[vsnet-grb-info 5671] Swift-XRT improved positions, and new redshift limits.

[GCN Circulars]

TITLE:   GCN CIRCULAR
NUMBER:  7328
SUBJECT: Swift-XRT improved positions, and new redshift limits.
DATE:    08/02/25 11:40:47 GMT
FROM:    Phil Evans at U of Leicester  <pae9 at star.le.ac.uk>

P.A. Evans, A.P. Beardmore, J.P. Osborne and M.R. Goad (U. Leicester),
J. Kennea and D.N. Burrows (PSU) and N. Gehrels (NASA/GSFC) report on
behalf of the Swift-XRT team:

The Swift-XRT team announces the availability of two new products:
significantly improved `enhanced' XRT positions, including promptly
available positions; and prompt XRT data analysis, including nH
estimates and redshift limits.

1) Improved XRT positions.

The technique of using the UVOT as a super-star-tracker to improve XRT
positions of GRBs was presented by Goad et al. (2007, A&A, 476, 1401).
We have made improvements to this process, reducing the typical position
uncertainty by a further 30-50% in radius. We now find positions with
90% confidence error radii <2" 90% of the time. The percentage of
Swift-observed GRBs with enhanced XRT positions has also risen from ~55%
to ~70%. To achieve this we use more UVOT filters and more Swift
observations per burst, and other algorithm improvements (see
http://www.swift.ac.uk/xrt_positions/changes.php).

The table of XRT GRB positions at http://www.swift.ac.uk/xrt_positions
has been updated using the new software. The old table of positions is
still available at http://www.swift.ac.uk/xrt_positions/v1/, this will
not be updated with new GRBs.

The new software can also be used to enhance the promptly available
ground-calculated XRT positions: those produced within ~20 minutes of a
trigger from the TDRSS single pixel Photon Counting mode event (SPER)
data, and reported in the first GCN circular from the Swift team and via
the GCN Notices system. These SPER positions are not as precise as those
described above which use the full ground-based data set (available ~2-3
hours post-trigger), however the 90% confidence position error radius is
<3" 90% of the time which represents a significant improvement on the
`standard' positions available at this early time.

A full table of prompt positions is available at
http://www.swift.ac.uk/sper, this table updates automatically as soon as
new positions are deduced. SPER-derived positions are distributed
automatically as GCN Position Notices, and the positions determined from
ground data will be announced in automatic GCN Circulars (see Osborne et
al., GCN Circ. 6726, for details).

Full documentation of the enhancement process, including verification of
the error-circle radius, is available at
http://www.swift.ac.uk/xrt_positions/docs.php
and for the prompt positions further information is available at
http://www.swift.ac.uk/sper/docs.php.

2) Prompt data analysis.

We have developed tools to produce light curves and spectra from the
promptly available SPER data. The light curve will be used by the XRT
team to rapidly determine whether a source is fading or variable; the
spectra will be used to measure the column density towards the GRB. If
the column density is in excess of the Galactic value we use the
relation given in Grupe et al. (2007, AJ, 133, 2216) to derive an upper
limit on the redshift of the GRB. We use the 90% confidence lower limit
on the excess column. If this provides a limit in the range 2-5, as it
does for 18% of GRBs, we will quote it in the initial GCN circular.

A summary page for each GRB with prompt event data, showing the prompt
position, light curve and spectrum, can be found at
http://www.swift.ac.uk/sper/ which also shows the XRT-derived redshift
limits. Full documentation of the prompt data analysis and verification
of the results is here as well.