ndep_nflds needs to be set by CAM/WACCM when ndep is provided so CTSM uses the ndep from CAM

[ekluzek]

What happened?

in current CLM: ctsm5_3_030, ndep_nflds needs to be set by CAM so that CTSM will use the ndep sent by it -- rather than the ndep it reads internally. This needs to be done for cases where ndep is calculated by CAM for sure. Eventually CLM will ensure it's set when coupled to CAM, but CLM needs to evaluate to make sure CAM/DATM are providing all the prescribed scenarios that are needed, and that answers don't change the climate from the current data that's used.

What are the steps to reproduce the bug?

Setup a simple case with WACCM where ndep is going to be calculated, and show that there aren't any settings in the drv_flds_in file for the case.

This was noticed by Mariana on the CTSM side here:

ESCOMP/CTSM#1900 (comment)

./create_test SMS.f19_f19_mg17.FWHIST_BGC.derecho_intel.cam-outfrq9s_waccm_mam4 -r . --walltime 00:20:00 --no-build
cd SMS.f19_f19_mg17.FWHIST_BGC.derecho_intel.cam-outfrq9s_waccm_mam4*
./preview_namelists
grep ndep CaseDocs/drv_flds_in # which shows nothing so the ndep fields won't be used inside of CTSM

I did this in a cesm3_0_alpha06b sandbox

What CAM tag were you using?

cam6_4_067

What machine were you running CAM on?

CISL machine (e.g. cheyenne)

What compiler were you using?

Intel

Path to a case directory, if applicable

No response

Will you be addressing this bug yourself?

No

Extra info

This is probably something that I should talk to whoever is going to do this on the CAM side, so we make sure CAM and CLM do the appropriate thing and will work correctly together.

[ekluzek]

One thing that's happening is that CAM is now always sending Faxa_ndep to CLM, but CLM only uses it if ndep_nflds is > 0 from the drv_flds_in ndep namelist.

[fvitt]

@ekluzek

I guess CTSM is expecting CAM always to set ndep_flds >0 in drv_flds_in and to pass the nitrogen dep fluxes to the mediator (via Faxa_ndep). Correct? This is for all cases where the nitrogen fluxes are prescribed and where the fluxes can be prognosed by the chemistry.

If CAM is always setting the Faxa_ndep field, is there a need to set the ndep options in drv_flds_in?

I would be happy to chat with you about the details.