J. Racusin (PSU), D. N. Burrows (PSU), P. A. Evans (U Leicester), C.
Pagani (PSU), report on behalf of the Swift-XRT Team:
We have analysed the first four orbits of Swift-XRT data obtained for GRB
080319B (Racusin et al. GCN Circ. 7427), totaling 1.1 ks of Windowed
Timing (WT) data beginning 64 s after the BAT trigger, and 4 ks of Photon
Counting (PC) data beginning 5 ks after the BAT trigger.
The UVOT-enhanced XRT position has been given by Evans et al. in GCN Circ
The bright X-ray light-curve can be fit by a broken power-law, with an
initial decay index of 1.46 +/- 0.01 followed, after a break at 6100 +/-
440 seconds, by a steeper decay index of 2.48 +/- 0.10.
Both the WT and PC spectra are strongly affected by photon pile-up, which
can alter the spectral fits. To eliminate these effects, we exclude the
central 8 pixel radius region of the WT data, and the central 3 pixel
radius region of the PC data when creating the spectra. Preliminary fits
to the WT spectrum (64-4943 seconds), modeled with an absorbed power-law,
result in a photon index of 1.65 +/- 0.02 and an absorbing column at
z=0.937 (Vreeswijk et al., GCN Circ. 7444) of NH = (5.28 +/- 1.03)e20
cm^-2, in addition to Galactic absorption of 1.12e20 cm^-2 in the
direction of the burst, with a reduced Chi^2=1.05. The PC spectrum (5-19
ks) can be modeled as an absorbed power-law, with photon index of 1.92 +/-
0.07 and an absorbing column at z=0.937 of NH = (23.5 +/- 5.6)e20 cm^-2,
in addition to the Galactic absorption, with a reduced Chi^2=1.06. If we
freeze the NH in the PC model to the WT fitted value, we fit a photon
index of 1.73 +/ 0.04, with a reduced Chi^2=1.4, which excludes this model
as a viable fit. As another attempt to model the spectra without NH
evolution, we freeze the WT NH to the value from the PC spectral fits and
add an additional low energy thermal component. The resulting fits yield
a photon index of 1.73 +/- 0.02, kT=0.06 +/- 0.01 keV, and a reduced
Chi^2=1.07. Therefore, the extra thermal component is a possible
explanation for the apparent spectral evolution.
Assuming the source continues to decay with the same decay index of 2.5,
we predict an XRT count rate of 8.8e-4 counts/s at T+24 hours, which
corresponds to an 0.3-10.0 keV observed (unabsorbed) flux of 6.1e-14
(6.2e-14) ergs cm^-2 s^-1.
This circular is an official product of the Swift-XRT team.