GCN Circular 3996

D. C. Morris (PSU), D. N. Burrows (PSU), and N. Gehrels (GSFC) report on 
behalf of the Swift XRT team:

We have analysed the first six orbits of data for GRB050916 (GCN 3991
Loading...
 
 , 
Morris et al., 2005). Using xrtcentoid, the refined position is:

RA(J2000) = 09h 03m 56.8s
Dec(J2000) = -51d 25' 50.3"

with an uncertainty of 8 arcsec. This is 3.5 arcsec from the original 
XRT position (GCN 3995
Loading...
 
 , Morris et al., 2005) and 23 arcsecs from the 
initial BAT position..

The XRT began taking data at 16:39:22, 210 seconds after the BAT 
trigger. The early data (<T+1 ks) show a fading lightcurve with decay 
index ~-1.5. Data from T+5ks to T+15ks also show a fading lightcurve 
consistent with decay index=-1.5 but at a flux level ~x3 greater than 
would be expected by extrapolating the decay index from the earlier 
data. A bright flare (peaking at x100 above the expected level) is seen 
at T+18ks and flux levels remain elevated for 5-7ks thereafter.

The overall spectrum of all the data is well fit (reduced chi-sq of 
1.04) by a power law with neutral hydrogen equal to the galactic column 
density of 1.1e22 :

NH=1.1e22(frozen)
gamma=1.77 + / - 0.09

Due to the low overall number of events, separate fits to the flaring 
and non-flaring portions of the lightcurve are poorly contrained and do 
not show any significant evidence of spectral evolution:

flaring:
NH=0.89e22 + / - 0.16e22
gamma= 1.54 + / - 0.19

non-flaring:
NH=1.12e22 + / - 0.34e22
gamma=1.85 + / - 0.37

Assuming that the decay slope has returned to ~-1.5 after the flare, the 
expected count rate at T+1e5s is ~2e-3 cts/s which converts to an 
unabsorbed flux of  2.4e-13 ergs cm^-2s^-1. The peak flux (seen at the 
start of the XRT observation and at the peak of the flare) is 0.5 cts/s, 
for a peak unabsorbed flux of 6e-11 ergs cm^-2s^-1.