Add files_for_e3sm step for remapping rms tidal velocity

[xylar]

This field is needed for any meshes that will run in the TMIX configuration described here:
E3SM-Ocean-Discussion/E3SM#76
The field gets remapped from a dataset that @irenavankova has preprocessed from the CATS tidal model,
see https://web.lcrc.anl.gov/public/e3sm/mpas_standalonedata/mpas-ocean/tidal_mixing/README.txt.

Checklist

• Developer's Guide has been updated
• API documentation in the Developer's Guide (api.rst) has any new or modified class, method and/or functions listed
• Documentation has been built locally and changes look as expected
• Document (in a comment titled Testing in this PR) any testing that was used to verify the changes

[xylar]

Testing

I ran the files_for_e3sm test case (the one for pre-existing meshes) and pointed it to the IcoswISC30E3r5 mesh.

I was able to generate the RMS tidal velocity:

My version differs only slightly from @irenavankova's:

The places where they differ are where there is very little overlap between the original grid and the MPAS mesh:

and results from a different approach to renormalization when there is less than 10% overlap between the meshes.

[xylar]

@irenavankova, this is ready for you to try out and review.

We should talk about the renormalization strategy. What I chose to do was to renormalize if there is more than 1% overlap and to set the velocity to zero if there is less.

It seemed like you had a different approach to cells with less than 10% overlap where you always renormalized as if the overlap were 1%. I think my approach is a little more standard and should lead to smoother results but I don't think it matters much because only a handful of cells are affected.

[xylar]

@irenavankova, I sent you an invitation to collaborate here. If you accept, I can add you as a reviewer but you can still comment without being an official reviewer.

[irenavankova]

All works for me

[xylar]

@irenavankova, thanks for testing this out!