1: % v2 Scott's comments (07/10/07)
2: % v3 Craig's comments (07/17/07)
3: % v4 Neil and Hans' comments (07/19/07)
4: % v5 Bing's Swift internal review (07/26/07)
5: % revised based on referee's comments
6:
7: \def\liso{L_{\rm iso}}
8: \def\eiso{E_{\rm iso}}
9: \def\egamma{E_{\gamma}}
10: \def\eop{E^{\rm obs}_{\rm peak}}
11: \def\esp{E^{\rm src}_{\rm peak}}
12: \def\kw{$Konus$-$Wind$ }
13: \def\ep{E_{\rm peak}}
14: \def\hete{{\it HETE$-$2 }}
15:
16: \documentclass[12pt,preprint]{aastex}
17:
18: %% preprint2 produces a double-column, single-spaced document:
19:
20: %\documentclass[preprint2]{aastex}
21:
22: %% Sometimes a paper's abstract is too long to fit on the
23: %% title page in preprint2 mode. When that is the case,
24: %% use the longabstract style option.
25:
26: %% \documentclass[preprint2,longabstract]{aastex}
27:
28: %% If you want to create your own macros, you can do so
29: %% using \newcommand. Your macros should appear before
30: %% the \begin{document} command.
31: %%
32: %% If you are submitting to a journal that translates manuscripts
33: %% into SGML, you need to follow certain guidelines when preparing
34: %% your macros. See the AASTeX v5.x Author Guide
35: %% for information.
36:
37: %\newcommand{\vdag}{(v)^\dagger}
38: %\newcommand{\myemail}{hullingerd@byui.edu}
39:
40: %% You can insert a short comment on the title page using the command below.
41:
42: %\slugcomment{Not to appear in Nonlearned J., 45.}
43:
44: %% If you wish, you may supply running head information, although
45: %% this information may be modified by the editorial offices.
46: %% The left head contains a list of authors,
47: %% usually a maximum of three (otherwise use et al.). The right
48: %% head is a modified title of up to roughly 44 characters.
49: %% Running heads will not print in the manuscript style.
50:
51: \shorttitle{BAT catalog}
52: \shortauthors{Sakamoto et al.}
53:
54: %% This is the end of the preamble. Indicate the beginning of the
55: %% paper itself with \begin{document}.
56:
57: \begin{document}
58:
59: %% LaTeX will automatically break titles if they run longer than
60: %% one line. However, you may use \\ to force a line break if
61: %% you desire.
62:
63: \title{The First {\it Swift} BAT Gamma-Ray Burst Catalog}
64:
65: %% Use \author, \affil, and the \and command to format
66: %% author and affiliation information.
67: %% Note that \email has replaced the old \authoremail command
68: %% from AASTeX v4.0. You can use \email to mark an email address
69: %% anywhere in the paper, not just in the front matter.
70: %% As in the title, use \\ to force line breaks.
71:
72: \author{
73: T. Sakamoto\altaffilmark{1,2},
74: S. D. Barthelmy\altaffilmark{1},
75: L. Barbier\altaffilmark{1},
76: J. R. Cummings\altaffilmark{3,10},
77: E. E. Fenimore\altaffilmark{4},
78: N. Gehrels\altaffilmark{1},
79: D. Hullinger\altaffilmark{8},
80: H. A. Krimm\altaffilmark{6,10},
81: C. B. Markwardt\altaffilmark{5,10},
82: D. M. Palmer\altaffilmark{4},
83: A. M. Parsons\altaffilmark{1},
84: G. Sato\altaffilmark{1,7},
85: M. Stamatikos\altaffilmark{1,2},
86: J. Tueller\altaffilmark{1},
87: T. N. Ukwatta\altaffilmark{1,9},
88: B. Zhang\altaffilmark{11}
89: }
90:
91: \altaffiltext{1}{NASA Goddard Space Flight Center, Greenbelt, MD 20771}
92: \altaffiltext{2}{Oak Ridge Associated Universities, P.O. Box 117,
93: Oak Ridge, Tennessee 37831-0117}
94: \altaffiltext{3}{Joint Center for Astrophysics, University of Maryland,
95: Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250}
96: \altaffiltext{4}{Los Alamos National Laboratory, P.O. Box 1663, Los
97: Alamos, NM, 87545}
98: \altaffiltext{5}{Department of Astronomy, University of Maryland,
99: College Park, MD 20742}
100: \altaffiltext{6}{Universities Space Research Association, 10211 Wincopin
101: Circle, Suite 500, Columbia, MD 21044-3432}
102: \altaffiltext{7}{Institute of Space and Astronautical Science,
103: JAXA, Kanagawa 229-8510, Japan}
104: \altaffiltext{8}{Moxtek, Inc., 452 West 1260 North, Orem, UT 84057}
105: \altaffiltext{9}{Center for Nuclear Studies, Department of Physics,
106: The George Washington University, Washington, D.C. 20052}
107: \altaffiltext{10}{Center for Research and Exploration in Space Science
108: and Technology (CRESST), NASA Goddard Space Flight Center, Greenbelt, MD
109: 20771}
110: \altaffiltext{11}{Department of Physics and Astronomy, University of Nevada,
111: Las Vegas, NV 89154}
112: %% Notice that each of these authors has alternate affiliations, which
113: %% are identified by the \altaffilmark after each name. Specify alternate
114: %% affiliation information with \altaffiltext, with one command per each
115: %% affiliation.
116:
117: %\altaffiltext{1}{Department of Physics, University of Maryland,
118: %College Park, MD 20742}
119: %\altaffiltext{2}{NASA Goddard Space Flight Center, Greenbelt, MD 85719}
120:
121: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
122: %% style, abstract will output a Received/Accepted line after the
123: %% title and affiliation information. No date will appear since the author
124: %% does not have this information. The dates will be filled in by the
125: %% editorial office after submission.
126:
127: \begin{abstract}
128: We present the first {\it Swift} Burst Alert Telescope (BAT)
129: catalog of gamma-ray bursts (GRBs), which contains bursts detected by the BAT
130: between 2004 December 19 and 2007 June 16. This catalog (hereafter BAT1
131: catalog)
132: contains burst trigger time, location, 90\% error radius, duration,
133: fluence, peak
134: flux, and time averaged spectral parameters for each of 237 GRBs, as measured
135: by the BAT. The BAT-determined position reported here is within 1.75$^{\prime}$ of
136: the Swift X-ray Telescope (XRT)-determined position for 90\% of these GRBs.
137: The BAT T$_{90}$ and T$_{50}$ durations peak at 80 and 20 seconds, respectively.
138: From the fluence-fluence correlation, we conclude that about 60\% of the
139: observed peak energies, $\eop$, of BAT GRBs could be less than
140: 100 keV. We confirm that GRB fluence to hardness and GRB peak flux to hardness
141: are correlated for BAT bursts in analogous ways to previous missions' results.
142: The correlation between the photon index in a simple power-law model
143: and $\eop$ is also confirmed. We also report the
144: current status for the on-orbit BAT calibrations based on observations of the
145: Crab Nebula.
146:
147: %We present the first 237 gamma-ray burst (GRB) catalog (BAT1 catalog)
148: %of the {\it Swift}
149: %Burst Alert Telescope (BAT) between 2004 December 19 and 2007 June 16,
150: %based on the analysis of the BAT event data for 237 GRBs.
151: %We present the tables of burst trigger time,
152: %locations, 90\% error radius, durations, fluences, peak fluxes, and the
153: %time-averaged spectral parameters. The BAT ground position is
154: %less than 1.75$^{\prime}$ from the position reported by the {\it Swift}
155: %X-ray Telescope for 90\% of GRBs. The BAT T$_{90}$ and T$_{50}$ durations
156: %for the long GRBs are peaked around 80 s and 20 s.
