First pass at getting fates-sp mode to run without soil BGC #1723
Conversation
ekluzek
added
the
enhancement
…he Fatest SP mode test-mod use the user-mod
ekluzek
added
the
next
Get FATES and carbon_only test-suite tests to pass with MIMICS active To pass with MIMICS active, FATES and carbon_only tests need calculations of: ligninNratioAvg and annsum_npp_col, both currently declared and calculated in CNVegCarbonFluxType.F90. ligninNratioAvg is a function of leafc_to_litter_patch, frootc_to_litter_patch, soilbiogeochem_cwdc_col, soilbiogeochem_cwdn_col, soilbiogeochem_decomp_cascade_ctransfer_col. I got both FATES and carbon_only tests to pass by declaring local copies (in MIMICS) of ligninNratioAvg and annsum_npp_col. Conflicts: src/main/clm_driver.F90
billsacks
removed
the
next
…s' into fates_sp_w_no_soil_decomp
…ake it more clear that it is a soilBGC call, since we don't allow soil_decomp==None for BGC it won't actually matter for any cases we test
subset_data allows zeroing out nonveg landunits without any dompft selected
|
I have this working, but I am showing that one of the externals updates is changing answers to roundoff for some fields for some tests. Such as : SMS_Ld12_Mmpi-serial.1x1_vancouverCAN.I1PtClm50SpRs.cheyenne_gnu.clm-output_sp_highfreq RMS lndImp_Fall_evap 1.0164E-20 NORMALIZED 1.5494E-15 This is likely OK, but I'd like to understand what change causes this. |
… add a preliminary testmod for FatesSp_prescribed mode
…es_sp_w_no_soil_decomp
…act is not true (hence not requiring short circuiting) fixes ESCOMP#1723
… add a preliminary testmod for FatesSp_prescribed mode
…done in aux_clm, for cheyenne_nvhpc since we know this fails
…nto fates_sp_w_no_soil_decomp
|
The roundoff difference for some reason did NOT hold out for final testing except with this test on izumi: SMS_D_Ld1_Mmpi-serial.f45_f45_mg37.I2000Clm50SpRs.izumi_gnu.clm-ptsROA I don't know if that means there's a reproducibility issue, or just an issue with some of my cases. I redid cases with differences many times, with the difference still showing. But, I also wasn't able to track down which externals were causing the issue. So this is all still disconcerting. |
Description of changes
Get FATES-SP mode to run without soil biogeochemistry
Specific notes
Set it up so that fates-SP mode with run with soil_decomp_method == None. I make some progress towards allowing soil decomposition to be off in general in FATES, but it's still hardcoded to assume this for FATES-SP mode and only allow it there.
Also add a user-mod directory for fates_sp
Contributors other than yourself, if any:
CTSM Issues Fixed (include github issue #):
Fixes: #1720
Fixes: #1726
Fixes: #1729
Fixes: #1610
Initial add of FatesSP compset: #1697
Starts adding a test mod for: #1722
Some part of #1736
Are answers expected to change (and if so in what way)? Yes
Any User Interface Changes (namelist or namelist defaults changes)? No
Testing performed, if any: Ran a couple cases with the testmod in place
SMS_Lm4.f45_f45_mg37.I2000Clm51Fates.cheyenne_intel.clm-fates_sp.20220425_151428_o9pw1i
which PASSes
SMS_D.f45_f45_mg37.I2000Clm51Fates.cheyenne_intel.clm-fates_sp
is currently dying with a subscript overflow error
(it looks like we don't run SatPhen with DEBUG on in the fates test list so this probably exists beforehand)