GCN Circular 3956

D. B. Fox (PSU), C. Pagani (PSU/INAF-OAB), L. Angelini (GSFC),
D. N. Burrows (PSU), J. P. Osborne (U. Leicester), and V. La Parola
(INAF-IASF) report:

"We have analyzed the Swift/XRT observations of the BAT localization
region for the short-hard burst GRB 050906 (Parsons et al., GCN 3935
Not found
 
 
).

During the first orbit only, there is evidence for an excess of counts
near the position:

     RA 03:31:15.6, Dec -14:36:37 (J2000)

where we have corrected the XRT coordinates by roughly 6-arcsec from
the native astrometry using a bright star / X-ray source in the field.
Within a 20-pixel (47-arcsec) radius of this position, 6 photons are
detected where 0.9 photons are expected from the background,
representing a detection at 99.7%-confidence when considered over the
full XRT field of view.  If due to an astrophysical source, the
source's estimated count rate is 7.5 +/- 5.2 cts/ksec (90%-confidence)
over the 798 seconds of observation, and the source positional
uncertainty is 18 arcsec (90%-confidence radius).  Subsequent XRT
observations reveal no further excess count detections within this
region.  The source location is not coincident with either the VLA
source (Cameron & Frail, GCN 3933
Not found
 
 
) or the bright galaxy (Levan &
Tanvir, GCN 3927
Not found
 
 
; Parsons et al., GCN 3935
Not found
 
 
) that have been identified
in observations of this burst.

We note that: (1) The location of the excess is 13 arcsec outside the
2.6-arcmin radius (90%-confidence) BAT localization region (Parsons et
al., GCN 3935
Not found
 
 
); (2) The temporal distribution of the counts during the
first orbit is consistent with the temporal distribution of the
detector background counts (88%-confidence via two-sided K-S test);
(3) The spatial distribution of the counts appears flat rather than
PSF-like (unquantified); and (4) The source region lies within a
region of increased local background, towards the end of the first
orbit, due to emission from the bright Earth limb (two counts arrive
during the last 100 seconds; however, the event grades are distinct
from those associated with the bright Earth limb).  Given these
caveats, we are not confident that the counts excess represents the
detection of an astrophysical source.  However, if it is due to a real
source, it is the most likely X-ray counterpart to GRB 050906."