157: %Based on the fluence-fluence correlation, we concluded that about 60\% of the
158: %peak energies, $\eop$, of the BAT GRBs could be less than 100 keV.
159: %We confirmed the correlations in
160: %the fluence - hardness and the peak flux - hardness for the BAT GRBs.
161: %The correlation between the photon index in a simple power-law model
162: %and the peak energy, $\eop$, is also confirmed.
163: %We also present the current status of the in-orbit calibration of the
164: %BAT instrument based on the Crab observations.
165: %(v5 T.S. 07/26/07)
166: \end{abstract}
167:
168: %% Keywords should appear after the \end{abstract} command. The uncommented
169: %% example has been keyed in ApJ style. See the instructions to authors
170: %% for the journal to which you are submitting your paper to determine
171: %% what keyword punctuation is appropriate.
172:
173: \keywords{gamma rays: bursts}
174: % Maybe I'll come up with more later
175:
176: %% From the front matter, we move on to the body of the paper.
177: %% In the first two sections, notice the use of the natbib \citep
178: %% and \citet commands to identify citations. The citations are
179: %% tied to the reference list via symbolic KEYs. The KEY corresponds
180: %% to the KEY in the \bibitem in the reference list below. We have
181: %% chosen the first three characters of the first author's name plus
182: %% the last two numeral of the year of publication as our KEY for
183: %% each reference.
184:
185:
186: %% Authors who wish to have the most important objects in their paper
187: %% linked in the electronic edition to a data center may do so by tagging
188: %% their objects with \objectname{} or \object{}. Each macro takes the
189: %% object name as its required argument. The optional, square-bracket
190: %% argument should be used in cases where the data center identification
191: %% differs from what is to be printed in the paper. The text appearing
192: %% in curly braces is what will appear in print in the published paper.
193: %% If the object name is recognized by the data centers, it will be linked
194: %% in the electronic edition to the object data available at the data centers
195: %%
196: %% Note that for sources with brackets in their names, e.g. [WEG2004] 14h-090,
197: %% the brackets must be escaped with backslashes when used in the first
198: %% square-bracket argument, for instance, \object[\[WEG2004\] 14h-090]{90}).
199: %% Otherwise, LaTeX will issue an error.
200:
201: \section{Introduction}
202:
203: The {\it Swift} mission \citep{gehrels2004} has revolutionized our
204: understanding of gamma-ray bursts (GRBs). Because of the sophisticated
205: on-board localization capability of the {\it Swift} Burst Alert Telescope
206: (BAT; \citet{barthelmy}) and the fast spacecraft pointing of {\it Swift},
207: more than 90\% (30\%) of {\it Swift} GRBs have an X-ray (optical)
208: afterglow observation from the {\it Swift} X-Ray Telescope (XRT;
209: \citet{burrows}) or the {\it Swift} UV/Optical Telescope (UVOT; \citet{roming})
210: within a few hundred seconds after the trigger. ${\it Swift}$ opens
211: a new opportunity to study the host galaxies and the distance scale
212: of the mysterious short duration GRBs.\footnote{The short burst GRB 050709,
213: which observed and promptly localized by {\it HETE-2} is also a good
214: example of a short GRB observation \citep{villasenor2005}.}
215: \citep[e.g.,][]{gehrels2005,barthelmy2005b} {\it Swift} allows us to use
216: GRBs as a tool for investigation of the early universe
217: (cf. detection of GRB 050904 at a redshift of 6.29; \citet{cusumano2006}).
218: {\it Swift} found the fourth GRB, GRB 060218, which is securely associated with
219: a supernova \citep{campana2006}.
220: %% referee's comment #1
221: Furthermore, {\it Swift} is providing X-ray
222: afterglow data \citep[e.g.,][]{zhang2006} which gives us insight into details
223: of the fireball model and to the nature of the central engine.
224: %%
225:
226: Here we present the first BAT GRB catalog including 237 GRBs detected
227: by BAT from 2004 December 19 to 2007 June 16. In \S 2, we describe
228: the BAT instrument. In \S 3, we show the current
229: status of the on-orbit calibration of the BAT based on the Crab observations.
230: In \S 4, we describe the analysis methods for the catalog. In \S 5, we
231: describe the content of the tables of the catalog and show the results of the
232: prompt emission properties of the BAT GRBs based on the catalog. Our
233: conclusions are summarized in \S 6. All quoted errors in this work
234: are at the 90\% confidence level.
235:
236: \section{Instrumentation}
237:
238: The BAT is a highly sensitive, large field of view (FOV) (1.4 sr for $>$50\%
239: coded FOV and 2.2 sr for $>$10\% coded FOV), coded-aperture
240: telescope, which detects and localizes GRBs in real time. The fast and accurate
241: BAT GRB positions with 1-3 arc-minute error radii are the key to
242: autonomously slewing the
243: spacecraft to point the XRT and the UVOT. The BAT GRB position, light
244: curves, and the detector plane image (BAT scaled map) are transmitted through
245: TDRSS to the ground within 20--200 s after the burst trigger.
246: The BAT detector plane is
247: composed of 32,768 pieces of CdZnTe (CZT: $4 \times 4 \times 2$ mm), and the
248: coded-aperture mask is composed of $\sim$ 52,000 lead tiles
249: ($5 \times 5 \times 1$ mm) with a 1-m separation between mask and
250: detector plane. The energy range is 15-150 keV for imaging or mask
251: weighting\footnote{Mask weighting is a
252: background subtraction technique based on the modulation resulting from
253: the coded mask.} with a non-coded response up to 350 keV.
254:
255: The CZT pixels are biased to $-200$ V with a nominal
256: operating temperature of 20$^{\circ}$C. The energy scale calibration
257: is performed automatically on the front end electronics (XA1) by injecting
258: calibration pulses. This electronic calibration task is executed every $\sim$
259: 5000 s during spacecraft slews. In addition to the electronic
260: calibration, there are two $^{241}$Am tagged sources mounted below the mask
261: for calibrating the absolute energy scale and the detector efficiency
262: for each CZT pixel in-flight.
263:
264: There are three types of triggers in the BAT flight software. Two of these
265: are rate triggers looking for excesses in the count rate from the
266: background, and one is an image trigger based on new significant sources
267: found in the sky images.
268: The rate triggers are divided into short
269: (foreground period $\le$ 64 ms) and long rate triggers (foreground period
270: $\ge$ 64 ms). Each trigger criterion is a specific combination
271: of choices of foreground interval, number of background samples, energy
272: band and one of nine different regions of the detector plane. Currently,
273: 494 trigger criteria have been running on-board for a long rate trigger, 36 for the
274: short rate trigger, and 1 for the image trigger. Once a rate trigger
275: has occurred, the BAT creates a sky image using the triggered foreground
276: and background intervals in the specified energy band to find a significant
277: source in the image. Failing to produce a significant image excess, the BAT will
278: check for trigger criteria that produce a more statistically significant image.
279: When a rate or image trigger finds a significant source in the image,
280: a location match to the on-board source catalog is executed to
281: exclude activity from known hard X-ray astronomical sources.\footnote
282: {If the significance of the activity of the known source is higher
283: than the threshold value set in the on-board catalog, the activity
284: will be reported to the GCN in real time with a different GCN notice type.}
285:
286: For each GRB trigger, the photon-by-photon data (event data) are
287: available with a time resolution of 100 $\mu$s. The duration of
288: the event data was initially from T-300 s to T+300 s (T as the BAT
289: trigger time). After March 17 2006, we have extended the duration
290: of the event data which are downlinked to from T-300 s to T+1000 s.
