222: Jakobsson, P., et al. 2004, ApJ, 617, L21
1223: 
1224: \bibitem[]{}
1225: Kamble, A., Resmi, L., \& Misra, K. 2007, ApJ, 664, L5
1226: 
1227: \bibitem[]{}
1228: Kouveliotou, C., Meegan, C.~A., Fishman, G.~J., Bhat, N.~P., Briggs, M.~S., Koshut, T.~M.,
1229:      Paciesas, W.~S., \& Pendleton, G.~N. 1993, ApJ, 413, L101
1230: 
1231: \bibitem[]{}
1232: Kulkarni, S.~R. 2005, astro-ph/0510256
1233: 
1234: \bibitem[]{}
1235: Lattimer, J.~M., \& Schramm, D.~N. 1974, ApJ, 192, L145
1236: 
1237: \bibitem[]{}
1238: Lee, W.~H., \& Ramirez-Ruiz, E. 2007, New J. Phys., 9, 17
1239: 
1240: %\bibitem[]{}
1241: %Levan, A.~J., et al. 2007, GCN Circ. 6630, http://gcn.gsfc.nasa.gov/gcn/gcn3/6630.gcn3
1242: 
1243: \bibitem[]{}
1244: Li, L.-X., \& Paczy\'nski, B. 1998, ApJ, 507, L59
1245: 
1246: \bibitem[]{}
1247: Liang, E.-W., Racusin, J.~L., Zhang, B., Zhang, B.-B., 
1248:            \& Burrows, D.~N.  
1249:               2008, ApJ, 675, 528
1250: 
1251: \bibitem[]{}
1252: Liang, E.-W., Zhang, B.-B., \& Zhang, B. 2007, ApJ, 670, 565
1253: 
1254: \bibitem[]{}
1255: MacFadyen, A.~I., \& Woosley, S.~E. 1999, ApJ, 524, 262
1256: 
1257: \bibitem[]{}
1258: Malesani, D., Stella, L., Covino, S., Lidman, C., \& Naef, D. 2006,
1259:                            GCN Circ. 5705, http://gcn.gsfc.nasa.gov/gcn/gcn3/5705.gcn3
1260: 
1261: \bibitem[]{}
1262: Malesani, D., et al. 2007, GCN Circ. 6565, http://gcn.gsfc.nasa.gov/gcn/gcn3/6565.gcn3
1263: 
1264: \bibitem[]{}
1265: Malesani, D., et al. 2007, GCN Circ. 6674, http://gcn.gsfc.nasa.gov/gcn/gcn3/6674.gcn3
1266: 
1267: \bibitem[]{}
1268: Malesani, D., et al. 2007, A\&A, 473, 77
1269: 
1270: \bibitem[]{}
1271: Mangano, V. et al. 2007, A\&A, 470, 105
1272: 
1273: %\bibitem[]{}
1274: %Melandri, A., Rol, E., \& Tanvir, N.
1275: %                         2007, GCN Circ. 6602, http://gcn.gsfc.nasa.gov/gcn/gcn3/6602.gcn3
1276: 
1277: \bibitem[]{}
1278: Melandri, A., Tanvir, N., \& Guidorzi, C.
1279:                          2006, GCN Circ. 5322, http://gcn.gsfc.nasa.gov/gcn/gcn3/5322.gcn3
1280: 
1281: \bibitem[]{}
1282: M\'esz\'aros, P.,  \& Rees, M.~J. 1997, ApJ, 476, 232
1283: 
1284: \bibitem[]{}
1285: Milne, P.~A.                2006, GCN Circ. 5127, http://gcn.gsfc.nasa.gov/gcn/gcn3/5127.gcn3
1286: 
1287: \bibitem[]{}
1288: Misra, K., \& Pandey, S.~B. 2005, GCN Circ. 3396, http://gcn.gsfc.nasa.gov/gcn/gcn3/3396.gcn3
1289: 
1290: \bibitem[]{}
1291: Misra, K., et al. 2007, A\&A, 464, 903
1292: 
1293: \bibitem[]{}
1294: Mochkovitch, R., Hernanz, M., Isern, J., \& Martin, X. 1993, Nature 361, 236
1295: 
1296: \bibitem[]{}
1297: Monfardini, A., et al. 2006, ApJ, 648, 1125
1298: 
1299: \bibitem[]{}
1300: Mundell, C.~G., et al.
