photometry of a cluster

[AndrewS]

I have reduced and coadded image of M13 in a red filter.
The individual frames all had BPMs associated with them. This BPM was to highlight saturated pixels.
when I coadd these to make the final coadded image of m13, there is not a BPM automatically associated with this coadded image.
However I think that for the photometry tasks in IRAF to perform correctly, they should have a BPM associated with the image of the star cluster that they are my processing.

In light of this my question is twofold:

1) What is the nature of the BPM that should be associated with the coadded image - should it be simply a coaddition of all the BPMs of the indivual frames that went into making the coadded image?

2) what is the best way to make the BPM for the coadded image?

[valdes]

If you use imcombine for coadding then you can select a bad pixel mask as another output. If you use individual bad pixels masks as input (an there are a few options for how to specify how the values are to be used) then the output mask will identify where no data is available. This could be because all images in the coadd had bad data at the same place (i.e. saturated pixels in a star) or at the edges because of filling the result into a rectangular image.

The output of imcombine will be 0 for good pixels (at least one good pixel contributing) or 1 for no good pixel in the coadd.

Frank Valdes