291: We also transmit 10 s
292: of the event data for failed triggers.\footnote{Failed triggers are those
293: which caused a rate trigger, but did not also have a significant point source
294: in the image formed with that data; or those for which the image coincided with
295: a known source in the on-board catalog.} In the survey mode,
296: the BAT produces detector plane histograms (DPHs). These histograms
297: have an 80 channel spectrum for each detector integrated typically over
298: five minutes.\footnote{The integration time of DPHs changes in various
299: operational conditions, but is 300 s for most of the
300: on-orbit operation time.} The DPHs are the primary data product when
301: the GRB prompt emission lasts longer than the stop time of the event
302: data collection \citep[e.g. GRB 060124;][]{romano2006}.
303:
304: \section{On-orbit calibration}
305:
306: The Crab nebula data collected for various positions in the BAT field of
307: view were used for the on-orbit calibration. We analyzed the
308: DPH data for this purpose. The standard BAT software (HEASoft 6.2) and the latest
309: calibration database (CALDB: 2006-10-14) were used for processing the data.
310: We first made the Good Time Interval (GTI) file
311: for each observation segment (each observation ID) excluding the periods
312: when 1) the spacecraft was not settled, 2) the spacecraft was in the South Atlantic
313: Anomaly (SAA), and 3) the Crab was occulted by the earth,
314: moon and/or sun.
315: {\tt batoccultgti} was used for excluding the time periods for
316: case 3.
317: {\tt baterebin} was applied to the DPH data to correct the energy scale.
318: %% referee's comment #2
319: %The individual rows in the DPH data were processed separately to reduce the systematic
320: %uncertainty of the spacecraft attitude.
321: %%
322: A Detector Plane Image (DPI)
323: file was created from the DPH using {\tt batbinevt} for each individual
324: row. The spacecraft attitude file was re-created using ftools {\tt aspect} by
325: specifying the observation start and stop time of the DPI. The detector
326: enable/disable map was created using {\tt bathotpix}. The BAT sky image
327: was created by {\tt batfftimage} using the DPI, the updated attitude file and
328: the enable/disable map. {\tt batcelldetect} was used to extract the position of
329: the Crab. The mask weight map of the Crab was created by {\tt batmaskwtimg}
330: using the ``true'' Crab position from SIMBAD (R.A.$_{\rm J2000}$ = 83.6332,
331: Dec.$_{\rm J2000}$ = 22.0144).
332: The spectrum (PHA file) was created by {\tt batbinevt} using the mask weight map for each
333: row of the DPH. All of the individual PHA files at the same sky coordinate
334: were added to create a single PHA file if the Crab was detected $>8 \sigma$ in
335: the image in the full energy band (15--350 keV). The detector energy response
336: file was created by {\tt batdrmgen} for each summed PHA file.
337:
338: \subsection{Position Accuracy}
339:
340: The histograms of the angular differences between the Crab detected position by
341: {\tt batcelldetect} and the ``true'' Crab position (the Crab position in SIMBAD)
342: are shown in the top of Figure \ref{fig:crab_pos_accuracy}. The
343: position differences are less than 1$^{\prime}$ for 95\% of the Crab
344: observations in both the right ascension (R.A.) and the declination (Dec.).
345: The bottom of Figure \ref{fig:crab_pos_accuracy} shows the BAT position errors as a
346: function of signal to noise ratio for the Crab. The signal to noise ratio
347: and the position error are correlated with a power-law index of $-0.7$ (see section 5).
348:
349: \subsection{Energy Response}
350:
351: Immediately after the first attempt to fit the Crab spectrum with the pre-launch
352: detector energy response matrices (DRM), we noticed that there were systematic
353: errors in the pre-launch DRM at low energies (below 25 keV) and also at
354: high energies (above 80 keV). The investigation of these problems is still
355: in progress. To overcome these problems, we applied corrections to force
356: the Crab to fit a canonical model, a power-law with a photon index of 2.15 and
357: a 15-150 keV energy flux of $2.11 \times 10^{-8}$ ergs cm$^{-2}$ s$^{-1}$
358: \citep[e.g.,][]{jung1989,rothschild1998,sax1999,hete2003,integral2005}.
359: Due to these corrections,
360: the BAT team has released the software tool, {\tt batphasyserr}, and the
361: CALDB file (swbsyserr20030101v002.fits) to apply the energy dependent
362: systematic errors to the PHA file. The systematic errors which should be
363: applied to the PHA file are shown in Figure \ref{fig:sysvector}. The Crab
364: spectra were fitted by a simple power-law model using {\tt Xspec 11.3.2}
365: including the systematic errors.
366:
367: Figure \ref{fig:crab_phindex_flux_theta} shows the Crab photon index and the
368: flux in the 15-150 keV band as a function of the incident angle. In the current
369: BAT DRM ({\tt batdrmgen v3.3} and CALDB: 20061014), the scatter of the
370: photon index and the flux are about 5\% and 10\% of the canonical
371: values.
372: Figure \ref{fig:2d_crab_phindex_flux} shows the contour maps of the Crab photon
373: index and the flux in the 15-150 keV band over the BAT field of view.
374: We note that the parameters tend to deviate
375: from the canonical values towards
376: the edge of the BAT field of view. Thus, the spectral
377: parameters could have a larger systematic error when the source is
378: at the edge of the field of view of BAT.
379:
380: \subsection{BAT GRB Response Time}
381:
382: Figure \ref{fig:bat_pos_notice_time} shows the histogram of the time delay
383: between the BAT GRB trigger time and the GCN BAT Position Notice.\footnote{This GRB sample
384: is from GRB 050215A to GRB 070616 (slightly reduced from the rest of the paper)
385: excluding the GRBs found on the ground process.}
386: The highest peak of the distribution is around 15 s. The BAT position has been
387: reported on the ground within 30 s after the burst trigger for half of the BAT GRBs.
388: Most of the longer delays ($>$300 s) are due to interruptions in TDRSS transmissions
389: during regular telemetry down links to the Malindi ground station.
390:
391: \section{Analysis for the GRB catalog}
392:
393: We used the standard BAT software (HEASoft 6.1.1) and the latest
394: calibration database (CALDB: 20061014) to process the BAT GRBs
395: from December 2004 (GRB 041217) to June 2007 (GRB 070616).\footnote{The GRB sample
396: includes bursts which were found in the ground processing.}
397: The burst pipeline script,
398: {\tt batgrbproduct}, was used to process the BAT event data.\footnote{By default, the minimum partial coding setting was 10\% to
399: remove portions of the light curve with poor sampling, and the
400: aperture setting was CALDB:FLUX in order to avoid passive materials
401: in the BAT field of view. Some bursts were initially in the extreme
402: partial coded field of view ($<$10\%). In those cases, we
403: re-ran {\tt batgrbproduct} specifying the options
404: pcodethresh=0.0 and aperture=CALDB:DETECTION.}
405: The {\tt Xspec} spectral fitting tool (version 11.3.1) was used to fit
406: each spectrum.
407: Since our analysis is restricted to use only the event data,
408: we present partial analysis based on the available
409: event data for bursts which last longer than the end period of the event data
410: (e.g. GRB 060124) or which have incomplete event data due to the various
411: reasons (e.g. GRB 050507). In some cases, especially for weak short GRBs, {\tt battblocks},
412: which is one
413: of the task run in {\tt batgrbproduct}, might fail to find the burst interval.
414: In those cases,
415: we fitted the mask-weighted light curve in the full BAT energy range by a
416: liner-rise exponential decay model (``BURS'' model in ftools
417: qdp\footnote{http://heasarc.gsfc.nasa.gov/docs/software/ftools/others/qdp/qdp.html})
418: to find the burst time intervals (T$_{100}$, T$_{90}$, T$_{50}$ and peak
419: 1-s intervals) and created the T$_{100}$ and peak 1-s PHA files based on
420: these time intervals. We put comments on Table \ref{tbl:bat_summary} for
421: the bursts which have a problem in either the data or the processing.