1301: %undell, C.~G., Gomboc, A., Guidorzi, C., Mottram, C.~J.
1302: %         Melandri, A.,   Steele, I.~A.,  Smith, R.~J.,  Monfardini, A.,  Carter, D.,
1303: %         Kobayashi, S.,  Bersier, D.,  O'Brien, P., 
1304: %    \&   Bannister, N.
1305:           2006, GCN Circ. 5700,
1306:           http://gcn.gsfc.nasa.gov/gcn/gcn3/5700.gcn3
1307: 
1308: 
1309: \bibitem[]{}
1310: Mundell, C.~G., et al. 2007, ApJ, 660, 489
1311: 
1312: 
1313: \bibitem[]{}
1314: Nakar, E., et al. 2007, Phys Reports, 442, 166
1315: 
1316: \bibitem[]{}
1317: Nava, L., Ghisellini, G., Ghirlanda, G., Tavecchio, F., 
1318:        \& Firmani, C.
1319:           2006, A\&A, 450, 471
1320: 
1321: \bibitem[]{}
1322: Nousek, J.~A., et al. 2006, ApJ, 642, 389
1323: 
1324: \bibitem[]{}
1325: Nysewander, M., et al.                     2007, astro-ph/0708.3444v2
1326: 
1327: \bibitem[]{}
1328: Nysewander, M., Fruchter, A.~S., \& Pe'er, A. 2008, astro-ph/0806.3607v1
1329: 
1330: \bibitem[]{}
1331: Oates, S.~R., et al. 2006, MNRAS, 372, 327
1332: 
1333: \bibitem[]{}
1334: Oechslin, R., Janka, H.-T., \& Marek, A. 2007, A\&A, 467, 395
1335: 
1336: \bibitem[]{}
1337: Paczy\'nski, B. 1986, ApJ, 308, L43
1338: 
1339: \bibitem[]{}
1340: Page, K.~L., et al. 2007, ApJ, 663, 1125
1341: 
1342: \bibitem[]{}
1343: Panaitescu, A. 2008, MNRAS, 383, 1143
1344: 
1345: \bibitem[]{}
1346: Panaitescu, A., et al. 2006, MNRAS, 366, 1357
1347: 
1348: \bibitem[]{}
1349: Pandey, S.~B., et al. 2006, A\&A, 460, 415
1350: 
1351: \bibitem[]{}
1352: %erley, D.~A., et al. 2007, astro-ph/0703538v2
1353: Perley, D.~A., et al. 2008, ApJ, 672, 449
1354: 
1355: \bibitem[]{}
1356: Perley, D.~A., Thoene, C. C., \& Bloom, J. S. 2007, 
1357:                    GCN Circ. 6774, http://gcn.gsfc.nasa.gov/gcn/gcn3/6774.gcn3
1358: 
1359: \bibitem[]{}
1360: Pian, E.,  et al.  2006, Nature, 442, 1011
1361: % An optical SN associated with the X-ray flash XRF 060218
1362: 
1363: \bibitem[]{}
1364: Piran, T., Kumar, P., Panaitescu, A., \& Piro, L. 2001, ApJ, 560, L167
1365: 
1366: \bibitem[]{}
1367: Press, W.~H., Flannery, B.~P., Teukolsky, S.~A., \& Vetterling, W.~T.
1368:       1986, Numerical Recipes (New York: Cambridge Univ. Press)
1369: 
1370: \bibitem[]{}
1371: Price, P.~A., et al.         2006, GCN Circ. 5077, http://gcn.gsfc.nasa.gov/gcn/gcn3/5077.gcn3
1372: 
1373: \bibitem[]{}
1374: Quimby, R., \& Rykoff, E. S. 2006, GCN Circ. 5377, http://gcn.gsfc.nasa.gov/gcn/gcn3/5377.gcn3
1375: 
1376: \bibitem[]{}
1377: Rol, E., et al. 2005, ApJ, 624, 868
1378: 
1379: \bibitem[]{}
1380: Rol, E., et al. 2007, ApJ, 669, 1098
1381: 
1382: \bibitem[]{}
1383: Roming, P.~W.~A., et al. 2005, Space Sci Rev, 120, 95
1384: 
1385: \bibitem[]{}
1386: Roming, P.~W.~A., et al. 2006, ApJ, 652, 1416
1387: 
1388: \bibitem[]{}
1389: Rosswog, S., Ramirez-Ruiz, E., \& Davies, M.~B. 2003, MNRAS, 345, 1077
1390: 
1391: \bibitem[]{}
1392: Rykoff, E.~S., et al. 2006, ApJ, 638, L5
1393: 
1394: \bibitem[]{}
1395: Sakamoto, T., et al. 2006, in ``Gamma-Ray Bursts in the {\it Swift} Era'', 
1396: ed. S.~S. Holt, N. Gehrels, and J.~A. Nousek (AIP: New York), p. 43
1397: 
1398: \bibitem[]{}
1399: %akamoto, T., et al. 2007, ApJ, submitted (astro-ph 0707.4626)
1400: Sakamoto, T., et al. 2008, ApJS, 175, 179
1401: 
1402: \bibitem[]{}
1403: Salmonson, J.~D., \& Galama, T.~J. 2002, ApJ, 569, 682
1404: 
1405: \bibitem[]{}
1406: Schlegel, D.~J., Finkbeiner, D.~P., \& Davis, M. 1998, ApJ, 500, 525
1407: 
1408: \bibitem[]{}
1409: Schmidt, B., \& Bayliss, D. 2006, 
1410:                                  GCN Circ. 4880, http://gcn.gsfc.nasa.gov/gcn/gcn3/4880.gcn3
1411: 
1412: \bibitem[]{}
1413: Schmidt, B., \& Mackie, G. 2007, GCN Circ. 6325, http://gcn.gsfc.nasa.gov/gcn/gcn3/6325.gcn3
1414: 
1415: \bibitem[]{}
1416: Shao, L., \& Dai, Z.~G. 2007, ApJ, 660, 1319
1417: 
1418: \bibitem[]{}
1419: Sharapov, D. et al. 2005,        GCN Circ. 3701, http://gcn.gsfc.nasa.gov/gcn/gcn3/3701.gcn3
1420: 
1421: \bibitem[]{}
1422: Shao, L., \& Dai, Z.~G. 2005, ApJ, 633, 1027
1423: 
1424: \bibitem[]{}
1425: Soderberg, A.~M., et al. 2006, ApJ, 650, 261
1426: 
1427: \bibitem[]{}
1428: Soderberg, A.~M., et al. 2007, ApJ, 661, 982
1429: 
1430: \bibitem[]{}
1431: Sollerman, J., et al. 2007, A\&A, 466, 839
1432: 
1433: \bibitem[]{}
1434: Soyano, T., Mito, H., \& Urata, Y. 2006, GCN Circ. 5548,
1435: http://gcn.gsfc.nasa.gov/gcn/gcn3/5548.gcn3
1436: 
1437: \bibitem[]{}
1438: Spearman, C. 1904, Am J Psychol, 15, 72
1439: 
1440: \bibitem[]{}
1441: Stanek, K.~Z., et al. 2003, ApJ, 591, L17
1442: 
1443: \bibitem[]{}
1444: Stanek, K.~Z., et al. 2007, ApJ, 654, L21
1445: 
1446: \bibitem[]{}
1447: Stefanescu, A., et al. 2006, GCN Circ. 5291, http://gcn.gsfc.nasa.gov/gcn/gcn3/5291.gcn3
1448: 
1449: \bibitem[]{}
1450: Stratta, G., Fiore, F., Antonelli, L.~A., Piro, L., 
1451:        \& de Pasquale, M. 2004, ApJ, 608, 846
1452: 
1453: \bibitem[]{}
1454: Terra, F., et al. 2007, GCN Circ., 6458, http://gcn.gsfc.nasa.gov/gcn/gcn3/6458.gcn3
1455: 
1456: \bibitem[]{}
1457: Thoene, C.~C., Fynbo, J.~P.~U., \& Jakobsson, P. 2006, GCN Circ. 5747,
1458: http://gcn.gsfc.nasa.gov/gcn/gcn3/5747.gcn3
1459: 
1460: \bibitem[]{}
1461: Thoene, C.~C., Fynbo, J.~P.~U., \& Williams, A. 2007, GCN Circ. 6389,
1462: http://gcn.gsfc.nasa.gov/gcn/gcn3/6389.gcn3
1463: 
1464: \bibitem[]{}
1465: Thoene, C.~C., Kann, D.~A., Augusteijn, T., \& Reyle-Laffont, C.
1466: 2007, GCN Circ. 6154, http://gcn.gsfc.nasa.gov/gcn/gcn3/6154.gcn3
1467: 
1468: \bibitem[]{}
1469: Uhm, Z.~L., \& Beloborodov, A.~M. 2007, ApJ, 665, L93
1470: 
1471: \bibitem[]{}
1472: van der Horst, A.~J., et al. 2008, ApJ, in prep.
1473: 
1474: %\bibitem[]{}
1475: %van Paradijs, J., Kouveliotou, C., \& Wijers, R.~A.~M.~J. 2000, ARAA, 38, 379
1476: 
1477: \bibitem[]{}
1478: Villasenor, J.~S., et al. 2005, Nature, 437, 851.
1479: 
1480: \bibitem[]{}
1481: Willingale, R., et al. 2007, ApJ, 662, 1093
1482: %> R. et al. 2007, ApJ, 662, 1093) with Liang et al.
1483: 
1484: \bibitem[]{}
1485: Woosley, S.~E., 1993, ApJ, 405, 273.
1486: 
1487: \bibitem[]{}
1488: Woosley, S.~E., \& Bloom, J.~S. 2006, ARAA, 44, 507
1489: 
1490: \bibitem[]{}
1491: Wo\'zniak, P.~R., Vestrand, W.~T., Wren, J.~A., 
1492: White, R.~R., Evans, S.~M., \& Casperson, D. 2005, ApJ, 627, L13
1493: 
1494: \bibitem[]{}
1495: Yost, S.~A., et al. 2007, ApJ, 657, 925
1496: 
1497: \bibitem[]{}
1498: Zhang, B., et al. 2006, ApJ, 642, 354
1499: 
1500: \bibitem[]{}
1501: Zhang, B., et al. 2007, ApJ, 655, 989
1502: 
1503: \bibitem[]{}
1504: %hang, Z.-B. \& Choi, C.-S. 2007, MNRAS, in press (astro-ph/0708.4049)
1505: Zhang, Z.-B., \& Choi, C.-S. 2008, A\&A, 484, 293
1506: 
1507: \bibitem[]{}
1508: Zhang, B., \& M\'esz\'aros, P. 2004, IJMPA, 19, 2385
1509: 
1510: \end{thebibliography}
1511: 
1512: 
1513: 
1514: 
1515: \vfil\eject
1516: \pagestyle{empty}
1517: \setlength{\voffset}{20mm}
1518: \LongTables
1519: \clearpage
1520: \begin{landscape}
1521: \voffset=-.235truein
1522: \input{tab1.tex}
1523: \clearpage
1524: \end{landscape}
1525: 
1526: 
1527: 
1528: 
1529: \vfil\eject
1530: \pagestyle{empty}
1531: \setlength{\voffset}{20mm}
1532: \LongTables
1533: \clearpage
1534: \begin{landscape}
1535: %vsize=10.5truein
1536: \voffset=-.235truein
1537: \input{tab2.tex}
1538: \clearpage
1539: \end{landscape}
1540: 
1541: 
1542: 
1543: \vfil\eject
1544: \clearpage
1545: \pagestyle{empty}
1546: \setlength{\voffset}{-15mm}
1547: \LongTables
1548: \clearpage
1549: %\begin{landscape}
1550: \input{tab3.tex}
1551: \clearpage
1552: %\end{landscape}
1553: 
1554: 
1555: \vfil\eject
1556: \clearpage
1557: \pagestyle{empty}
1558: \setlength{\voffset}{20mm}
1559: \input{tab4.tex}
1560: \clearpage
1561: 
1562: 
1563: 
1564: 
1565: 
1566: 
1567: \end{document}
1568: 
1569: 
1570: