... spectra

See http://swift.gsfc.nasa.gov/docs/swift/analysis/bat_digest.html

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... excluded

Failure of the Thermo Electric Cooler (TEC) during the verification phase of the mission has resulted in operational temperatures for the XRT CCD in the range -40 to -70 degrees Celsius. While above the nominal operational temperature of -100 degrees Celsius, the XRT operates well within pre-flight specifications for temperatures below -50 degrees Celsius.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... data

XSPEC does not bin data. It groups data together between start and stop times such that a specified number of counts are contained within that bin. XSPEC then integrates a model fit to the data. The difference between an XSPEC fit and a fit to binned data is rather subtle, the main difference being that the XSPEC fit to the data is not constrained to pass through the centre of the time bin. This difference is extremely important when fitting GRB light-curves, as they decay rather rapidly, typically $f_{\nu}(t)\propto t^{-1}$ or faster. The use of overly large bins at early times can drastically alter the derived slope of the light-curve if model integration is not performed.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... absorbed

Here we use the XSPEC wabs Wisconsin absorber model.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...$t_{\rm a}$)

This is not true for GRB050219A, for which such a model significantly underpredicts the flux in the late-time data.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... general

The Ghirlanda relation is derived from bursts the majority of which have $E_{\rm peak}$ outside of the 10-100 keV range, where the BAT has the largest effective area.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.