Update CMEPS - Githubissues

DeniseWorthen commented 4 years ago

Issue #1

Removes use of ifrac_nstod in med_fraction_set

Issue #2

Implements correct mapping for fractional land mode when coupling_mode is set to 'nems_frac'

Additional changes:

Creates separate DataInitialize and prep_ocn code blocks for nems modes (nems_orig, nems_frac or nems_orig_data) which are distinct from corresponding CESM modes.
Implements bulkflux fix for datm as in NEMS (PR # 69)
Adds field dictionary changes to fd_nems.yaml for CICE6.
Removes aoflux associated field bundles from nems_orig and nems_frac modes.
Minor cleanup of mis-spellings or other text.

New baselines will be required.

Related PR: ufs-s2s-model

DeniseWorthen commented 4 years ago

Baselines are running on Hera and Orion.

DeniseWorthen commented 4 years ago

New Baselines are created on Hera and Orion. Ready for merge.

DeniseWorthen commented 4 years ago

The changes to med.F90 were tested on cheyenne and did not change answers. When I ran the RT against the new (20200701) baselines I had created for this commit, the answers for some of the tests changed: cpld_fv3_ccpp_384_mom6_cice_cmeps_1d_bmark_rt cpld_fv3_ccpp_384_mom6_cice_cmeps_2d_atm_flux cpld_fv3_ccpp_384_mom6_cice_cmeps_ww3_1d_bmark_rt.

I am now testing in a sandbox with and without this med.F90 change; I have not yet identified where the answers are changing. My initial tests in my sandbox are that all answers are reproducible.

DeniseWorthen commented 4 years ago

I believe this can be explained by an error in the RT dependency when creating a new baseline. The cold mediator re-starts were not being copied in from the newly created baseline but from the previous baseline. @MinsukJi-NOAA has identified a fix and once that is committed, I will re-create the baselines.

DeniseWorthen commented 4 years ago

@junwang-noaa : My compile on Orion failed with the change to the Makefile. The compile on Hera was fine. Did you test the RT on both machines?

I would like to revert this change so I can get the baselines made and the S2S updated while we figure this out.

The error on Orion was:

med.F90(746): error #6580: Name in only-list does not exist or is not accessible. [ESMF_STATESET] use ESMF , only : ESMF_StateSet, ESMF_StateIntent_Import, ESMF_StateIntent_Export ----------------------^ med.F90(780): error #6632: Keyword arguments are invalid without an explicit interface. [STATEINTENT] call ESMF_StateSet(is_local%wrap%NStateImp(n), stateIntent=ESMF_StateIntent_Import, rc=rc) --------------------------------------------------------^ med.F90(780): error #6632: Keyword arguments are invalid without an explicit interface. [RC] call ESMF_StateSet(is_local%wrap%NStateImp(n), stateIntent=ESMF_StateIntent_Import, rc=rc) ---------------------------------------------------------------------------------------------^ med.F90(790): error #6632: Keyword arguments are invalid without an explicit interface. [STATEINTENT] call ESMF_StateSet(is_local%wrap%NStateExp(n), stateIntent=ESMF_StateIntent_Export, rc=rc) ---------------------------------------------------------^ med.F90(790): error #6632: Keyword arguments are invalid without an explicit interface. [RC] call ESMF_StateSet(is_local%wrap%NStateExp(n), stateIntent=ESMF_StateIntent_Export, rc=rc) ----------------------------------------------------------------------------------------------^ compilation aborted for med.F90 (code 1)

junwang-noaa commented 4 years ago

It should have no impact on Orion since the ESMF version is still 800 in s2s. I will test it on Orion with the top of s2s.

On Wed, Jul 8, 2020 at 4:32 PM Denise Worthen notifications@github.com wrote:

@junwang-noaa https://github.com/junwang-noaa : My compile on Orion failed with the change to the Makefile. The compile on Hera was fine. Did you test the RT on both machines?

I would like to revert this change so I can get the baselines made and the S2S updated while we figure this out.

The error on Orion was:

med.F90(746): error #6580: Name in only-list does not exist or is not accessible. [ESMF_STATESET] use ESMF , only : ESMF_StateSet, ESMF_StateIntent_Import, ESMF_StateIntent_Export ----------------------^ med.F90(780): error #6632: Keyword arguments are invalid without an explicit interface. [STATEINTENT] call ESMF_StateSet(is_local%wrap%NStateImp(n), stateIntent=ESMF_StateIntent_Import, rc=rc) --------------------------------------------------------^ med.F90(780): error #6632: Keyword arguments are invalid without an explicit interface. [RC] call ESMF_StateSet(is_local%wrap%NStateImp(n), stateIntent=ESMF_StateIntent_Import, rc=rc)

---------------------------------------------------------------------------------------------^ med.F90(790): error #6632: Keyword arguments are invalid without an explicit interface. [STATEINTENT] call ESMF_StateSet(is_local%wrap%NStateExp(n), stateIntent=ESMF_StateIntent_Export, rc=rc) ---------------------------------------------------------^ med.F90(790): error #6632: Keyword arguments are invalid without an explicit interface. [RC] call ESMF_StateSet(is_local%wrap%NStateExp(n), stateIntent=ESMF_StateIntent_Export, rc=rc)

----------------------------------------------------------------------------------------------^ compilation aborted for med.F90 (code 1)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/NOAA-EMC/CMEPS/pull/3#issuecomment-655740985, or unsubscribe https://github.com/notifications/unsubscribe-auth/AI7D6TNXN5SEHFNVS2XAYYDR2TJVXANCNFSM4OLNXJMQ .

DeniseWorthen commented 4 years ago

New baselines have been created for both orion and hera. @junwang-noaa are we ready to merge this branch to develop?

NOAA-EMC / CMEPS

Update CMEPS #3