PFT_CROOP in Transition run

[Esther_Bender]

I am running a stand alone CLM simulation, and i am Trying to run a single point Transition run from 1901 to 2001 using my own atmospheric data set.
I also created my own Landuse file which somehow is causing troubles.
I successfully run a small simulation, bur rerunning the same simulation (same setup script, same input data) the model suddenly stops, while getting the PFT_CROOPS from the land use file. It is the same value as it used in the Surface data and i am running a cold start case.
This happened to me a couple of times now, that sometimes the model gets the value and sometimes it doesn't even with using the same script.
Is there anything i can do about that? Or does anyone know what could cause the problem?

I am getting the following error in the cesm.log file: Caught signal 11 (Segmentation fault: address not mapped to object at address 0x1cd5871974)

I am using the following script to set up the run:


"
#!/bin/bash

#create a single point run using PTSm mode with ExcessIce

cd ~/CTSM_ExcessIce/cime/scripts


echo "create case" $1


./create_newcase --machine fram --case ~/cases/$1 --res CLM_USRDAT --compset HIST_DATM%GSWP3v1_CLM50%BGC_SICE_SOCN_SROF_SGLC_SWAV>
--run-unsupported --user-mods-dirs /cluster/projects/nn2806k/estherbe/Input_Data/1x1Iskoras --project nn2806k \
--compiler intel --handle-preexisting-dirs r --walltime 01:00:00 -q normal
#--queue devel


cd ~/cases/$1

cp /cluster/projects/nn2806k/estherbe/Input_Data/1x1Iskoras/user_mods/user_nl_datm_streams ~/cases/$1


#./case.setup

#change long and lat parameters
#for samalovy

./xmlchange STOP_N=25
./xmlchange STOP_OPTION=nyears
./xmlchange PTS_LON=24.250
./xmlchange PTS_LAT=69.383
./xmlchange RESUBMIT=3
./xmlchange NTASKS=1
./xmlchange RUN_TYPE=startup
./xmlchange JOB_WALLCLOCK_TIME="04:00:00"
./xmlchange RUN_STARTDATE="1901-01-01"
./xmlchange DATM_YR_ALIGN="1901"
./xmlchange DATM_YR_START="1901"
./xmlchange DATM_YR_END="2001"
./xmlchange CLM_ACCELERATED_SPINUP="off"
./xmlchange CLM_USRDAT_DIR=1x1Iskoras
./xmlchange CLM_USRDAT_NAME=1x1Iskoras


./xmlchange LND_DOMAIN_PATH=1x1Iskoras
./xmlchange LND_DOMAIN_FILE="domain.lnd.fv0.9x1.25_gx1v7_Iskoras_c240325.nc"
./xmlchange DIN_LOC_ROOT_CLMFORC=1x1Iskoras

cat > user_nl_clm << EOF
use_excess_ice = .true.
use_excess_ice_tiles = .true.
use_excess_ice_streams = .true.
use_tiles_snow=.true.
use_tiles_lateral_water=.true.
excess_ice_split_factor=0.9
soil_layerstruct_predefined = '49SL_10m'
use_tiles_lateral_heat=.true.
fsurdat = '/neu_cdf5.nc'


flanduse_timeseries='Landuse_cdf5.nc'
hist_dov2xy=.false.
hist_nhtfrq=0
hist_mfilt=12
hist_fincl1

EOF
./case.setup

echo "BUILDING CASE"

./case.build

./case.submit

 

[oleson]

I assume this is a typo in your post, but there is not a PFT_CROOP variable in CLM, there is a PCT_CROP variable. If that is not the problem, then can you attach all of your log files? What version of CLM are you running?

 

[Esther_Bender]

Hei,
Yes that's a typo, I meant the PCT_CROP variable. I attached the log files and also the Landuse_change_file
I am running the CTSM 5.1. version with some adaptions.

Also is it possible to to a transient run without land-use change? as the land-use is not changing anyway i could also just turn it off if possible

 

[oleson]

Thanks for the files. I don't see anything obviously wrong. I guess make sure that PCT_CROP appears in your landuse file as:

double PCT_CROP(time, lsmlat, lsmlon) ;
PCT_CROP:long_name = "total percent crop landunit" ;
PCT_CROP:units = "unitless" ;

I'm not sure how the txt file from the landuse file you attached was generated, but I see:

PCT_CROP
_FillValue ø long_name total percent crop landunit units unitless "

which doesn't indicate any dimensions, although I don't see dimensions for other variables either.

Yes, you can turn off transient landuse in a transient run. To do that, I think you can just set:

flanduse_timeseries = ' '

in your user_nl_clm.

Your lnd_in would look like:

&dynamic_subgrid
flanduse_timeseries = ' '
/

Might look different depending on your version of the model.