Bursts since the end of the 1B catalog (March 1992) occurred when the GRO tape recorders were experiencing numerous errors. Consequently, there are gaps in the data of many bursts that preclude valid measurement of peak flux, peak rate, fluence, or duration. Peak rates on the 1 second timescale from each detector are almost always available. These data (called MAXBC rates) can be used to determine burst location. Previous difficulties with this data type have been largely removed, and we now believe that the systematic errors for MAXBC-located bursts are the same as for bursts located with other data types. It is still true however, that the MAXBC-located bursts usually have larger statistical errors than would be the case if another data type were available. The parameter comments_position provides information for MAXBC-located bursts. A number of CGRO and BATSE flight software changes have significantly reduced the problem of data gaps since March of 1993.
Naming Convention Contents tte_bfits_YYYY.fits time-sequenced 4-energy channel data bracketing trigger; combines discla, preb, discsc, and tte. discsc_bfits_YYYY.fits time-sequenced 4-energy channel data bracketing time of burst trigger for triggered detectors; combines data types discla, preb and discsc. (s)her_bfits_X_YYYY.fits time-sequenced 128 energy channel data bracketing time of burst trigger for specified detector; combines datatypes her and herb - suggested data type for 128 energy channel burst analysis. mer_bfits_YYYY.fits time-sequenced 16-energy channel data for times bracketing burst trigger - triggered detectors only; combines mer and cont datatypes - suggested for 16-energy channel burst analysis. discsc_drm_YYYY.fits detector response matrix for 4 energy channel burst data triggered detectors only; used with 4-energy channel data to determine burst photon spectra. her_drm_X_YYYY.fits detector response matrix for 128-energy channel burst data for specified detector; used with 128-energy channel counts data to generate burst photon spectra. mer_drm_YYYY.fits detector response matrix for 16-energy channel burst data triggered detectors only; used with 16-energy channel data to generate burst photon spectra. cont_TJD_fits 8 BATSE detectors, 2.048s resolution/16 energy channels. discla_TJD.fits 8 BATSE detectors, 1.024 s resolution/4 energy channels. XXX_TJD1_TJD2_his.fits Occultation histories. XXX_TJD1_TJD2_nhis.fits Data for one or more BATSE detectors for available energy channels as source count rate (counts/sec) from which the background has been subtracted - used for light curves. XXX_TJD_lad_p11.fits Pulsar low level data. XXX_TJD1_TJD2_lad_olc Light curve file for a given pulsar.where YYYY = trigger number, XXXX = source name, TJD = Truncated Julian Day, X = detector number.
Notice that, because of the instrument configuration at the time of the event, the same files are not available for all triggers. Spectral (SD) data are prefixed with an 's' (e.g. 'sher').
Available data taken prior to the trigger may contain the beginning of the triggering event before it satisfied the triggering criteria. Background-type files can be used to remove background signal levels from the triggered period. The BFITS data files - containing burst and background spectral data as a function of time - and the detector response matrices (DRM) - modeling the instrument response to account for scattering and other effects - are extremely useful for gamma-ray burst analysis. Also, the BFITS and DRM files can be converted to PHA-II and RMF format for analysis with XSPEC using available FTOOLS. Other file types exist, notice, and advice on their use is obtainable at the CGRO help desk.
The BATSE trigger number is a running sequence of BATSE triggers which include cosmic bursts, solar flares and other events. The sequence begins with trigger 105 and ends with trigger 3174. Each burst has a unique catalog name. These BATSE catalog names later may be incorporated into a multi-spacecraft catalog with "GB" or "GRB" replacing this designation of "3B". The characters "3B " begin every BATSE catalog burst name, followed by the "yymmdd" of the burst. "yymmdd" is the two digit year, two digit month, and two digit day. When more than one gamma-ray burst occurs on one day, those bursts have a single letter suffix (B,C,D...), generally in order of intensity. Example: 3B 920503B refers to the second brightest burst that triggered BATSE May 3, 1992. The brightest burst on that day will have no suffix.
The BATSE on-board software tests for bursts by comparing the count rates on the eight large-area detectors to threshold levels for three separate time intervals: 64 ms, 256 ms, and 1024 ms. A burst trigger occurs if the count rate is above threshold in two or more detectors simultaneously. The thresholds are set by command to a specified number of standard deviations above back- ground (nominally 5.5 sigma). Background rates are recomputed every 17 seconds. The thresholds exhibit a coarse quantization that results from truncating the square root of the 64 ms count rate. Since we require that rates be above the thresholds of at least two detectors, the trigger threshold is determined by the threshold of the second most brightly illuminated detector.
When a burst trigger occurs, subsequent triggers are disabled during the accumulation period when the BATSE burst memories accumulate data. These data are then transmitted. During this readout period, the 64 ms threshold is revised to correspond to the maximum rate attained by the current burst, and triggering is disabled on the 256 ms and 1024 ms timescales. Bursts intense enough to trigger during this readout period are termed "over- writes". They are recognized in the table by the value of -999 in the threshold columns for 256 ms and 1024 ms.
Since a trigger can occur on any of the three timescales, there are often cases in which the maximum rate will be below threshold on one or two of the timescales. The value of V/Vmax can be determined for any burst by selecting the maximum of the three peak rates, raised to the -3/2 power. Many bursts have unknown counts or thresholds on one or more timescales. These are marked by a "-999" in the table. This can happen for one of the follow- ing reasons: (1) If the trigger occurs on the 64 ms timescale during the peak 256 ms rate, then the peak 256 ms rate is not found; (2) If the 64 ms peak rate never exceeds the 64 ms threshold, and it occurs before the trigger time, then the peak 64 ms rate is not found; (3) If the 256 ms peak rate never exceeds the 256 ms threshold, and it occurs before the trigger time, then the peak 256 ms rate is not found. Note that items 2 and 3 do not affect V/Vmax, since these peak rates do not exceed threshold; item 1 can, on rare occasions, lead to an overestimate of V/Vmax.
Users should note that T90 and T50 are not available for those bursts which suffer from data gaps during the event; the integration procedure inherently fails in these cases. However, visual estimates of the burst duration are provided in the parameter Comments_Duration for those bursts with sufficient data coverage. Users may also find other pertinent comments concerning the calculated value of T90 and T50 therein, and it is highly recommended that this parameter be consulted before any distribution selected on T90 or T50 is used. Access to high time resolution data has been improved since the 2B catalog, therefore the values of T90 and T50 have been recalculated for a number of short duration bursts.
The preliminary data in the online version of the 3B Durations Table were replaced on 13 October 1995 with the current data. The final data set was calculated using an improved algorithm, primarily affecting the measurements of T90 and T50 for short bursts for which high time resolution data (TTE) are not available. Only three values of T90, and three values of T50, changed by more than 2 sigma. Each of these changes were less than 5 sigma. The start and end times of the T90 and T50 intervals that were previously reported for bursts with trigger numbers larger than 2230 were the midpoints of the bins in which they were found, rather than the start and end times, respectively, of the bins. The start and end times of all intervals in this final version are the start and end times, repectively, of the bins in which they are found. In the preliminary version, the start times of the T90 interval reported for bursts with trigger numbers larger than 2230 were found using a different method than that used in the 1B and 2B catalogs. In this final version, the start times of the T90 interval are found using the same method as used in these earlier catalogs. Finally, in the preliminary version, the uncertainties reported for bursts with trigger numbers larger than 2230 were underestimated by, on average, 20%. This also has been corrected in this final version. The data on durations were revised on 3/7/96, resulting from a thorough review by the BATSE team of the previous measurements. Improvements were made to the measurements for 14 bursts.
Trigger
The BATSE trigger number; this is a running sequence number of BATSE triggers
which include cosmic bursts, solar flares and other events. The sequence
begins with trigger 105 and ends with trigger 3174.
Name
The BATSE Catalog burst name. Each burst has a unique catalog name. These
BATSE catalog names later may be incorporated into a multi-spacecraft
catalog with "GB" or "GRB" replacing this designation of "3B". The characters
"3B " begin every BATSE catalog burst name, followed by the "yymmdd" of
the burst, where "yymmdd" is the two digit year, two digit month, and two
digit day. When more than one gamma-ray burst occurs on one day, those bursts
have a single letter suffix (B,C,D...) that is generally in order of intensity.
Example: 3B 920503B refers to the second brightest burst that triggered BATSE
on May 3, 1992. The brightest burst on that day will have no suffix.
RA
Right Ascension in the default equinox.
Dec
Declination in the default equinox.
LII
Galactic longitude in decimal degrees.
BII
Galactic latitude in decimal degrees.
Day_Trigger
The truncated Julian Date (TJD) of the trigger: TJD = JD - 2440000.5
Time
The burst trigger time, converted to shf for the BROWSE database or to
MJD for the W3Browse database, for the convenience of searches and
cross-correlations. The burst trigger time is the end of the interval (64,
256 or 1024 ms) in which the burst triggered the detector.
Seconds_Trigger
The burst trigger time, in decimal seconds of the day (UT) on which it
occurred. The burst trigger time is the end of the interval (64,
256 or 1024 ms) in which the burst triggered the detector.
Error_Radius
The radius in decimal degrees of the positional error box.
The error in angular location is the radius of a circle having
the same area as the 68% confidence ellipse defined by the formal
covariance matrix from a chi^2 fit on the assumption of normal
errors. The error is based solely on the Poisson uncertainty in the
BATSE measurement of burst flux by each Large Area Detector. There is,
in addition, an RMS systematic error of approximately 1.6 degrees. Adding
1.6 degrees in quadrature to the error in the table yields our estimate
of the 68% confidence interval for the burst location error. The statistical
error is believed to be Guassian. The systematic error distribution is not
known very well, but may have a more extended tail than a Gaussian. The
improvements in the 3B location algorithm are known to have significantly
reduced this extended tail of the systematic error distribution.