422:
423: For the time-averaged spectral analysis, we use the time interval from
424: the emission start time to the emission end time (T$_{100}$ interval).
425: Since the BAT energy response generator, {\tt batdrmgen}, performs the
426: calculation for a fixed single incident angle of the source, it will be
427: a problem if the position of the source is moving during the time interval
428: selected for the spectral analysis because of the spacecraft slew.
429: In this situation, we created the DRMs for each five-second
430: period during the time interval taking into account the position
431: of the GRB in detector coordinates. We then weighted these DRMs
432: by the five-second count rates and created the averaged DRM using {\tt addrmf}.
433: Since the spacecraft slews about one degree per second in response to a GRB
434: trigger, we chose five second intervals to calculate
435: the DRM for every five degrees.
436:
437: The spectrum was fitted by a simple power-law (PL) model,
438: \begin{eqnarray}
439: f(E) = K_{50}^{\rm PL}\left(\frac{E}{50 \: {\rm keV}}\right)^{\alpha^{\rm PL}}
440: \label{eq:pl}
441: \end{eqnarray}
442: where $\alpha^{\rm PL}$ is the power-law photon index and $K_{50}^{\rm
443: PL}$ is the normalization at 50 keV in units of
444: photons cm$^{-2}$ s$^{-1}$ keV$^{-1}$, and by a cutoff power-law (CPL) model,
445: \begin{eqnarray}
446: f(E) = K_{50}^{\rm CPL}\left(\frac{E}{50 \: {\rm keV}}\right)^{\alpha^{\rm CPL}}
447: \exp\left(\frac{-E\,(2+\alpha^{\rm CPL})}{\ep}\right)
448: \label{eq:cpl}
449: \end{eqnarray}
450: where $\alpha^{\rm CPL}$ is the power-law photon index, $\ep$ is the
451: peak energy in the $\nu$F$_{\nu}$ spectrum and $K_{50}^{\rm CPL}$ is the
452: normalization at 50 keV
453: in units of photons cm$^{-2}$ s$^{-1}$ keV$^{-1}$. We also systematically
454: fitted the spectrum with the Band function \citep{band1993}. However, none of the
455: BAT spectra show a significant improvement in $\chi^{2}$ with
456: a Band function fit compared to that of a CPL model fit.
457: %
458: Note that this is equivalent to saying that a CPL model and a Band function
459: represent equally well the observed spectrum. However, we only present
460: the results based on
461: a CPL model throughout the paper due to its simplicity in the functional
462: form.\footnote{A Band function has 4 parameters, whereas, a CPL model
463: has 3 parameters.}
464: %
465: The best fit spectral model is determined
466: based on the difference in $\chi^{2}$ between a PL and a CPL fit. If
467: $\Delta$ $\chi^{2}$ between a PL and a CPL fit is greater than 6 ($\Delta \chi^{2}
468: \equiv \chi^{2}_{\rm PL} - \chi^{2}_{\rm CPL}$ $>$ 6), we determined that a
469: CPL model is a better
470: representative spectral model for the data. To quantify the significance of this
471: improvement, we performed 10,000 spectral simulations taking into account the
472: distributions of the power-law photon index in a PL fit, the fluence in the 15-150 keV band
473: in a PL fit and the T$_{100}$ duration of the BAT GRBs, and determined
474: how many cases a CPL fit gives
475: $\chi^{2}$ improvements of equal or greater than 6 over a PL fit. The BAT
476: DRM used in the simulation was for an incident angle of 30$^{\circ}$
477: which was an averaged incident angle of the BAT GRB sample
478: (see Figure \ref{fig:bat_theta}). We found equal or higher
479: improvements in $\chi^{2}$ in 62 simulated spectra out of 10,000. Thus,
480: the chance
481: probability of having an equal or higher $\Delta \chi^{2}$ of 6 with a CPL model when
482: the parent distribution is a case of a PL model is 0.62\%.
483:
484: The fluence and the peak flux are derived from the spectral fits. The
485: fluences are calculated fitting the time-averaged spectrum by the best
486: fit spectral model. The peak fluxes are calculated fitting the spectrum
487: of the one-second interval bracketing the highest peak in the light
488: curve (hereafter peak spectrum). Again, we used the best fit spectral
489: model for calculating the peak fluxes. To correctly reflect the incident
490: angle of the source during the period of the peak spectrum, the DRM
491: for the peak spectrum was created updating the keywords of the peak
492: spectrum file by {\tt batupdatephakw} and running {\tt batdrmgen} using
493: this updated spectral file.
494:
495: \section{The Catalog}
496:
497: The first BAT catalog includes the time period between 2004 December
498: 19 and 2007 June 16. The total number of GRBs including five untriggered
499: GRBs\footnote{GRBs which found in ground processing.} and four
500: possible GRBs is 237. 237 GRBs are listed in Table
501: \ref{tbl:bat_summary}. The first column is the GRB name. The next column
502: is the BAT trigger number. The next column specifies the BAT trigger time
503: in UTC in the form of {\it YYYY-MM-DD hh:mm:ss} where {\it YYYY}
504: is year, {\it MM} is month, {\it DD} is day of month, {\it hh} is hour,
505: {\it mm} is minute, and {\it ss} is second. Note that the definition
506: of the BAT trigger time is the start time of the foreground time interval
507: of the image from which the GRB is detected on-board. The next four columns give
508: the locations by the ground process in equatorial (J2000) coordinate,
509: the signal-to-noise ratio of the BAT image at the location, and the radius
510: of the 90\% confidence region in arcmin. The 90\% error radius is calculated
511: based on the signal-to-noise ratio of the image using the following equation
512: which derived from the BAT hard X-ray survey
513: process\footnote{http://heasarc.gsfc.nasa.gov/docs/swift/analysis/bat\_digest.html}$^{,}$\footnote{SWIFT-BAT-CALDB-CENTROID-v1 \\
514: http://swift.gsfc.nasa.gov/docs/heasarc/caldb/swift/docs/bat/index.html}
515: \begin{displaymath}
516: r_{90\%} = 10.92 \times {\rm SNR}^{-0.7}\;({\rm arcmin}),
517: \end{displaymath}
518: where SNR is the signal-to-noise ratio of the BAT image. However, due to the
519: limitation of the BAT point spread function, we decided to quote the minimum
520: allowed value of $r_{90\%}$ as 1$^{\prime}$ in the catalog.
521: The next two columns specify the
522: burst durations which contain from 5\% to 95\% (T$_{90}$) and from 25\% to
523: 75\% (T$_{50}$) of the total burst fluence. These durations are calculated in
524: the 15--350 keV band.\footnote{The coded mask is transparent to photons above 150 keV.
525: Thus, photons above 150 keV are treated as background in the mask-weighted
526: method. The effective upper boundary is $\sim$ 150 keV.} The next two columns are
527: the start and stop time from the BAT trigger time of the event data. The last column is
528: the comments.
529:
530: The energy fluences calculated in various energy bands are summarized in
531: Table \ref{tbl:bat_fluence}. The first column is the GRB name. The next column specifies
532: the spectral model which used in deriving the fluences (PL: simple power-law model;
533: Eq.(\ref{eq:pl}), CPL: cutoff power-law model; Eq.(\ref{eq:cpl})). The next five columns
534: are the fluences in the 15-25 keV, the 25-50 keV, the 50-100 keV, the 100-150 keV, and the
535: 15-150 keV band. The unit of the fluence is 10$^{-8}$ ergs cm$^{-2}$.
536: The last two columns specify the start and the stop time from the
537: BAT trigger time which used to calculate the fluences. Note that since our analysis is
538: based on the available event data, 6 bursts with the incomplete data (see the 12th column
539: of Table \ref{tbl:bat_summary}) might not include the whole burst emission.
