Run fails due to sea ice mesh setting in tag noresm2_5_alpha08d

[jmaerz]

Describe the bug
Please provide a clear and concise description of what the bug is.

• NorESM version: noresm2_5_alpha08d
• HPC platform: betzy
• Compiler (if applicable):
• Compset (if applicable): NOINYOC
• Resolution (if applicable): T62_tn21
• Error message (if applicable): at run-time

ERROR: CICE n, mesh_lon , lon, diff_lon =      108    249.3936783249953      -110.4974239761087            -0.10890D+00   
(abort_ice)ABORTED: 
(abort_ice) called from ice_mesh_mod.F90
(abort_ice) line number          641
(abort_ice) error =  ice_mesh_check:
ERROR: (abort_ice) ice_mesh_check:

To Reproduce
Steps to reproduce the behavior:

1. ${SRCROOT}/cime/scripts/create_newcase --case ${CASEDIR} --res ${GRID} --mach ${MACHINE} --compset ${COMPSET} --project ${PROJECT} --pecount L --run-unsupported
   where:

COMPSET="NOINYOC"
GRID="T62_tn21" 
MACHINE="betzy"

2. cd CASEDIR
3. ./case.setup
4. ./xmlchange BLOM_UNIT=cgs
5. ./xmlchange STOP_OPTION=${PERIOD} with PERIOD="nmonths"
6. ./xmlchange STOP_N=${NSTEP} with NSTEP="1"
7. ./case.build
8. ./case.submit

Expected behavior
A clear and concise description of what you expected to happen.

• Run the simulation for 1 month.

Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
I hope I haven't missed something for the setup that is now needed in noresm2_5_alpha08d (checkout of fresh NorESM and update via git fleximod and compilation went well). If so, please apologize and let me know.

[TomasTorsvik]

I can recreate this bug.

The CICE model should have support for the "tnx2v1" grid, as far as I can see
NorESMhub/NorESM_CICE@fe8bee5

[mvertens]

@jmaerz @TomasTorsvik - there is support for tnx2v1 - but I think what happened is that it was not crashing when there was a mismatch between the internal grid points and the mesh points. I think that a recent update I did to cice made this crash rather than just issue a warning. @JensBDebernard - the ice mesh and ocn mesh are the same for this case - but I'm pointing to cice binary files:

   grid_file = "/cluster/shared/noresm/inputdata/ocn/blom/grid/horiz_grid_tnx2v1_20130206.ieeer8"
   kmt_file="/cluster/shared/noresm/inputdata/ocn/blom/grid/topography_tnx2v1_20130206.ieeei4"

that seem not to be consistent with the mesh file I created:
/cluster/shared/noresm/inputdata/share/meshes/tnx2v1_20230815_cdf5_ESMFmesh.nc

Can someone please point me to a correct SCRIP file for tnx2v1 and I will try generating the tnx2v1 mesh again.

[jmaerz]

Hi @JensBDebernard , would you mind having a look into this? - admittedly, I am not (yet) familiar with the ice grid/meshes, and hence it would take me too long to get into for this question.

[TomasTorsvik]

Moving this from NorESM2.5 milestone. Maybe we can try to fix this for noresm3_0_alpha02?