doing photometry on distorted star images?
cong wrote on Jul 24, 2008
I got a lot of star images from an old piece of hardware. Because of the telescope's problem with tracking, and because of the terrible seeing, the stars (fairly bright objects) on the CCD images are all distorted and looks like ellipses or ovals.
The thing I'm going to do is to extract the magnitudes of them from each image without transforming to standard system (we are doing relative photometry on variable sources). Since the star images are distorted by the telescope's "lagging", I guess the standard procedures of aperture photometry may not work well with the images. I would like to hear your suggestions on dealing with those data.
BTW with the help of SExtractor I'm able to find the stars' locations as well as a fitting ellipse for each star (read: the center coords., eccentricity, orientation, and the FWHM if assuming a Guassian PSF). However, I am yet to have some working idea about taking advantage of them in my photometry.
Thanks in advance.
The thing I'm going to do is to extract the magnitudes of them from each image without transforming to standard system (we are doing relative photometry on variable sources). Since the star images are distorted by the telescope's "lagging", I guess the standard procedures of aperture photometry may not work well with the images. I would like to hear your suggestions on dealing with those data.
BTW with the help of SExtractor I'm able to find the stars' locations as well as a fitting ellipse for each star (read: the center coords., eccentricity, orientation, and the FWHM if assuming a Guassian PSF). However, I am yet to have some working idea about taking advantage of them in my photometry.
Thanks in advance.
Francisco Valdes wrote on Jul 24, 2008
There are two thoughts I have. First, you could use circular apertures anyway. Since you are measuring stars (self-similar PSF) the same fraction of light will be used and so the relative photometry should be correct. The main problem might be that the centroiding is not consistently at the same place in the ellipse. You could also use a large aperture though that would introduce noise and you probably are better off with smaller apertures and a large aperture correction.
The second approach is, obviously, to use an aperture that matches the shape. The apphot.polyphot tasks is supposed to do that. So investigate that option. Note that there is an interactive, GUI-based tool that also supports polygonal apertures call xapphot. This is part of the guapps external package.
Yours,
Frank Valdes
The second approach is, obviously, to use an aperture that matches the shape. The apphot.polyphot tasks is supposed to do that. So investigate that option. Note that there is an interactive, GUI-based tool that also supports polygonal apertures call xapphot. This is part of the guapps external package.
Yours,
Frank Valdes
Last post on Jul 24, 2008