540:
541: Table \ref{tbl:bat_phflux} and \ref{tbl:bat_peakeneflux}
542: summarize the 1-s peak photon and energy fluxes in various energy
543: bands. The first column is the GRB name. The next column specifies
544: the spectral model used in deriving the 1-s peak flux. The next five column
545: are the 1-s peak photon and energy fluxes in the 15-25 keV, the 25-50 keV, the 50-100 keV,
546: the 100-150 keV, and the 15-150 keV band. The unit of the flux is
547: photons cm$^{-2}$ s$^{-1}$ for the peak photon flux and 10$^{-8}$ ergs cm$^{-2}$ s$^{-1}$
548: for the peak energy flux.
549: The last two columns specify the start and the stop time from the BAT trigger time
550: which were used to calculate the peak fluxes.
551:
552: The time-averaged spectral parameters are listed in Table \ref{tbl:bat_spec}.
553: The first column is the GRB name. The next three columns are the photon index,
554: the normalization at 50 keV and $\chi^{2}$ of the fit for a PL model. The degree of
555: freedom is 57 for all bursts in a PL fit. The next four columns are the photon
556: index, $\eop$, the normalization at 50 keV and $\chi^{2}$ of the fit in a CPL model.
557: The degree of freedom is 56 for all bursts for a CPL fit. The spectral parameters in
558: a CPL are only shown for the bursts which meet the criteria described in the section 4.
559:
560: In the following subsections of investigating the relationship among fluences,
561: peak fluxes, and the spectral parameters, we excluded 10 GRBs based on an incomplete
562: data set and also those labeled as possible GRBs/SGRs
563: (see the 12th column of Table \ref{tbl:bat_summary}).
564:
565: \subsection{BAT GRB Position and Sky Locations}
566:
567: Figure \ref{fig:bat_xrt_pos_diff} shows the angular difference between the
568: BAT ground position and the first reported XRT position.\footnote{The XRT position
569: is either from the on-board or from the XRT event-by-event data downloaded from the
570: TDRSS satellite (Single-Pixel-Event-Report).} The BAT ground position is within
571: 0.95$^{\prime}$ and 1.75$^{\prime}$ from the XRT position for 68\% and 90\% of the
572: bursts, respectively. The distribution of the incident angles ($\theta$) is shown in Figure
573: \ref{fig:bat_theta}. The $\theta$ distribution is peaked around 30$^{\circ}$ with a spread
574: from 0$^{\circ}$ to 60$^{\circ}$. The sky map of the BAT 237 GRBs in galactic
575: coordinates is shown in Figure \ref{fig:bat_sky_map}.
576:
577: \subsection{Durations and Hardness}
578:
579: The histograms of T$_{90}$ and T$_{50}$ measured by the mask weighted
580: light curve in the BAT full energy band
581: are shown in Figure \ref{fig:bat_t90_t50}. The BAT T$_{90}$ and T$_{50}$
582: durations are peaked around 80 s and 20 s respectively. The BAT duration distributions
583: do not show the clear bimodality seen in the BATSE sample \citep[e.g.,][]{kouveliotou1993}
584: due to the smaller samples of short duration bursts (hereafter short GRBs).
585: This is a well-known
586: selection effect for an imaging GRB instrument like the BAT. For most of the GRB imaging
587: instruments, two triggering processes have to be met to be determined a GRB trigger.
588: The first step is the increase in the count rate from the background level, and
589: the second step is the significant signal in the image. Although
590: the short duration bursts would have triggered as an excess in the count rate,
591: it could be very
592: difficult to meet the criterion for a significant signal in the image
593: because of the smaller number of photons
594: to do the imaging compared to those of the long duration bursts (hereafter long GRBs).
595: However, because of the
596: large effective area and also the sophisticated flight software, the BAT has
597: been triggering and localizing short GRBs in a much higher
598: fraction than other GRB imaging instruments (e.g. {\it Beppo}SAX, {\it HETE-2}).
599:
600: Figure \ref{fig:bat_t90_t50_vs_s23} shows the T$_{50}$ and T$_{90}$ durations
601: versus the fluence ratio between the 50-100 keV and the 25-50 keV band. Although
602: the short GRBs tend to be systematically harder than the long GRBs,
603: the separation in the hardness between these two classes is not
604: obvious, at least in the BAT GRB sample. Note that there are several
605: works that question the hardness of the BATSE short GRBs
606: reported on the BATSE catalog \citep{sakamoto2006,ohno2007,donaghy2007}.
607: Therefore, as mentioned in \citet{donaghy2007} and also seen in the BAT
608: short GRB samples,
609: %% referee's comment #4
610: the duration and the
611: hardness are insufficient to resolve the long and short populations,
612: since the two components have so much overlap.
613: %not enough criteria to distinguish between long and short GRBs.
614: %%
615: \subsection{Peak Fluxes and Fluences}
616:
617: Figure \ref{fig:bat_peakflux_fluence} shows the distribution of
618: the fluence versus the peak photon flux in the 15-150 keV band.
619: The positive correlation is seen in these two parameters (correlation
620: coefficient of +0.912 in a 220 burst sample).
621: The peak photon flux of about 70\% of the BAT GRBs
622: is less than 2 photons cm$^{-2}$ s$^{-1}$ in the 15-150 keV band. If we
623: assume a GRB detector with a sensitivity of 1 photons
624: cm$^{-2}$ s$^{-1}$ in the 50-300 keV band (trigger band of BATSE),
625: the peak photon flux in the 15-150 keV band is 2.5 photons
626: cm$^{-2}$ s$^{-1}$ assuming the Band function parameters of $\alpha = -1$,
627: $\beta = -2.5$, and $\eop = 100$ keV. Thus, the majority of the BAT GRBs
628: might be too weak to trigger a typical GRB detector which is sensitive at
629: $>$50 keV. This rough estimate
630: is consistent with the observation that only 20-30\% of the BAT GRBs are
631: simultaneously triggered successfully by the currently active GRB detectors
632: such as ${\it Konus}$-${\it Wind}$ and ${\it Suzaku/WAM}$.
633:
634: Figure \ref{fig:s12_s34} shows the fluence in the 15-50 keV band versus that in
635: the 50-150 keV band. The blue dash-dotted line is the case of the Band
636: function parameters of $\alpha = -1$, $\beta = -2.5$, and $\eop = 100$ keV.
637: The blue dotted line is the case of the Band
638: function parameters of $\alpha = -1$, $\beta = -2.5$, and $\eop = 30$ keV.
639: With the assumption of the Band parameters of $\alpha = -1$ and $\beta = -2.5$,
640: the fraction of the long GRBs (T$_{90}$ $>$ 2 s) with $\eop < 100$ keV can
641: be estimated to be about 60\% of the total long GRB samples.
642: On the other hand,
643: according to the BATSE spectral catalog \citep{kaneko2006},
644: there are only 3\% of the long BATSE GRBs with $\eop < 100$ keV.
645: Therefore, the $\eop$ distribution of the BAT GRBs could be systematically
646: lower than the BATSE $\eop$ distribution because of the relatively
647: lower energy coverage of the BAT
648: %% referee's comment #5
649: \footnote{Note that the estimated number of {\it Swift} GRBs with
650: $\eop < 100$ keV depends on the assumption of $\alpha$. If we estimate the
651: number of {\it Swift} GRBs of $\eop < 100$ keV for $\alpha = -1.6$, which
652: is the peak of
653: the PL photon index distribution of BAT (Figure \ref{fig:pl_phoindex_hist}),
654: and $\beta =-2.5$,
655: it will be 10\% of the total population. However, based on the Band $\alpha$
656: distribution of the BATSE spectral catalog \citep{kaneko2006} and also on the
657: detailed spectral
658: simulation study of the BAT data \citep{sakamoto2007}, the BAT PL photon
659: index very likely does not correspond to $\alpha$ of the Band function
660: (see section 5.4).}.
