Skip to content

Conversation

@glemieux
Copy link

No description provided.

@glemieux glemieux force-pushed the lnd/fates-bc-refactor-patch-direct branch from 1a338a0 to c6c5b78 Compare July 2, 2025 16:37
rljacob and others added 29 commits August 13, 2025 23:08
Remove rof_comp_esmf.F90 from MOSART. Never used.
Remove lnd_comp_esmf.F90 from ELM.  Never used.
Remove atm_comp_esmf.F90 from EAM.  Never used.
[BFB] for most resolutions (NE30, NE256, NE1024), may be CC for some tests
Use the newly installed ADIOS2 libraries under /ccs/proj/cli115 to
avoid potential system enforced purges on /lustre/orion/cli115.

[BFB]
If ekat or eamxx's main CMakeLists change, it's probably
a good idea to test
Remove cime_model=cesm options in namelist_definition_drv that will
never be used.  Should have no "cime_model" conditions.

Also remove some NUOPC options.
Fix a few comments.
…roject#7605)

EAMxx: fix scripts tests

Fix issue in scripts tests following ekat update.

[BFB]
In buildlib.spio, NCZarr support was previously detected using:
netcdf4_nczarr_string = "NetCDF_C_HAS_NCZARR:BOOL=TRUE"

On some machines like Perlmutter and Aurora, NCZarr is enabled by
default in the NetCDF libraries, so this variable is always TRUE.
This causes buildlib.spio to incorrectly assume that NCZarr support
is enabled in SCORPIO.

Since NCZarr support in SCORPIO is still experimental and disabled
by default, detection should instead rely on the PIO_ENABLE_NCZARR
configuration flag.

Updated search string:
netcdf4_nczarr_string = "PIO_ENABLE_NCZARR:BOOL=ON"
Remove code that is unused in both mct and moab drivers.

We never used or tested NUOPC and NUOPC_TIGHT coupling sequence options so remove them from driver
and namelist options.
They were used by CESM while developing the NUOPC-based coupler.

Remove test for (trim(cime_model) == 'e3sm') in seq_flds_mod.F90. This is always true.
Remove cime_model=cesm options in namelist_definition_drv because its never true.

Remove config_components_cesm.xml from driver/cime_config. This is never read.

Remove HAVE_MOAB ifdefs from driver-moab.

Also remove model_comp_esmf.F90 from EAM, ELM and MOSART. This never-used code is left over
from an attempt to build a coupler with ESMF (before NUOPC existed).

[BFB]
This will be handled by the fates site initialization routine
While the associated boundary condition is currently set with a min check against this and nlevsoil, this check should be moved inside fates
Since the max rooting depth is calculated using multiple bc_in variables, handle the logic on the fates side
This will be handle on the fates-side
These have moved into fates
… can/should take place during init_coldstart instead of init

This will mostly be the non-global data type variables
Since we pass vert_soilc namelist tag, we can replace this init_soil_depths logic on the fates side
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.