WF/PC Instrument Report 90-05:
A Pre-Flood Study of WF/PC Photometric Stability Using Aperture Photometry on NGC 188 Data Taken in August and September
-Eric Wyckoff, Roberto Gilmozzi, Keith Horne
November, 1990
Excerpts from the report Summary:
A relatively uncrowded field of the Galactic cluster NGC 188 was studied in order to learn about the level of photometric stability of the Wide Field Camera and OTA. Different methods of aperture photometry were applied to images taken a month apart in order to determine which was the most successful at producing repeatability in the data which, as a result of spherical aberration, suffers most dramatically from a position dependent point spread function (PDPSF).
The results depend heavily on the size of the aperture radius and the position and size of the background annulus. The selection of these parameters can create quite different results that range from systematic shifts of several percent to scatter about the fit of up to 45% rms. Of the setups...tried [they] find that the parameter selections that produce the best result are an aperture radius of 0.15", annulus radius of 0.2", annulus width of 0.1", and a mean background determination of the pixels in the annulus. This is based on the assumption that the 'best' result is the one where the line fit to the data has the smallest offset from zero balanced with the smallest amount of scatter about this fit.
A compromise must be made between two major difficulties that affect the quality of aperture photometry that can be done with HST. On one side there is the PDPSF which could be compensated for by using a large aperture to include all flux independent of shape. This, however, is complicated by the fact that the S/N ratio decreases very quickly with distance from the core so large apertures produce noisy measurements. On the other, photometry could be done in the core only where the S/N is high; however, the PDPSF causes a different fraction of the PSF to [be] measured in a small aperture at different positions. [They] believe that this effect can be calibrated using a position dependent zero point for the photometry. Even though the core aperture photometry is the method most sensitive to the variable PSF the fact that [they] find results as good as or better than large aperture photometry implies that the results will be much better when [they] can correct for a variable zero point.
For a copy of this report, please contact help@stsci.edu.