Issue with Long-Term CTSM Run (ctsm5.3.025): CN Soil Matrix Solution Off and Simulation Fails Beyond One Year

[uzairrahil]

What version of the code are you using?
ctsm5.3.025


Have you made any changes to files in the source tree?
- Describe any changes (code, xml files, etc.)
No, I have not changed.


Describe every step you took leading up to the problem:

1----------------------------------------------------------------------------------------------Creating the Regional Case:
cd /glade/work/rahilmoh/CLM/Model/my_ctsm/cime/scripts

./create_newcase --case /glade/work/rahilmoh/CLM/Cases/MI27_121y109x_MI_IHistClm50BgcCrop_0.05d --res CLM_USRDAT --compset IHistClm50BgcCrop --project UMSU0017 --run-unsupported

cd /glade/work/rahilmoh/CLM/Cases/MI27_121y109x_MI_IHistClm50BgcCrop_0.05d
./case.setup
scp ../MI19_115x108_MI_I2000Clm50BgcCrop_0.05d_2000/6_update_env_run.sh ./ # just to request resources


scp ../MI09_115x108_MI_I2000Clm50BgcCrop_0.05d_2000/SourceMods/src.clm/*.F90 SourceMods/src.clm/

./6_update_env_run.sh
./case.setup --reset


./xmlquery -p MOSART_MODE
./xmlchange MOSART_MODE=ACTIVE


*******
./xmlquery -p DATM to see the DATM_YR_ALIGN and DATM_YR_END, set it to the same as the forcing. -- Thanks to Aman.
./xmlchange DATM_YR_ALIGN=1980
./xmlchange DATM_YR_END=2022

qcmd -A UMSU0017 -- ./case.build
Now i modified the runtime in env_run_xml to run from 2000-2009.

./case.submit

2--------------------------------------------------------------------- user_nl_clm --------------------------------------
I am using Conus404 data prepared from 1980-2022, and the relevant files are mentioned in the user_nl_datm-stream file along with the meshfile path.
I have also created my own surface data and land use time series; the paths of these two are mentioned in the user_nl_clm.

fsurdat='/glade/work/rahilmoh/CLM/Model/my_ctsm/tools/mksurfdata_esmf/surfdata_LM_Rahil_5km_hist_1980_2022_78pfts_c250520.nc'
flanduse_timeseries = '/glade/work/rahilmoh/CLM/Model/my_ctsm/tools/mksurfdata_esmf//landuse.timeseries_LM_Rahil_5km_hist_1980-2022_78pfts_c250520.nc'

Detailed user_nl_clm:

limit_irrigation_if_rof_enabled = .true.
use_groundwater_irrigation = .false.

lower_boundary_condition = 3
use_bedrock = .false.
soilwater_movement_method = 1

hist_empty_htapes = .true.

! Monthly average output for both streams
hist_nhtfrq = 0, 0

! Stream 1: 1 file per year (grid-averaged output)
hist_mfilt = 12, 12

! Stream 1: grid-averaged output (.true.), Stream 2: subgrid/PFT-level output (.false.)
hist_dov2xy = .true., .false.

! Stream 1: all your standard variables (grid-averaged)
hist_fincl1 = 'EFLX_SOIL_GRND','Qh', 'FSH', 'FCEV', 'FGEV', 'FCTR', 'FSAT',
'EFLX_LH_TOT', 'H2OCAN', 'H2OSNO', 'H2OSOI',
'HK', 'QDRAI', 'QFLX_EVAP_TOT', 'QFLX_EVAP_VEG', 'QFLX_SNOW_GRND', 'QFLX_LIQ_GRND',
'QH2OSFC', 'QINFL', 'QINTR', 'QIRRIG_FROM_SURFACE', 'QIRRIG_DEMAND',
'QIRRIG_DRIP', 'QIRRIG_SPRINKLER',
'QDRAI_PERCH',
'Qle', 'TBOT', 'GRAINC',
'RH', 'Tair', 'WIND', 'FLDS', 'FSDS', 'QBOT',
'QOVER', 'QRUNOFF', 'SNOWLIQ', 'QSOIL', 'QTOPSOIL', 'QVEGE', 'QVEGT',
'RAIN', 'SNOW', 'RAIN_FROM_ATM', 'SNOW_FROM_ATM',
'VOLR', 'VOLRMCH', 'TWS', 'TOTSOILLIQ',
'ZWT', 'ZWT_CH4_UNSAT', 'ZWT_PERCH', 'H2OSOI',
'TLAI', 'GPP', 'QCHARGE'

! Stream 2: subgrid (PFT) level output for GRAINC_TO_FOOD,
hist_fincl2 = 'GRAINC_TO_FOOD'


Describe your problem or question:

The cesm.log file shows that the "CN Soil matrix solution is off." Although the case successfully runs for one year and produces output, it fails to run for longer periods, such as the 2000–2009 simulation I initially intended. I want to run this case for 1980–2022, but for now, I’m trying to get it to work for the 2000–2009 period.

The log files are attached.

Thank you so much for your support.

Sincerely,

Mohammad Uzair Rahil
MSU

 

[slevis]

I do not see the log files. Regardless, it's a good sign that the simulation works for a year. Then you can troubleshoot from something that initially works.

 

[uzairrahil]

Dear Sir Slevis,

Thank you very much for your response.

The case now runs successfully for a longer period, which I realized was due to not setting the DATM_ALIGN parameter correctly with my CONUS404 data earlier and I also changed the ctsm model to an earlier version.

However, I’ve encountered a new issue: the streamflow output from the IHISTBGC compset appears to be significantly lower than expected. Previously, when I used the I2000BGCCrop compset, the streamflow ranged from 0 to several thousand CMS. In contrast, the IHISTBGC compset is now producing streamflow values in the range of 0 to a few hundred CMS, which is much lower than the corresponding USGS observations.

This discrepancy arises even though I’m using dynamic land use time series and the correct surface data in the IHIST configuration.

/glade/work/rahilmoh/CLM/Cases/MI34_121y109x_MI_IHistClm50BgcCrop_0.05d_CTSM_aman_version
This is the case directory.

I would greatly appreciate any insights or suggestions you might have regarding this issue.

Best regards,
Rahil

 

[slevis]

I do not know why the streamflow decreased from the I2000BgcCrop to the IHistBgc compset. Maybe because precipitation is different?

However, we do not recommend using the river routing model in any configuration other than global. If you search through the Forums, you will find past threads discussing this.