Single Point Quick Start
The instructions below are tested with ctsm5.1.dev098
, and ran on fram
machine.
Change the machine and module names and paths accordingly.
The instructions assume CTSM repo is cloned in ~/ctsm_escomp
.
```{keypoints} Note
Check out [this tutorial](https://ncar.github.io/CTSM-Tutorial-2022/notebooks/Day2a_GenericSinglePoint.html) by NCAR for more details.
Note that the tutorial is written for NCAR's machines.
```
Create data for a point
If you haven’t set up CTSM yet, check out steps 1 to 8 on the quick start guide.
Load the required modules: If any of the modules in the following list is not available, search for the module name with
module spider <module_name>
and use the latest available version instead.module load git/2.33.1-GCCcore-11.2.0-nodocs module load Anaconda3/2019.07
Set up and Install dependencies for
subset_data
:eval "$(/cluster/software/Anaconda3/2019.07/bin/conda shell.bash hook)" conda create --name subset_data python=3.9 xarray netcdf4 -y conda activate subset_data
Update the input data paths for
./subset_data
:sed -i 's/glade\/p\/cesmdata\/inputdata/cluster\/shared\/noresm\/inputdata/' ~/ctsm_escomp/tools/site_and_regional/default_data.cfg sed -i 's/glade\/p\/cgd\/tss\/CTSM_datm_forcing_data/cluster\/shared\/noresm\/inputdata\/atm\/datm7/' ~/ctsm_escomp/tools/site_and_regional/default_data.cfg
Generate input data for a single point: This should generate all the data for the given latitude and longitude and period in ~/finse.
~/ctsm_escomp/tools/site_and_regional/subset_data \ point \ --site finse \ --lat 60.59383774 \ --lon 7.527008533 \ --create-user-mods \ --create-domain \ --create-surface \ --create-landuse \ --create-user-mods \ --create-datm \ --datm-syr 1970 \ --datm-eyr 1972 \ --dompft 7 \ --outdir ~/finse \ --overwrite
Run a case for the given point:
~/ctsm_escomp/cime/scripts/create_newcase \ --case ~/cases/finse \ --compset I2000Clm51BgcCrop \ --res CLM_USRDAT \ --machine fram \ --run-unsupported \ --user-mods-dirs ~/finse/user_mods \ # This contains the generated namelists and a shell command that CTSM uses to configure your case with the given input data. --handle-preexisting-dirs r \ --project <project_name> \ # e.g. nn2806k cd ~/cases/finse ./xmlchange MPILIB=impi ./xmlchange STOP_OPTION=nyears ./xmlchange STOP_N=1 ./xmlchange RUN_STARTDATE='1970-01-01' ./xmlchange DATM_YR_ALIGN=1970 ./xmlchange DATM_YR_START=1970 ./xmlchange DATM_YR_END=1971 ./case.setup ./case.build ./case.submit