Earth_Angle
The angle in decimal degrees of the geocenter, i.e., the angle between the
burst and the nadir, as measured from the satellite.
Overwrite
The overwrite flag: this is Y (true) if this burst overwrote an earlier,
weaker trigger, N (false) otherwise.
Overwritten
The overwritten flag: this is Y (true) if this burst was overwritten by
a later, more intense trigger, N (false) otherwise.
Max_Cts_64
The maximum counts in the second most brightly illuminated detector divided by
the threshold count rate on the 64 ms timescale.
Threshold_64
The trigger threshold on the 64 ms time-scale. It is the number of counts
in 64 ms required to trigger the second most brightly illuminated detector
for this particular burst.
Flux_64
The peak flux on the 64 ms timescale in units of photons/cm^2/sec.
Flux_64_Error
The one-sigma statistical error in the peak flux on the 64 ms timescale.
Flux_64_Time
The time of the peak flux on the 64 ms timescale, in decimal seconds
relative to the burst trigger time for the end of the interval in which
the flux was calculated.
Max_Cts_256
The maximum counts in the second most brightly illuminated detector divided by
the threshold count rate on the 256 ms timescale.
Threshold_256
The trigger threshold on the 256 ms timescale. It is the number of counts in
256 ms required to trigger the second most brightly illuminated detector for
this particular burst.
Flux_256
The peak flux on the 256 ms timescale in units of photons/cm^2/sec.
Flux_256_Error
The one-sigma statistical error in the peak flux on the 256 ms timescale.
Flux_256_Time
The time of the peak flux on the 256 ms timescale, in decimal seconds
relative to the burst trigger time for the end of the interval in which
the flux was calculated.
Max_Cts_1024
The maximum counts in the second most brightly illuminated detector divided by
the threshold count rate on the 1024 ms timescale.
Threshold_1024
The trigger threshold on the 1024 ms timescale. It is the number of counts in
1024 ms required to trigger the second most brightly illuminated detector for
this particular burst.
Flux_1024
The peak flux on the 1024 ms timescale in units of photons/cm^2/sec.
Flux_1024_Error
The one-sigma statistical error in the peak flux on the 1024 ms timescale.
Flux_1024_Time
The time of the peak flux on the 1024 ms timescale, in decimal seconds
relative to the burst trigger time for the end of the interval in which
the flux was calculated.
T50
The 50% duration of the burst in seconds. T50 measures the duration of
the time interval during which 50% of the total observed counts have been
detected. The start of the T50 interval is defined by the time at which
25% of the total counts have been detected, and the end of the T50 interval
is defined by the time at which 75% of the total counts have been detected.
T50_Error
The uncertainly in the T50 duration.
T50_Start
The start time of the T50 interval, relative to the trigger time (Time),
in seconds.
T90
The 90% duration of the burst in seconds. T90 measures the duration of
the time interval during which 90% of the total observed counts have been
detected. The start of the T90 interval is defined by the time at which 5%
of the total counts have been detected, and the end of the T90 interval is
defined by the time at which 95% of the total counts have been detected.
T90_Error
The uncertainty in the T90 duration.
T90_Start
The start time of the T90 interval, relative to the trigger time (Time),
in seconds.
Fluence_1
The fluence for Channel 1 (energy range 20-50 keV), in units of ergs/cm^2.
Fluence_1_Error
The error in the fluence for channel 1.
Fluence_2
The fluence for Channel 2 (energy range 50-100 keV), in units of ergs/cm^2.
Fluence_2_Error
The error in the fluence for channel 2.
Fluence_3
The fluence for Channel 3 (energy range 100-300 keV), in units of ergs/cm^2.
Fluence_3_Error
The error in the fluence for channel 3.
Fluence_4
The fluence for Channel 4 (energy range E > 300 keV), in units of ergs/cm^2.
Since channel 4 is an integral channel, fluences given for this channel are
quite sensitive to the assumed spectral form. Spectral analyses in this
energy range should be performed with higher resolution data types.
Fluence_4_Error
The error in the fluence for channel 4.
Comments_Quality
Comments on data quality: not all gamma-ray burst have such comments.
Comments_Otherobs
Comments on additional observations by other instruments: not all gamma-ray
burst have such comments.
Comments_General
General comments: not all gamma-ray burst have such comments.
Comments_Position
Comments on the gamma-ray burst co-ordinates: not all gamma-ray burst
have such comments.
Comments_Duration
Comments on the gamma-ray burst duration: not all gamma-ray burst have
such comments.
Class
The browse classification; all entries in this database have been classed
as gamma-ray bursts.