It might be useful to proceed with a layer coordinate PanAntarctic config while we try to sort out the ALE coord issues in the idealised cases.
A plan for this could be:
Implement a panan-01 case without ice shelf cavities using parameters from Matt Harrison's 1/2deg panan case. See how this compares to our usual panan-01. Does it degrade the simulation much?
It might be useful to proceed with a layer coordinate PanAntarctic config while we try to sort out the ALE coord issues in the idealised cases.
A plan for this could be: