Note that the Lick MDS group's correction flatfield files described in this memo have been installed into the Calibration Database (CDB) with extensions of r8h/r8d; the archive class, as for all reference files, is "CDB".
-S. Baggett and J. Biretta, STScI, WF/PC-1 Group
filter rootname
F555W --> e2a1258lw
F785LP --> e2a1258ow
Any questions or problems, please contact the Lick MDS group:
phillips@lick.ucsc.edu
forbes@lick.ucsc.edu
mab@lick.ucsc.edu
Text provided by Lick MDS group:
The flatfield used in the standard STScI reduction pipeline contains significant background gradients of about +/-10%, which if uncorrected can lead to photometric errors in a single filter of about 0.10 mag and about 0.15 mag in a color term.
We have created correction flatfields for the F555W and F785LP filters (details are described in Phillips et al. 1993, submitted to A. J.). The internal chip photometric accuracy after applying the correction flat is about 1-2% (a factor of 5 or more improvement). The chip-to-chip accuracy is about 1.2% rms.
HOW THEY WERE MADE:
These correction frames were created from combined multi-orbit F555W and F785LP images of a high galactic, low ecliptic latitude field acquired in January, 1992, as part of the MDS. The sky background in each filter/CCD combination was fit interactively with a bi-cubic spline surface fit, carefully rejecting all objects. We estimate these fits model the sky to within 1-2% everywhere except the very edges of each chip, with the largest errors occuring where a bright extended object fell near the edge.
The F555W surface fits for WF1, WF2, WF3 were modified to remove the ''doughnuts'' or pupil images (Hester 1992) introduced by the F122M filter which was used as a neutral density filter in creating the pipline flat-field, c191513jw.r6h. We modeled the doughnut structures semi-empirically, assuming circular inner and outer edges and constant amplitude (in the uncalibrated data), and with the approximate positions and amplitudes determined by comparison with the pipeline flat-field image and the MDS data. The three vanes in each reflected image were likewise modeled. Generally, residuals from these pupil images are quite small in corrected data, but near the edges of the doughnuts residuals may be as high as 3-5% due to the diffraction effects being poorly approximated.
A modification was made to the F785LP surface fits (all CCDs) in order to remove a small degree of ''odd-even'' pattern that was detected in the pipeline flat-field, c1915143w.r6h. This pattern can be easily seen by comparing an average of all odd columns with all even columns. The odd-even pattern has a peak-to-peak amplitude of only about 1% and so is negligible for many applications.
Both surface fits were then normalized by the mean value over columns 41-780 and rows 51-780, a reasonable characterization of the useable area of each chip and avoiding any poorly-constrained portions of the fits. The inverse was then taken so that pipeline-calibrated data can by multiplied by the correction frames. We normalize each chip independently so that noise characteristics in the data will be preserved as much as possible.
These flats are INVERSE CORRECTION FLATS, i.e. the standard STScI pipeline-calibrated data should be multiplied by our correction flats. Our flats are appropriate for the STScI flatfield c191513jw.r6h (F555W) and c1915143w.r6h (F785LP).
Users may also choose to apply the STScI 'delta flats' to the F555W data for the appropriate date of observation (see Baggett and MacKenty 1993, WFPC Instrument Science Report 92-10). Again your data should be multiplied by the delta flat.
After the above flatfield corrections, the appropriate zeropoints are:
Chip Filter Zeropoint Chip Filter Zero
---- ------ --------- ---- ------ ---------
WF1 F555W 22.90 WF1 F785LP 21.56
WF2 F555W 23.04 WF2 F785LP 21.64
WF3 F555W 23.04 WF3 F785LP 21.44
WF4 F555W 22.96 WF4 F785LP 21.47
We ask that people who publish work that use our correction flats reference: Phillips, A. C., Forbes, D. A., Bershady, M. A., Illingworth, G. D., Koo, D. C. 1993 (Submitted to A. J.)
Files provided by the Lick group (Phillips, A. C., Forbes, D. A., Bershady, M. A., Illingworth, G. D., Koo, D. C.; January 1994). Please acknowledge use of correction flats in any publication based on data calibrated with them.
-- Drew Phillips and Duncan Forbes, 09 December 1993
We would also appreciate hearing from users about the level of photometric improvement they achieved, so we can pass this information on to others.