GREAT3 PSF footprints

[barnabytprowe]

This is a question we received via email at GREAT3 HQ:

would you mind answering two small questions about Great3 :-)? Sorry if this is already explained somewhere; may be I missed something but I could not find that in the Handbook. While testing various PSFEx settings (and as expected), we noticed that the results were quite sensitive to the exact extent of the footprint we would use for the PSF model. In PSFEx, the PSF is modelled inside a disk to minimize directional biases while convolving this model with limited footprint. Now, the Great3 point source samples are cropped inside square images. So here comes my questions!

• while making the galaxy images, did you use areas of the PSF which are not inside those stellar images, and which we would therefore have to "invent" (or say, "recover") ?
• if not then, did you simply convolve the galaxy models with the aforementioned squarish PSF footprint or did you crop the PSF model inside a disk?

Thanks!

[barnabytprowe]

• while making the galaxy images, did you use areas of the PSF which are not inside those stellar images, and which we would therefore have to "invent" (or say, "recover") ?

The answer is basically yes, the images which are shown are a postage stamp view onto the PSF but not technically the entire PSF: if the postage stamps had been larger you would see a little more.

I say a little more because there are actually a couple of technical details to mention... In generating the PSFs we used a two component model. The atmospheric component was a simple Kolmogorov profile, but the optical part of the PSF was constructed from a Zernike model of an aberrated pupil, evaluated using the FFT on a square grid.

Before being convolved with the atmospheric component, the optical component is therefore first generated as a large 2D lookup table of finite size. Once this profile is convolved with a Kolmogorov, it looks like an blurred optical PSF framed in a blurred box made up of the edges of the lookup table.

We checked all of our PSFs quite carefully to ensure that this edge-of-lookup table feature happens beyonds the postage stamp edges to avoid any obvious artifacts in the stars, but while it does do this it generally it does not do so by very much (to save computation time - the bigger the lookup table the slower the image rendering). Therefore, while the PSF does extend beyond the edges of the postage stamp, it does often not do this by very much, and at some point soon after the window.

As the flux in the PSF wings out at the edge of the postage stamp is so low, if I remember rightly we made a crude estimate that the impact of these effects and it seemed to be lower than was important for a stage IV survey.

Sorry that was a long answer! Yes you have to invent / guess / recover / reconstruct stuff, but not very much. Unfortunately the PSF generation part of the problem was actually a real bottleneck in the image generation, and we simply didn't have the computing resources to make the optical PSF grid be twice the galaxy postage stamp image size (which would be the formally correct thing to do).

[barnabytprowe]

@rmandelb added this:

I’ll just add that when we did things like “compare shear estimation if we use the given postage stamp size vs. a larger size” then results for size/shape-related quantities were changing at the 10^-4 level at most, might have been 10^-5. It’s a really tiny effect according to my notes.