[Aavso-photometry] Uncertainty limits

[arne]

Keith Graham wrote:
> I have a question regarding useful observations and uncertainty limits. I have noticed, not surprisingly, that my uncertainty readings on moonlit nights are higher than on darker nights. When I get down to 16th mag and dimmer on "bright sky" nights, the uncertainty can sometimes be .2, and I have even gotten as high as .8. These numbers are certainly worse than expected visual uncertainties, but, I can see from my images that the mags are pretty reasonable when the variable is compared to the dimmest comps on the image. So, I was wondering if it is best to simply scrap observations with such a high uncertainty? My thinking is that even with such a high uncertainty, the information is still better than  a < observation. If I can see the star on an image, is it better to attempt a measurement with a high uncertainty than to simply use  < reading? I do realize I can improve the uncertainty with either longer exposures or by stacking images, but I sometimes do not catch how dim
 the target is when I am doing a run of many different stars in one evening.
> 
If you can measure the variable, then you should do so and report whatever
uncertainty you get.  This is always better than a "fainter than" report.
I usually use the smallest aperture I can on such occasions (about 2fwhm),
or else use psf-fitting (equivalent to 1-fwhm aperture) to cut down on
the sky background included in the aperture and thereby improve the
estimate.
Arne