661: %%
662: \subsection{Time-averaged Spectral Parameters}
663:
664: The histogram of the photon index in a PL fit is shown in Figure
665: \ref{fig:pl_phoindex_hist}. The photon index distribution has a
666: single broad peak which centroid around $-1.6$. The peak value
667: of the PL photon index of $-1.6$ is close to the mean value of the
668: low energy photon index of $-1.0$ and the high energy
669: photon index of $-2.5$ of the typical GRB spectrum
670: \citep{kaneko2006,sakamoto2005}. Since we would expect a photon
671: index based on a PL model to be $-1.0$ or $-2.5$ if $\eop$ is above or below
672: the BAT energy band, this result clearly demonstrates that majority
673: of $\eop$ of the BAT GRBs is likely to be within the BAT energy band.
674: The reason why the BAT can not measure $\eop$ for the majority of GRBs
675: is due to its narrow energy band \citep{sakamoto2007}.
676: Although the sample of bursts is very limited, the short GRBs
677: tend to have a harder PL photon index than the long GRBs.
678: However, the PL photon index distributions have a significant
679: overlap between the short and long GRBs.
680:
681: The top panel of Figure \ref{fig:fluence_pflux_phindex} shows the photon
682: index versus the fluence in the 15-150 keV band in a PL fit.
683: There is a correlation between
684: these two parameters for the long GRBs. The correlation coefficient
685: is +0.458 for the sample of 206 bursts. The probability of such a correlation
686: occurring by chance is $<$ 0.001\%. If we include the short duration bursts,
687: however, the correlation becomes weaker (correlation coefficient of +0.228 for the
688: sample of 220 bursts). Therefore, we have confirmed the fluence - hardness
689: correlation \citep[e.g.,][]{lloyd2000} especially for the
690: BAT long GRB sample.
691: %
692: The bottom panel of Figure \ref{fig:fluence_pflux_phindex} shows the time-averaged
693: photon index in a PL fit
694: versus the 1-s peak energy flux in the 15-150 keV band. The
695: correlation coefficient of these two parameters are +0.397 for
696: the GRB sample without the short bursts (total 206 GRBs) and +0.376
697: for the sample with the short bursts (total 220 GRBs).
698: The probabilities of a chance coincidence
699: of the correlation between parameters are $<$ 0.001\% for both cases. Therefore,
700: we also confirmed the correlation between the peak flux and the hardness either
701: with or without the short GRBs \citep[e.g.,][]{lloyd2000}. Note that both
702: the fluence and the peak flux measured by the BAT are not the bolometric values
703: and may introduce the systematic errors in the correlations.
704:
705: For the limited number of GRBs (32 GRBs) which have a significant improvement
706: in $\chi^{2}$ by a CPL fit,
707: we investigated the relationship between $\eop$ and other parameters.
708: %
709: The top panel of Figure \ref{fig:fluence_pflux_ep} shows the distribution of $\ep$
710: and the energy fluence in the 15-150 keV band. The correlation coefficient
711: between these two parameters is +0.580 (a chance probability of 0.2\%). The
712: bottom panel of Figure \ref{fig:fluence_pflux_ep} shows the distribution of $\ep$
713: and the 1-s peak energy flux in the 15-150 keV band. The correlation
714: coefficient is +0.563 (a chance probability of 0.2\%). Note that the fluence
715: and the peak flux are calculated from the BAT energy range. Therefore, these
716: values are not bolometric.
717: %
718: Figure \ref{fig:cpl_alpha_ep}
719: shows the relationship between the photon index in a CPL model and $\eop$.
720: The $\eop$ distribution is spread from 10 keV to 500 keV for the BAT
721: GRB sample. There is no correlation between the photon index and $\eop$.
722: This is consistent with the measurements of other missions
723: \citep[e.g.,][]{kippen2001,sakamoto2005}.
724:
725: Figure \ref{fig:ep_phindex} shows the correlation between the photon index
726: in a PL fit and $\eop$ derived from a CPL fit. As we mentioned in the first
727: paragraph of this section, the variation in the photon indices derived
728: from a PL fit is very likely to reflect the differences in $\eop$
729: energies which are within the BAT energy band.
730: This correlation between the PL photon
731: index and $\eop$ was also recognized by \citet{zhang2007} and their best fit
732: correlation is shown as a dashed line in Figure \ref{fig:ep_phindex}.
733: %
734: Since
735: the correlation found by \citet{zhang2007} is based on $\ep$ derived from a
736: Band function, the slight offset between the dashed line and the data could be due
737: to the systematic difference of $\ep$ based on the choise of the spectral model
738: \citep[e.g.,][]{kaneko2006}.
739: %
740: The detailed study of this correlation based on the spectral simulations will be
741: presented elsewhere \citep{sakamoto2007}.
742:
743: \section{Summary}
744:
745: The first BAT catalog includes 237 GRBs detected by BAT during two and a half
746: years of operation.
747: The BAT ground positions are $<$ 1.75$^{\prime}$ from the XRT position
748: for 90\% of GRBs. We presented the observed temporal and spectral properties
749: of the prompt emission based on the analysis of the BAT event data. Taking into
750: account the difficulty in triggering short GRBs with the imaging instrument
751: like BAT, the duration distributions are consistent with other missions.
752: We showed that the BAT long GRB sample is systematically softer than that of
753: the BATSE bright GRB sample. The correlations such as the fluence- hardness
754: and the peak flux - hardness have been confirmed by the BAT GRB samples.
755:
756: \acknowledgements
757: We would like to thank J.~A. Nousek, J.~P. Osborne, G. Chincarini, Y. Kaneko,
758: and D. Malesani for valuable comments. We also would like
759: to thank the anonymous referee for comments and suggestions that materially
760: improved the paper.
761: This research was performed while T.~S. held a NASA Postdoctoral Program
762: administered by Oak Ridge Associated Universities at NASA Goddard Space
763: Flight Center formerly the National Research Council program.
