[vsnet-grb-info 20095] Possible blackbody component in the X-ray spectrum of GRB171205A

[GCN Circulars]

TITLE:   GCN CIRCULAR
NUMBER:  22191
SUBJECT: Possible blackbody component in the X-ray spectrum of GRB171205A
DATE:    17/12/06 17:31:56 GMT
FROM:    Sergio Campana at INAF-OAB  <sergio.campana at brera.inaf.it>

S. Campana (INAF-OAB), A. Beardmore (U. Leicester), A. D'Ai (INAF-IASFPA),
V. D'Elia (ASDC), P. Evans (U. Leicester), J. Kennea (PSU), J.P. Osborne
(U. Leicester), K.L. Page (U. Leicester), R.L.C. Starling (U. Leicester)

report on a preliminary analysis of the Swift XRT data. Swift XRT started
observing GRB171205A (D'Elia et al. 2017, GCN 22177) 145 s after the 
trigger.
XRT detected a bright X-ray source decreasing from ~80 c/s to ~10 c/s
from 150 s to 400 s, remaining in Windowed Timing (WT) mode (Kennea et al.
2017, GCN 22183).
The hardness ratio light curve does not imply appreciable spectral
variations during this time interval.
We fit the WT spectrum binned to 20 counts per energy bin with an absorbed
power law (with the column density fixed to 5.9e20 cm-2, Willingale et al.
2013, MNRAS 431 394).
We obtain a reduced chi2=1.23 (239 degrees of freedom, dof, and 9e-3 null
hypothesis probability, nhp).
The inclusion of an intrinsic absorption component at the GRB redshift
(z=0.0368, Izzo et al. 2017, GCN 22180) does not improve the fit, rather
we can only put a 90% confidence level upper limit of N_H(z)<0.3e20 cm-2.

Given the relatively low nhp of the fit we include a soft blackbody 
component.
The fit improves considerably with kT_BB=0.10+-0.02 keV (90% c.l.) and
equivalent radius R_BB~5e11 cm (for a source distance of 163 Mpc).
The fit is good with a reduced chi2=1.09 (237 dof and 0.15 nhp).
F-test provides a probability for a chance improvement of 6e-8.
The 0.3-10 keV blackbody contribution to the total flux is ~4%.

We checked that the fit does not improve using different values for the
Galactic column density (NH_DL=4.82e20 cm-2 or NH_Kalberla=4.91e20
cm-2), reaching similar results.
Using a spectrum made from grade 0 events does not change the
results and the blackbody component is still significant.