Refactoring xwrf

[lpilz]

          @lpilz - I have a sample dataset we can pull in here - I wonder if we just make the projection optional as well, as we do with other postprocessing steps?

Originally posted by @mgrover1 in #168 (comment)

Max suggested that we make the projection/grid generation optional, too. I think this issue would be a good opportunity to discuss whether we want to refactor/add some other things, too. I have a couple of ideas:

• Splitting projection and grid generation (if no projection, grid is just lat/lon with cen_lat/cen_lon as center)
• Add a table of base diagnostics (with formulas and references) to docs
• pint apparently doesn't support factors in units anymore, so we should fix that for ppm units (maybe separate issue?)
• make other .postprocess steps optional, too
• Maybe look at [MISC]: More careful consideration of different xarray options #98 and [FEATURE]: Pull in coordinates from separate file #36 again

Do we want additional features in xwrf like plotting routines, adding sophisticated interpolation support (one feature I think would be cool is interpolation between different CRS using xugrid), some interactive data exploration? Or do we want xwrf to stay as lightweight as possible?

[mgrover1]

@lpilz - these are all great points and questions - I think it would be helpful to setup a meeting to discuss a roadmap for a "1.0" version of xwrf... do you think this would be of interest?

[lpilz]

Yes, I think that would be great. Not sure who would be interested in joining but I'd be happy to have this discussed for sure :)

[lpilz]

Hey @xarray-contrib/xwrf-devs is anybody interested in joining this discussion? :)

[jthielen]

I'd definitely be interested, but unfortunately I don't have great availability until the second half of June (so if you need to move ahead on this sooner, don't worry about me).

[mgrover1]

Could folks interested please fill out this availability poll?
https://www.when2meet.com/?25256760-mOpoG