Add freshwater tracers

[cbegeman]

This PR adds a freshwater tracer group that tracks the concentration of all surface mass fluxes to the ocean. The nonlocal flux term is not included because it would require a new nonLocalSurfaceTracerFlux struct array to separate the salinity flux into its constituent mass source terms. That term is also not included for other passive tracers.

[cbegeman]

@jonbob When you have time, would you be willing to take a look at this PR and let me know if you see any obvious issues why E3SM builds would fail while MPAS-O builds are successful? I imagine I might have missed something that your namelist-related scripts would pick up.

Here's the build log if that's helpful. I didn't see anything that looked informative to me. /lcrc/group/e3sm/ac.cbegeman/scratch/chrys/SMS_Lm1_P2048.ne30pg2_ECwISC30to60E2r1.CRYO1850-DISMF.chrysalis_intel.G.20240702_105706_xqyt5o/bld/e3sm.bldlog.240702-110049

Thank you so much!

[cbegeman]

@irenavankova Here's the branch in case you'd like to start taking a look at it.

[darincomeau]

@cbegeman I see this in the bld log:

/lcrc/group/e3sm/ac.cbegeman/scratch/chrys/SMS_Lm1_P2048.ne30pg2_ECwISC30to60E2r1.CRYO1850-DISMF.chrysalis_intel.G.20240702_105706_xqyt5o/bld/cmake-bld/core_ocean/shared/mpas_ocn_subgrid.f90(550): 
error #6404: This name does not have a type, and must have an explicit type.   [CONFIG_USE_SUBGRID_WETTING_DRYING]

I don't see that config option anywhere in this PR, are you perhaps running this on top of a different branch?

[jonbob]

good catch, @darincomeau. Let me see if running the scripts to make bld files helps

[jonbob]

@cbegeman -- the problem is that we haven't brought the wetting_drying namelist into E3SM yet, so the

config_use_subgrid_wetting_drying

namelist setting isn't currently available. We can bring it in, but it might take some discussion

[cbegeman]

Thanks, @darincomeau and @jonbob! This PR doesn't depend on wetting and drying at all. How can I get around this? Is there a particular commit/tag I should rebase this onto?

[xylar]

@cbegeman, it looks like that's the base case you're trying to compare this branch to (based on the .G. in the name). So this is with master, right?

Could it be that the test case you're trying to run isn't a good choice? SMS_Lm1_P2048.ne30pg2_ECwISC30to60E2r1.CRYO1850-DISMF.chrysalis_intel? I think we want to be testing with E3SM v3 grids. Even if so, we probably want to know why this test case is broken and fix it.

[xylar]

I've been using:

SMS_D_P480_Ld1.ne30pg2_r05_IcoswISC30E3r5.CRYO1850-DISMF.chrysalis_intel

[xylar]

For what it's worth, I was able to build e3sm.exe for SMS_Lm1_P2048.ne30pg2_ECwISC30to60E2r1.CRYO1850-DISMF.chrysalis_intel with the current master:

/lcrc/group/e3sm/ac.xylar/scratch/chrys/SMS_Lm1_P2048.ne30pg2_ECwISC30to60E2r1.CRYO1850-DISMF.chrysalis_intel.G.20240712_102150_3q1ix1/bld

@cbegeman, it might be worth giving it another try.

[cbegeman]

@jonbob I'm not seeing the new tracer group config options appear in the namelist for e3sm cases. When you have a chance, can you give this branch a look and see if I've missed anything in the e3sm build namelists/scripts?

[jonbob]

@cbegeman -- I'll look now. Oh, you have to add the new group to the groups list in build-namelist

[irenavankova]

Past line 1819 in build-namelist add a new line tracer_forcing_freshwatertracers

[cbegeman]

Thanks for pointing this fix out @irenavankova! 81f93b2

[irenavankova]

@cbegeman, the subglacial runoff is in master now. Do you want me to add the subglacial tracers to this pull request before moving it on?

[irenavankova]

The branch I used for the gcase runs with subglacial runoff tracers is here:

E3SM-Project/E3SM@master...irenavankova:E3SM:sgr_tracers

Probably a bit messy, so let me know if you want me to pick out something specific

[cbegeman]

@cbegeman, the subglacial runoff is in master now. Do you want me to add the subglacial tracers to this pull request before moving it on?

@irenavankova I'll rebase this branch then compare the code against your branch. I'll let you know if I have any questions or issues. Thanks!

[cbegeman]

@irenavankova Can you rerun one of your isomip_plus configurations with subglacial fluxes with this new branch to test that capability? As far as I can tell, there's not an easy way to turn sgr on in this test case. I have rerun an isomip_plus case and it seems to be working as expected for ismf.

[irenavankova]

I ll rerun it, I have a compass branch modified for that. It might not be till tomorrow though.

[cbegeman]

@irenavankova No rush! Thanks!

[irenavankova]

@cbegeman, the isomip case with subglacial tracers runs fine on your branch.

Here is subglacial freshwater tracer in the top layer (first month average):

And here the associated landice freshwater tracer:

[cbegeman]

@irenavankova and @xylar Would you like to see anything further before I migrate this to E3SM? I just ran the e3sm_cryo_developer's suite and all tests pass on chrys.

[irenavankova]

@cbegeman, have you tried to run with DSGR compset to make sure that the tracers work there in the current implementation? I had put the subglacial tracers in onto your branch back in July and have been running fine with that, but I am not sure if you implemented that over or used mine, the now very old branch I was running with is here: https://github.com/irenavankova/E3SM/tree/sgr_tracers

[cbegeman]

@irenavankova Here is the routine that implements the subglacial runoff freshwater tracer if you want to take a look.

[cbegeman]

@irenavankova If the DSGR capability is not tested in the e3sm_cryo_developer's suite, can you suggest a test to run?

[irenavankova]

I am not familiar with the standardized test suits, there was one added for DSGR though:
https://github.com/E3SM-Project/E3SM/blob/c1dca98083c6c3d88ac6f6909bdbd6caa68780d4/cime_config/tests.py#L298

If you want to I can just try to setup and run a case with DSGR manually and check that

[cbegeman]

@irenavankova Thanks for pointing me to that. I'll run the whole e3sm_ocnice_extra_coverage suite

[cbegeman]

@irenavankova All e3sm_ocnice_extra_coverage tests pass.

[irenavankova]

Great, thanks for checking that!

[cbegeman]

@xylar I'm migrating this to E3SM. If you have any suggested testing or feedback you can provide it there.

[xylar]

Sorry, yes, that's fine!