Adjust DEEP2 degrader for LSST filters

[sschmidt23]

The current DEEP2 degrader produces a large population of low redshift interlopers when applied to LSST bands, I'm fairly sure this is because the color cuts as implemented are in done in terms of the CFHT BRI colors, which are somewhat different from LSST gri. A quick color transform and maybe some tweaking of the cut values should produce a selection closer to what is actually observed by DEEP2

Before submitting
Please check the following:

• I have described the purpose of the suggested change, specifying what I need the enhancement to accomplish, i.e. what problem it solves.
• I have included any relevant links, screenshots, environment information, and data relevant to implementing the requested feature, as well as pseudocode for how I want to access the new functionality.
• If I have ideas for how the new feature could be implemented, I have provided explanations and/or pseudocode and/or task lists for the steps.

[sschmidt23]

We can leave the current degrader in case we have access to CFHT12k magnitudes, I'll create an additional degrader called DEEP2_LSST or something for use with gri bands.

[janewman-pitt-edu]

This is a good point. You can find conversions between sdss and 12K BRI in the DEEP2 photometry file headers as I recall.

That said, I am pretty sure the inaccuracy of the colors in the catalog is likely to be a bigger issue (e.g. applying the DESI ELG color cuts to the catalogs give crazy results).

[sschmidt23]

I computed some color corrections based on the CWWSB SEDs and the CFHT and LSST filter curves, results look better with those, though I did also tweak the B-R cut slightly because the low-z interloper fraction was still a bit high, and visually I could see those galaxies seemed to be in that B-R region of space.

And I'm testing with Cardinal data rather than OpenUniverse, since that should have more accurate colors (pending further updates from the CSS group to the diffsky mocks).

This isn't perfect, but it's a much better match than without the transforms, likely good enough for the data challenge to get similar redshift/color coverage (the other rail_astro_tools-implemented degraders look good but not perfect, DEEP2 was the outlier looking quite different from the actual selections).

[janewman-pitt-edu]

OK, Cardinal should be a whole lot better than openuniverse for this.

I checked, it's actually the AB-Vega offsets that are in the file headers. I think one could get the empirical transformations by combining different results from section 4.1 of https://arxiv.org/pdf/1210.2405 (along with LSST-SDSS offsets) but that is likely a lot of work.