Problem with Roman WFI PSFs and distortion

[ariedel]

The problems, specifically, are that for SCA03, SCA06, SCA09, SCA12, SCA15, and SCA18, bluer monochromatic PSFs (around 0.8 microns and below) sum to values larger than 1, even when the flux is normalized to 1 at the pupil plane. The problem goes away when distortion effects are switched off.

This was discovered within the context of generating new PSFs for Pandeia, which does use a heavily customized script to create PSFs.