The Neil Gehrels Swift Observatory

Return to: UVOT Digest

Small Scale Sensitivity Check for UVOT

The UVOT detector has some small-scale sensitivity inhomogeneities. Sources falling within the small areas of low sensitivity can give count rates of, in the worst case, up to 39% lower than in other areas and thus the photometry is not accurate. The UV filters show the effect more strongly than the optical. The bad areas cover just a small percentage of the whole field of view, however in UVW2 they affect nearly 10% of the central 5'x5' region. For the optical filters, up to 5% of the central area is affected. See SWIFT-UVOT-CALDB-17-02 for more information.

Heasoft 6.30 and Later

From Heasoft version 6.30, and CALDB version 20211108 (released 2022-02-25), a small-scale-sensitivity check is performed by default by the software.

If you are running Heasoft 6.29c or earlier, or have an old version of CALDB please see below.

The UVOT ftools now by default use a mask to screen out sources detected in areas of low sensitivity on the detector. The magnitude of the source will be given as 99 and the corrected count rates for that source will appear negative. Additionally, the Small Scale Sensitivity factor (SSS_FACTOR) in the FITS file will be -99.9. Sources with these values should generally be discarded.

A set of three masks are defined for each filter. The default level (Low) removes the smallest number of sources and is designed to cover all the most strongly affected locations. Users can choose instead to use NONE (i.e. disable the use of a mask), or a MID or HIGH level of screening.

The tools (specifically uvotsource, uvotevtlc, uvotlc, uvotmaghist, uvotproduct, uvotpipeline, uvotexpmap) have been updated to use the parameters sssfile (default CALDB) and ssstype (default LOW).

That is, running uvotsource is by default equivalent to running:

$ uvotsource sssfile=CALDB ssstype=low

For more information on which level of screening is appropriate for your science requirements please see SWIFT-UVOT-CALDB-17-02.

For those running Heasoft 6.29c or earlier:

A preliminary release of the SSS check was made available for earlier versions of HEASOFT. To check whether your data may be affected by the areas of lower throughput, you should download this caldb-style file: sssfile5.fits.gz In this SSS file, the SSS factor images are almost entirely filled with 1.0s but wherever there is a bad patch, the SSS factor is set to -99.9. You can use fv or ds9 to look at the images in the file.

The check can be performed by running any of the UVOT routines (uvotsource, uvotmaghist, uvotlss, uvotdetect) as normal but defining the parameter LSSFILE=sssfile5.fits. This replaces the normal CALDB LSS file with this special SSS file.

In the event that your source is located on a bad patch, the LSS factor in the output file will be -99.9 and the magnitude of the source will come out as 99. The photometry for these sources should be discarded.

e.g. > uvotmaghist lssfile=sssfile5.fits infile=sw00055761030uw1_sk.img.gz outfile=maghist.fits srcreg=55761.reg bkgreg=55761-bk.reg plotfile=plotout.gif

NOTE: When using this SSS file in this way, you are disabling the normal LSS (large scale sensitivity) correction. Therefore, it is important to re-run the analysis using the normal LSSFILE (LSSFILE=CALDB, or just use the default setting) to get the best photometry for those sources that are not being discarded.

For more information on this earlier version of the check see SWIFT-UVOT-CALDB-17.

Return to: UVOT Digest