764:
765: \begin{thebibliography}{99}
766: \bibitem[Barthelmy et al. (2005a)]{barthelmy} Barthelmy, S. D., et al. 2005,
767: \ssr, 120, 143
768: \bibitem[Barthelmy et al. (2005b)]{barthelmy2005b}
769: Barthelmy, S. D., et al. 2005, Nature, 438, 994
770: \bibitem[Band et al.(1993)]{band1993} Band, D.L., et al., ApJ, 413, 281
771: \bibitem[Burrows et al. (2005a)]{burrows} Burrows, S. D., et al. 2005a,
772: \ssr, 120, 165
773: \bibitem[Campana et al.(2006)]{campana2006}
774: Campana, S. et al. 2006, Nature, 442, 1008
775: \bibitem[Cusumano et al.(2006)]{cusumano2006}
776: Cusumano, G. et al. 2006, Nature, 440, 164
777: \bibitem[Donaghy et al.(2007)]{donaghy2007} Donaghy, T. Q. et al,
778: submitted to ApJ (astro-ph/0605570)
779: %\bibitem[Fenimore et al.(2003)]{fenimore2003}
780: % Fenimore, E.E. et al. 2003, in API Conf. Proc. 662,
781: % Gamma-Ray Burst and Afterglow Astronomy 2001,
782: % ed. G.R.Ricker \& R.K. Vanderspek (New York: AIP), 491
783: \bibitem[Fiore, Guainazzi \& Grandi(1999)]{sax1999}
784: Fiore, F., Guainazzi, M., Grandi, P. 1999,
785: {\it Cookbook for BeppoSAX NFI Spectral Analysis}
786: (http://heasarc.gsfc.nasa.gov/docs/sax/abc/saxabc/saxabc.html)
787: \bibitem[Gehrels et al.(2004)]{gehrels2004}
788: Gehrels, N. et al. 2004, ApJ, 611, 1005
789: \bibitem[Gehrels et al.(2005)]{gehrels2005}
790: Gehrels, N. et al. 2005, Nature, 437, 851
791: \bibitem[Jung (1989)]{jung1989}
792: Jung, G.V. 1989, ApJ, 338, 972
793: \bibitem[Kaneko et al.(2006)]{kaneko2006}
794: Kaneko, Y. et al. 2006, ApJS, 166, 298
795: \bibitem[Kippen et al.(2001)]{kippen2001}
796: Kippen, R.M., Woods, P.M., Heise, J., in't Zant, J.J.M.,
797: Briggs, M.S., Preece, R.D. 2003, in API Conf. Proc. 662,
798: Gamma-Ray Burst and Afterglow Astronomy 2001,
799: ed. G.R.Ricker \& R.K. Vanderspek (New York: AIP), 244
800: \bibitem[Kouveliotou et al.(1993)]{kouveliotou1993}
801: Kouveliotou, C. et al. 1993, ApJ, 413, L101
802: \bibitem[Lloyd et al.(2000)]{lloyd2000}
803: Lloyd, N.M., Petrosian, V., Mallozi, R.S. 2000,
804: ApJ, 534, 227
805: %\bibitem[Lloyd-Ronning \& Petrosian (2002)]{lloyd2002}
806: % Lloyd-Ronning, N.M., Petrosian, V. 2002, ApJ, 565, 182
807: \bibitem[Lubi\'nski, Dubath \& Paltani(2005)]{integral2005}
808: Lubi\'nski, P., Dubath, P., Paltani, S. 2005, {\it INTEGRAL
809: cross-calibration status for OSA 5.1} (http://isdc.unige.ch/?Support+documents)
810: %\bibitem[Markwardt et al.(2005)]{markwardt2005}
811: % Markwardt, C. et al. 2005, ApJL, 633, 77
812: \bibitem[Markwardt et al.(2005)]{markwardt2005}
813: Markwardt, C. et al. 2005, GCN Circ. 4037, http://gcn.gsfc.nasa.gov/gcn3/4037.gcn3
814: \bibitem[Ohno et al.(2007)]{ohno2007}
815: Ohno, M. et al. 2007, PASJ in press
816: \bibitem[Olive et al.(2003)]{hete2003}
817: Olive, J.-F. et al. 2003, in API Conf. Proc. 662,
818: Gamma-Ray Burst and Afterglow Astronomy 2001,
819: ed. G.R.Ricker \& R.K. Vanderspek (New York: AIP), 88
820: %\bibitem[Preece et al.(1998)]{preece1998}
821: % Preece, R.D. et al. 1998, ApJL, 506, 23
822: \bibitem[Romano et al.(2006)]{romano2006}
823: Romano, P. et al. 2006, A\&A, 456, 917
824: \bibitem[Roming et al.(2005)]{roming}
825: Roming, P.W.A. et al. 2005, \ssr, 120, 95
826: \bibitem[Rothschild et al.(1998)]{rothschild1998}
827: Rothschild, R.E. et al. 1998, ApJ, 496, 538
828: \bibitem[Sakamoto et al.(2005)]{sakamoto2005}
829: Sakamoto, T. et al. 2005, ApJ, 629, 311
830: \bibitem[Sakamoto et al.(2006)]{sakamoto2006}
831: Sakamoto, T. et al. 2006, in API Conf. Proc. 836,
832: Gamma-Ray Bursts in the Swift Era,
833: ed. S.S. Holt, N. Gehrels \& J.A. Nousek
834: (New York: AIP), 43
835: \bibitem[Sakamoto et al.(2007)]{sakamoto2007}
836: Sakamoto, T. et al. 2007, submitted to ApJ
837: %\bibitem[Sari et al.(1998)]{sari1998}
838: % Sari, R., Piran, T., and Narayan, R. 1998, ApJL, 497, 17
839: %\bibitem[Vestrand et al.(2006)]{vestrand2006}
840: % Vestrand, W. T. 2006, Nature, 442, 172
841: \bibitem[Villasenor et al.(2005)]{villasenor2005}
842: Villasenor, J.S. et al. 2005, Nature, 437, 855
843: \bibitem[Zhang et al.(2006)]{zhang2006}
844: Zhang, B. et al. 2006, ApJ, 642, 354
845: \bibitem[Zhang et al.(2007)]{zhang2007}
846: Zhang, B. et al. 2007, ApJL, 655, 25
847: \end{thebibliography}
848:
849: \clearpage
850:
851: %% Tables
852: %%% BAT summary table
853: \input{tab1.tex}
854:
855: %%% BAT energy fluence table
856: \input{tab2.tex}
857:
858: %%% BAT 1-s peak phton flux table
859: \input{tab3.tex}
860:
861: %%% BAT 1-s peak ene flux table
862: \input{tab4.tex}
863:
864: %%% BAT time-ave spec
865: \input{tab5.tex}
866:
867: %% Figures
868: %%% Crab position accuracy
869: \newpage
870: \begin{figure}[p]
871: \centerline{
872: \includegraphics[width=9cm,angle=-90]{f1a.eps}}
873: \vspace{1cm}
874: \centerline{
875: \includegraphics[width=9cm,angle=-90]{f1b.eps}}
876: \caption{Top: The difference in R.A. (top) and Dec. (bottom) between
877: the BAT detected ($> 8 \sigma$) and the SIMBAD ('true') Crab position. The dotted
878: line is the best fit gaussian model. The centroid and sigma of the
879: best fit gaussian are $-0.07^{\prime}$ and $0.33^{\prime}$ for
880: R.A., and $0.08^{\prime}$ and $0.27^{\prime}$ for Dec., respectively. Bottom:
881: BAT position errors as a function of signal to noise ratio for the Crab. The red dashed
882: line is the position error $\propto$ SNR$^{-0.7}$.}
883: \label{fig:crab_pos_accuracy}
884: \end{figure}
885:
886: %%% Systematic error vector
887: \newpage
888: \begin{figure}[p]
889: \centerline{
890: \includegraphics[width=10cm,angle=-90]{f2.eps}}
891: \caption{Systematic error as a function of energy. The systematic
892: error vectors must be applied to the BAT spectral file created by
893: the BAT software, HEASoft 6.2 and CALDB: 2006-10-14, due to the
894: current uncertainly in the energy calibration.}
895: \label{fig:sysvector}
896: \end{figure}
897:
898: %%% Crab phindex/flux vs theta
899: \newpage
900: \begin{figure}[p]
901: \centerline{
902: \includegraphics[width=9cm,angle=-90]{f3a.eps}}
903: \vspace{1cm}
904: \centerline{
905: \includegraphics[width=9cm,angle=-90]{f3b.eps}}
906: \caption{The power-law photon index (top) and the flux in the
907: 15-150 keV band as a function of the incident angle ($\theta$) of the Crab
908: observed December 2004-April 2005 (black), November 2005 (red), and
909: August 2006 (blue). The horizontal dashed lines are the Crab canonical
910: values of $-2.15$ for the photon index and $2.11 \times 10^{-8}$ ergs
911: cm$^{-2}$ s$^{-1}$ for the flux. The dashed dotted lines are $\pm$5\% of
912: the photon index and $\pm$10\% of the flux canonical values.}
913: \label{fig:crab_phindex_flux_theta}
914: \end{figure}
915:
916: %%% 2D Crab phindex/flux
917: \newpage
918: \begin{figure}[p]
919: \centerline{
920: \includegraphics[width=18cm,angle=0]{f4a.eps}}
921: \vspace{1cm}
922: \centerline{
923: \includegraphics[width=18cm,angle=0]{f4b.eps}}
924: \caption{The contour maps (sparsely sampled) of the Crab photon index (top) and the flux in the
925: 15-150 keV band in units of 10$^{-8}$ ergs cm$^{-2}$ s$^{-1}$
926: (bottom) in the BAT field of view in tangent plane coordinates (IMX and IMY).}
927: \label{fig:2d_crab_phindex_flux}
928: \end{figure}
929:
930: %%% BAT postion notice time
931: \newpage
932: \begin{figure}[p]
933: \centerline{
934: \includegraphics[width=10cm,angle=-90]{f5.eps}}
935: \caption{The time delay from the BAT trigger time to the GCN BAT Position Notice
936: (the BAT burst sample from GRB 050215A to GRB 070616 excluding the GRBs found
937: in ground processing).}
938: \label{fig:bat_pos_notice_time}
939: \end{figure}
940:
941: %%% BAT - XRT position (position accuracy)
942: \newpage
943: \begin{figure}
944: \centerline{
945: \includegraphics[width=10cm,angle=-90]{f6.eps}}
946: \caption{The histogram (top) and the cumulative fraction (bottom)
947: of the angular difference between the BAT ground position and the
948: XRT position. 68\% and 90\% of BAT ground positions are within
949: 0.95$^{\prime}$ and 1.75$^{\prime}$ from the XRT position, respectively.}
950: \label{fig:bat_xrt_pos_diff}
951: \end{figure}
952:
953: %%% BAT theta hist
954: \newpage
955: \begin{figure}
956: \centerline{
957: \includegraphics[width=10cm,angle=-90]{f7.eps}}
958: \caption{The incident angle ($\theta$) distribution of the BAT GRBs.
