Closed xylar closed 1 month ago
I created a new SORRM r3 (#807) land-ice mask during the cull_mesh
step with this procedure. Whereas before this fix there is a region of isolated open ocean surrounded by land or land ice near Stange Ice Shelf:
this is filled in with land-ice with this fix:
Results are in:
/lcrc/group/e3sm/ac.xylar/compass_1.4/chrysalis/test_20240718/fix_land_ice_flood_fill/ocean/global_ocean/SOwISC12to30/mesh/cull_mesh
My actual preferred fix is to add a critical passage to keep this location (called the Carroll Inlet) open, see https://github.com/MPAS-Dev/geometric_features/pull/203. But the algorithm here is behaving consistently with what we want the general solution to these kinds of issues to be.
Thanks @darincomeau!
We need to address land-locked and land-ice-locked cells within the process for flood-filling to ensure that the land-ice mask is connected. Otherwise, land-ice-locked cells added to the land-ice mask can create new regions of open ocean within the land ice.
Checklist
Testing
in this PR) any testing that was used to verify the changes