New Mesh: SOwISC12to30E3r4

[xylar]

Long name: SOwISC12to30kmL80E3SMv3r4

This version of the Southern Ocean Regionally Refined Mesh (SORRM) has resolution that is:

• 12 km resolution around Antarctica
• 30 km elsewhere

It is intended to be similar to the Icos30 mesh except in the Southern Ocean and around Antarctica.

This mesh differs from r3 in #807 in that:

• MALI topography as well as masks for land ice, floating ice and ocean get combined with BedMachine Antarctica/GEBCO.

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
• The E3SM-Project submodule has been updated with relevant E3SM changes
• Document (in a comment titled Testing in this PR) any testing that was used to verify the changes

[xylar]

This is based off of #807 (which is based off of #813) so it will be a bit before this is ready to go in. It also needs documentation.

[xylar]

Comparison with SOwISC12to30E3r3

Here are the landIceMask fields for the SORRM mesh with MALI topography:

And the same but with BedMachine from #807:

[xylar]

Updated Comparison with SOwISC12to30E3r3

Here are the landIceMask fields for the SORRM mesh with MALI (AIS_2to10km) topography:

And the same but with BedMachine from #807:

[xylar]

I was able to run the workflow through the performance test. In a discussion with @cbegeman, @matthewhoffman and @trhille, we decided we'd like to work with a 4km initial condition that's more similar to the 2km one I used in the plots above, and that that initial condition would also be swapped into E3SM-Project/E3SM#6440, the PR bringing in the new MALI AIS meshes and initial conditions.

[xylar]

Another Updated Comparison with SOwISC12to30E3r3

Here are the landIceMask fields for the SORRM r4 mesh with MALI (AIS_4to20km) topography:

And the same but SORRM r3 with BedMachine topography from #807:

[xylar]

The same as above but zooming in on the Amundsen Sea.

SORRM r4:

SORRM r3:

[cbegeman]

@xylar have you been looking at all at landIceFloatingFraction?

[xylar]

The land-ice mask for the Amundsen sea is looking reasonable with the latest MALI topography dataset (AIS_4to20km_20241118_relaxation_0TGmelt_10yr.20241224.nc):

[xylar]

I made a version of the culled MPAS-Ocean mesh that includes grounded ice below sea level:

It looks like this adds only about 5% more cells (even though all added cells are in the high-res region). That certainly seems acceptable.

[cbegeman]

@xylar Great! I'm wondering about the distribution of 0 < landIceGroundedFraction < 1. It looks from the figure above that these cells are only along the transition between grounded and floating ice or open ocean, and not along the transition from grounded ice to land above sea level. Is this intentional?

[xylar]

@cbegeman, I think we do a flood fill to remove all land-ice-to-land so, yeah, that transition is intentionally not there and I don't think we want it.

[xylar]

I think that might have been one of the important fixes I had to make in SORRMr3 -- if I accidentally didn't eliminate the land-ice-to-land transition, I ended up with weird ice drafts above sea level in some cases because of interpolation between areas that are above and below sea level.

[cbegeman]

On giving this a little more thought I agree that the hard transition is what we want. Thanks for explaining the reasoning a bit more.