959: The red line shows the calculated $\theta$ distribution in the case
960: of the uniform sky distribution.}
961: \label{fig:bat_theta}
962: \end{figure}
963:
964: %%% BAT sky map
965: \newpage
966: \begin{figure}
967: \centerline{
968: \includegraphics[width=15cm,angle=0]{f8.eps}}
969: \caption{Sky distribution of the 237 BAT bursts in Galactic coordinates.}
970: \label{fig:bat_sky_map}
971: \end{figure}
972:
973: %%% BAT T90/T50
974: \newpage
975: \begin{figure}
976: \centerline{
977: \includegraphics[width=10cm,angle=-90]{f9.eps}}
978: \caption{T$_{90}$ (top) and T$_{50}$ (bottom) distributions from the BAT
979: mask-weighted light curves in the 15-350 keV band.}
980: \label{fig:bat_t90_t50}
981: \end{figure}
982:
983: %%% BAT T90/T50 vs. S23
984: \newpage
985: \begin{figure}
986: \centerline{
987: \includegraphics[width=10cm,angle=-90]{f10a.eps}}
988: \vspace{0.5cm}
989: \centerline{
990: \includegraphics[width=10cm,angle=-90]{f10b.eps}}
991: \caption{T$_{90}$ (top) and T$_{50}$ (bottom) versus the fluence ratio between
992: the 50-100 keV and the 25-50 keV bands.}
993: \label{fig:bat_t90_t50_vs_s23}
994: \end{figure}
995:
996: %%% BAT peak flux vs fluence
997: \newpage
998: \begin{figure}
999: \centerline{
1000: \includegraphics[width=10cm,angle=-90]{f11.eps}}
1001: \caption{The distribution of the energy fluence in the 15-150 keV band
1002: versus 1-s peak photon flux in the 15-150 keV band.}
1003: \label{fig:bat_peakflux_fluence}
1004: \end{figure}
1005:
1006: %%% S12 vs. S34
1007: \newpage
1008: \begin{figure}
1009: \centerline{
1010: \includegraphics[width=10cm,angle=-90]{f12.eps}}
1011: \caption{The distribution of the energy fluence in the 15-50 keV band
1012: versus that in the 50-150 keV band. Long GRBs (T$_{90}$ $\ge$ 2 s) are in
1013: black and short GRBs (T$_{90}$ $<$ 2 s) are in red. The blue
1014: dash-dotted line is the case of the Band function of $\alpha=-1$,
1015: $\beta=-2.5$, and $\eop=100$ keV. The blue dashed line is the case
1016: of the Band function of $\alpha=-1$, $\beta=-2.5$, and $\eop=30$ keV.}
1017: \label{fig:s12_s34}
1018: \end{figure}
1019:
1020: %%% PL photon index
1021: \newpage
1022: \begin{figure}
1023: \centerline{
1024: \includegraphics[width=10cm,angle=-90]{f13.eps}}
1025: \caption{The histogram of the photon index in a PL fit for long GRBs
1026: (black) and short GRBs (red). The short GRB which has a PL photon index
1027: of $-2.5$ is GRB 050906.}
1028: \label{fig:pl_phoindex_hist}
1029: \end{figure}
1030:
1031: %%% fluence vs. PL photon index
1032: %%% Peak flux vs. PL photon index
1033: \newpage
1034: \begin{figure}
1035: \centerline{
1036: \includegraphics[width=8cm,angle=-90]{f14a.eps}}
1037: \vspace{1cm}
1038: \centerline{
1039: \includegraphics[width=8cm,angle=-90]{f14b.eps}}
1040: \caption{Top: The distribution of the PL photon index versus
1041: the energy fluence in the 15-150 keV band
1042: for long GRBs (black) and short GRBs (red). Bottom: The
1043: distribution of the PL photon index versus
1044: the 1-s peak energy flux in the 15-150 keV band
1045: for long GRBs (black) and short GRBs (red).}
1046: \label{fig:fluence_pflux_phindex}
1047: \end{figure}
1048:
1049: %%% fluence/peak flux vs. Ep
1050: \newpage
1051: \begin{figure}
1052: \centerline{
1053: \includegraphics[width=8cm,angle=-90]{f15a.eps}}
1054: \vspace{1cm}
1055: \centerline{
1056: \includegraphics[width=8cm,angle=-90]{f15b.eps}}
1057: \caption{Top: The distribution of $\ep$ versus
1058: the energy flux in the 15-150 keV band. Two GRBs which locate in
1059: $\sim$ 7 $\times$ 10$^{-8}$ erg cm$^{-2}$ are GRB 050815 and GRB 050925. One GRB
1060: which has $\ep$ of $\sim$ 20 keV is GRB 060428B. Bottom: The distribution
1061: of $\ep$ versus the 1-s peak energy flux in the 15-150 keV band.}
1062: \label{fig:fluence_pflux_ep}
1063: \end{figure}
1064:
1065: %%% CPL alpha vs. Epeak
1066: \newpage
1067: \begin{figure}
1068: \centerline{
1069: \includegraphics[width=10cm,angle=-90]{f16.eps}}
1070: \caption{The distribution of $\eop$ versus the CPL photon index for 32
1071: GRBs which have a significant improvement in $\chi^{2}$ by a CPL fit over
1072: a PL fit.}
1073: \label{fig:cpl_alpha_ep}
1074: \end{figure}
1075:
1076: %%% Ep vs. Photon index
1077: \newpage
1078: \begin{figure}
1079: \centerline{
1080: \includegraphics[width=10cm,angle=-90]{f17.eps}}
1081: \caption{The distribution of $\eop$ versus the PL photon index for 32
1082: GRBs. The dashed line is the PL photon index - $\eop$ correlation of
1083: \citet{zhang2007}: $\log \eop = 2.76 - 3.61 \log (-\alpha_{PL}$).}
1084: \label{fig:ep_phindex}
1085: \end{figure}
1086:
1087: \end{document}
1088: