generalize logic for coupling frazil fluxes #458
Conversation
columnphysics/icepack_therm_itd.F90
Outdated
| @@ -2140,6 +2138,9 @@ subroutine icepack_step_therm2 (dt, ncat, nltrcr, & | |||
| !----------------------------------------------------------------- | |||
|
|
|||
| if (icepack_chkoptargflag(first_call)) then | |||
| if (present(update_ocn_f)) then | |||
| call icepack_init_parameters(update_ocn_f_in=update_ocn_f) | |||
| endif | |||
There was a problem hiding this comment.
This new code needs to be outside icepack_chkoptargflag. That interface is for checking whether optional arguments that are needed are passed, and the settings reflect never checking to always checking. The icepack_init_parameters call should be outside that if block. If a user passes update_ocn_f in, then the code should update the value in parameters. Otherwise, just continue. And it might vary by gridcell or timestep (even if it shouldn't).
| @@ -369,6 +369,9 @@ forcing_nml | |||
| "", "``minus1p8``", "constant ocean freezing temperature (:math:`-1.8^{\circ} C`)", "" | |||
| "", "``mushy``", "matches mushy-layer thermo (ktherm=2)", "" | |||
| "``trestore``", "integer", "sst restoring time scale (days)", "90" | |||
| "``cpl_frazil``", "``external``", "frazil water/salt fluxes are handled outside of Icepack", "``fresh_ice_correction``" | |||
| "", "``internal``", "send full frazil water/salt fluxes for mushy physics", "" | |||
| "", "``fresh_ice_correction``", "correct fresh-ice frazil water/salt fluxes for mushy physics", "" | |||
There was a problem hiding this comment.
These should be alphabetical to be consistent with the rest of the document (hopefully). So external, fresh_ice_correction, then internal. Should just have to swap the order of internal and fresh_ice_correction is you want to follow that format.
|
Base_suite tests passed. @apcraig please go ahead and run your complete set of tests -- thank you. I'll begin porting the driver changes into CICE. |
|
Full test suite run on cheyenne for Icepack and CICE (with the Icepack updates only). Everything is bit-for-bit and looks good. |
|
@eclare108213, I think this can be merged if you feel it's ready. |
Draft PR for sanity review and further testing in Icepack and CICE.
Adds a new namelist parameter (cpl_frazil) and generalizes the logic for coupling frazil fluxes
@eclare108213
These changes have been tested in E3SM.
Base_suite with regression:
178 measured results of 178 total results
178 of 178 tests PASSED
0 of 178 tests PENDING
0 of 178 tests MISSING data
0 of 178 tests FAILED
https://github.com/CICE-Consortium/Test-Results/wiki/e4bc61184b.chicoma.intel.23-09-11.142102.0
Adds namelist parameter cpl_frazil, needed to enable E3SM's style of coupling.
Uses update_ocn_f in icepack_parameters rather than icedrv_flux.
Removes update_ocn_f from icedrv/step_therm2 argument lists and makes it optional in icepack_step_therm2.
Changes frazil dfresh, dsalt logic to include cpl_frazil option but the calculations should be BFB backward compatible.
Updates interfaces.include and documentation.