Add isunnguata sermia mesh generation test case

[trhille]

Add a new issunguata_sermia test case with a mesh_gen test group.

Checklist

• User's Guide has been updated
• 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

[trhille]

force-push from 1bddfd9 to 728cc4a was a rebase onto compass/main to incorporate changes after merge of #590.

[trhille]

Testing

Running mesh_gen created the mesh without issues. It's closer to 0.8–8 km than 1–10km, but at only 3900 cells this doesn't seem like a major issue.

[matthewhoffman]

This looks pretty standard, so I don't have any concerns with the implementation.

Two observations:

• Official name of this glacier has two n's and one s: Isunnguata Sermia (https://tc.copernicus.org/articles/9/2215/2015/tc-9-2215-2015.pdf)
• The cell spacing near the terminus looks strange and not similar to the velocity field. Do you know why that is?

I'm happy to entertain your opinion about to what extent either of these items should be corrected.

[trhille]

Thanks @matthewhoffman. I'm happy to change the name of the test case. Good catch on the velocity-cell spacing mismatch. I believe this stems from the 8km dataset that is used to set cell spacing, while the 1km dataset is used to populate the velocity field after the mesh has been created and culled. I notice that this also leads to an issue with ice thickness abutting the mesh boundary, so I'll fix that.

[trhille]

Force-push was a rebase onto origin/main in order to get new python cell culler.
Changing from the 8km to 2 km data set to define cell width results in much better cell spacing:

Note: the gridSpacing variable is now identical to the cellQuality variable. This seems like a bug introduced elsewhere in COMPASS.

[xylar]

Note: the gridSpacing variable is now identical to the cellQuality variable. This seems like a bug introduced elsewhere in COMPASS.

That would probably be in MPAS-Tools in the cell culler? If you want to follow up, that would be appreciated. But I also don't think we use the cell quality for anything...

[matthewhoffman]

@trhille , this looks ready to merge with exception of correcting the glacier spelling (which is a lot of changes in file names, variables, docs, etc.). Do you want to do that or leave it as is? My preference would be to fix it, and I'm happy to help do that.

[trhille]

@matthewhoffman, thanks for the reminder. I'll fix that.

[trhille]

@matthewhoffman, I went through and fixed that spelling error, and also updated the gridded data set being used to avoid the negative basal heat flux error. This is ready for you whenever.

[matthewhoffman]

Thanks for the updates, @trhille ! I confirmed I can reproduce the same result.