[Aavso-photometry] Alternates Photometric Growth curve fitting

[Brad Walter]

 When doing (or at least trying to do) millimag time series photometry
(exoplanet light curves, for example), I have been manually picking the
inner-most aperture on every individual image so that my measurement is
always at the same point on the growth curve. This is exceedingly time
consuming and exceptionally tedious. I have tried a "short cut" of analyzing
the images with the smallest FWHM, the median FWHM and the largest FWHM in
the time series; then finding the ratio of the measurement apertures to
FWHMs that accomplishes the following:
1. Puts all three sets of measurements as close as possible to the same
point on the growth curve of the target star measured as a percentage of the
maximum net counts obtained with an aperture that slightly exceeds the
radius needed to reach the second minima inflection point on the horizontal
and vertical line profiles through the centroid of the star; 
2. Falls on or near the maximum SNR point on the growth curve of the comp
star;
3. If using ensemble photometry, I will adjust condition in #2 to pick an
aperture that falls near some weighted average of the highest SNR points of
the ensemble stars (usually using a weighting that is the square of the
ratio of each star's best SNR to the best SNR of the "brightest" star, i.e.
the one with the highest overall SNR). 

Once I have found these three Aperture to FWHM ratios, I average them and
set the inner apertures for each image in the series at the FWHM (or
weighted average FWHM, in the case of ensemble photometry) times this ratio.


This is faster than making growth curves of each image, but it is still
pretty tedious stuff, and I am not convinced it is rigorous. Does anyone
know a faster, less labor intensive method that is rigorous? I would like to
avoid as much of this drudgery as possible, since I am basically lazy, and I
find it hard to carve out the time to do all of the spreadsheet work this
process takes. On the other hand, the process allows you to achieve
precisions of a few millimag standard deviation even with a 10 inch
telescope as long as you have reasonable consistent skies, even when the
seeing isn't great (e.g. 3.5-5.5 arcsec FWHM).