Improved model for MIRI cruciform artifact, part 2

[mperrin]

[Relocated from https://github.com/spacetelescope/webbpsf/pull/838]

This is the second portion of an improved model for the MIRI cruciform artifacts, began in #837 . This PR should stay draft and not be merged until after #837 is merged; this branch is branched off of the PR #837 branch. Besides it's still more of a work in progress.

This PR adds an attempt at modeling the peaks or bumps within the cruciform, as predicted in Gaspar et al. 2021. See their Figure 12 in particular.

I have attempted to take the values stated in section 4.3 of that paper, for the m=4 to 10 order peaks, and apply some scale factors to get models of those in webbpsf, and tune a bit to match the empirical PSFs. This is complex and messy however because the physics is very complex, there are many free parameters in this ad hoc parameterizatation, and the available MIRI ePSFs are of modest SNR. Not sure how tractable this is; for now it's a work in progress.

Motivated in part by seeking to understand the complex diffractive structure seen in dispersed LRS PSFs at short wavelengths.

Examples

• Data: miri_epsfs_F560W_measured.pdf
• With PR #837: miri_epsfs_F560W_sim_new.pdf
• with this PR: miri_epsfs_F560W_sim_new_v2.pdf