Update default data ocean for v12?

[mathomp4]

Recently, @AgilentGCMS discovered that the Reynolds o1 ocean seems to...stop in 2022. If you try running in 2023+ it crashes. I've asked @rcullath to see if this is expected or not, but I think this brings up another question, mainly for @wmputman but also @sdrabenh: Should we change the default data ocean in v12?

At the moment in v12 (and v11), the default data ocean is just Reynolds; from C12 to C5760, it's Reynolds. This seems...dumb. If nothing else, we should probably default to Ostia if we want to default to a lat-lon (I think Ops uses it?) and maybe even CS ocean?

I think we have good remapped ocean data of C90 up for the "usual" uniform grids. The ultra-high-res and stretched are in sort of a worse shape but we can make those if needed.

I could work with @sshakoor1 to fix up the scripts pretty easily to make the change. But just need guidance.

[mathomp4]

Also something @AgilentGCMS reminded me of. Whatever we choose here for a default should also be what remap_restarts.py does. It looks like it might choose CS as the default to remap to, so we should probably echo that here.

A question for @weiyuan-jiang would be, is it possible for that script to default to:

• CS ocean if C90+
• Reynolds or MERRA2 if lower?

[weiyuan-jiang]

Also something @AgilentGCMS reminded me of. Whatever we choose here for a default should also be what remap_restarts.py does. It looks like it might choose CS as the default to remap to, so we should probably echo that here.

A question for @weiyuan-jiang would be, is it possible for that script to default to:

• CS ocean if C90+
• Reynolds or MERRA2 if lower?

Yes, I think it is not hard

[weiyuan-jiang]

The script already did what you want:
default_ = 'CS (same as atmosphere OSTIA cubed-sphere grid)'
if resolution in ['C12','C24', 'C48'] : default_ = '360x180 (Reynolds)'