Hybrid KAMPHOT/Aperture Photometry

The analysis of the M92 photometry indicates that for point sources with 13.5 aperture photometry provides measurements with better internal consistency than psf-fit pphotometry. For fainter sources, psf-fitting performs better. Consequently, optimum photometric measurements should use aperture photometry for bright sources, and transition smoothly to psf-fitting for fainter objects. This can be accomplished in one of two ways:

1. Measure both the aperture and psf-fit magnitudes, and intelligently interpolate between the two. The interpolation would need to be calibrated using measurements on additional large sets of multiply scanned regions such as the M92 fields. This approach could be keyed to source density and revert to pure psf-fitting in very crowded fields.

2. Apply a weighting function to pixel values in a source to be photometered that varies with source brightness. For bright sources, the weighting function is a boxcar, so each pixel is given uniform weight and the integrated brightness gives the same value as simple aperture photometry. For successively fainter sources, the weighting function gradually "softens" to a more psf-like profile, until for faint sources it becomes identically the psf. Again, the brightness dependence of the weighting function would need to be calibrated using measurents on additional large sets of multiply scanned regions.