Automatically set update_ocn_f = .true. when coupling with MOM

[gustavo-marques]

Is it possible to automatically set update_ocn_f = .true. when coupling with MOM?
I thought this had already being done but in my latest cases with cesm2_3_alpha05c, update_ocn_f = .false. by default.
pinging @alperaltuntas and @dabail10

[dabail10]

Not sure what happened here. I don't have a record of this changing in namelist_defaults_cice.xml. Was this in the MOM scripts? I know we talked about keying off the grid, but I felt like this was not the best way. Ideally, it should key off comp_ocn = mom6 or whatever. I thought we lumped this in with tfrz_option?

[alperaltuntas]

I just sent a PR that fixes this issue. As for tfrz_option, are we happy with mushy for mom6 coupling, or should it be compset dependent?

[gustavo-marques]

@alperaltuntas, thank you very much. I do not know what should be done for tfrz_option so I will let @dabail10 answer that.

[dabail10]

I'm pretty sure you do not have tfrz_option = 'mushy' in MOM. So, I think we had to do linear_salt. We get this from the mediator:

call NUOPC_CompAttributeGet(gcomp, name="tfreeze_option", value=tfrz_option, isPresent=isPresent, rc=rc)
if (ChkErr(rc,__LINE__,u_FILE_u)) return
if (.not. isPresent) then
   tfrz_option = 'linear_salt'  ! TODO: is this right? This must be the same as mom is using for the calculation.
end if

[gustavo-marques]

I can confirm that TFREEZE_SALTWATER_OPTION = linear_salt in my out-of-the-box MOM cases. This is a CIME variable. To clarify, does that overwrites tfrz_option in CICE?

[dabail10]

Yes. It is done in the cap. I thought update_ocn_f was also overwritten in the cap, but it does not appear to be. I think this is the way we had wanted to go. Nonetheless, give me a few weeks. I will be switching CICE6 and CICE5 as the defaults at which point I will make sure that update_ocn_f is set correctly.