README on the despsf.fits file

This is a simulated DES camera PSF file.

It is a 1201x1201 pixel image with the PSF centered on pixel (600,600)
(if you count 1 as the first pixel, if you count 0 as the first, then it's
in 599,599, I think).

It is composed of 4 parts:

1. A central Gaussian of FWHM 1.15 arcsec for r < 1.1''
2. A power law of slope -4.13 for 1.1'' < r < 7.5''
3. A power law of slope -2.46 for 7.5'' < r < 50''
4. A power law of slope -1.16 for r > 50'' 
(the edge of the image is at r=600 pixels = 157'')
The scale of the image is 0.263 arcsec/pixel.

The sum of all the flux under the psf profile is normalized to (almost) 1.0.
In fact the summed flux is as follows as a function of radius:

Light inside 1'' = 82% = 0.9219 counts
Light inside 5'' = 98.3% = 1.115 counts
Light inside 50'' = 99.6% = 1.13 counts
Light inside 157'' = > 99.9%  = 1.134 counts

If you care about the difference between 1.0 and 1.134, you should normalize the image by dividing by 1.134 before using as a PSF or your magnitudes will
change a bit.

You should be able to convolve this with your images, or deconvolve.

If it's too big, you can probably just use the central 100 pixels or less
and still get essentially the same results.

Let me know if you have questions or problems.

This is pretty close to the PSF quoted/measured in the paper.