Activate MALI thermal solver and update configuration to be relevant to both AIS and GIS grids #6559
Conversation
Enable iceberg and small island checkers for stability. Enable thermal solver to allow ice-sheet extent to evolve without issues.
matthewhoffman
added
mpas-ocean
mpas-albany-landice
BFB
matthewhoffman
requested review from
stephenprice,
mperego,
jonbob and
trhille
Also create the corresponding albany_input.*.yaml files
This was not possible when the mesh was added due to issues with the treatment of temperature in MALI. Now that those issues have been resolved, this test can be updated to the newer mesh.
matthewhoffman
force-pushed
the
matthewhoffman/mali/enable_thermal_solver
branch
from
777b747
to
57c057e
Compare
|
@stephenprice , here is the branch we were talking about that enables the thermal solver that you can test out with your various Greenland configurations. @trhille , @stephenprice , @mperego , look over the changes to the namelist defaults and the new fields added to the history files and let me know if there is anything else you think we should include. @mperego and @jonbob , let me know if you are happy with the way I've implemented supporting mesh-specific albany_input.yaml files. @jonbob , I've only tested the test that is updated as part of this PR, so I will need help finding and running all the tests that use MALI. |
|
components/mpas-albany-landice/bld/namelist_files/albany_input.gis_1to10km_r02.yaml
Outdated
Show resolved
Hide resolved
components/mpas-albany-landice/bld/namelist_files/albany_input.gis_20km_r01.yaml
Outdated
Show resolved
Hide resolved
…#6559) Activate MALI thermal solver and update configuration to be relevant to both AIS and GIS grids This PR updates a number of details about the MALI configuration to support both Antarctica and Greenland Ice Sheet configurations: * Enables the MALI thermal solver, which is needed to allow evolving calving fronts, in preparation for enabling calving * Adds new logic to make the albany_input.yaml file specific to a mesh. This is needed because AIS and GIS meshes use different solver boundary conditions that are specified in that file. * Enables config_remove_icebergs and config_remove_small_islands, two options that improve convergence of the velocity solver and are typically used in standalone MALI simulations * Adds additional fields to the MALI history files related to the now activated thermal solver, as well as ice-shelf basal melting. Also updates the MALI globalStats files to write out all global stats instead of the subset that was hardcoded before. * Updates the ERS_Ld5.TL319_oQU240wLI_ais20.MPAS_LISIO_JRA1p5.mpaso-ocn_glcshelf test to the MALI ais8to30 grid that was recently added. When the mesh was added, the test was failing due to issues with the thermal solver, but those have been resolved between MALI thermal solver bug fix and improved iceberg handling #6547 and this PR, so we can now update that test to the newer mesh. [NML] for all configurations with MALI [non-BFB] for all configurations with active MALI
|
Passes:
Also successfully ran new test:
merged to next |
|
merged to master and expected DIFFs and NML DIFFs blessed except for mappy and pm-cpu_developer, which have not yet reported. But the new test is failing -- we probably will need to work on a new PR to change the test layouts as quickly as possible |
|
The new test isn't failing -- a recent CIME update changed the behavior for BFAILs. A fix is being merged and I'll add bless requests for the new test as soon as it changes the dashboard |
This PR updates a number of details about the MALI configuration to support both Antarctica and Greenland Ice Sheet configurations:
[NML] for all configurations with MALI
[non-BFB] for all configurations with active MALI