Extended version of the Sandbox to run the COSMO chain. Typically this code is used to run high-resolution simulations over the European and alpine regions, but it can be easily adapted to any other region.
TODO soil moisture
This code is prepared to run at Piz Daint from the login nodes. The only extra requirement is to have python3 installed and the list of libraries below ( preferable installed with miniconda).
- pandas
- subprocess
- numpy
- sys
- os
- glob
- calendar
- netCDF4
The code is structured in different steps that need to be executed to simulate the standard chain
- 0_get_data: Scripts to copy automatically ERA or MPI files
- 1_ifs2lm: Generate 12km boundary conditions from GCM or reanalysis fields.
- 2_lm_c: Run Cosmo at 12km.
- 3_lm2lm: Generate high resolution boundary conditions for the nested run.
- 4_lm_f: Run cosmo at high-resolution (standard 2.2km)
- x_chain: Check that the simulation step has correctly been performed and resubmit the next step.
1-Copy the cosmo and int2lm executables into the bin/ folder. The current standard executables can be found in /store/c2sm/pr04/jvergara/base_executables
2-Edit define_simulation.py to set the steps of the simulation.
3-Run it with python define_simulation.py
4-Run python control_simulation.py which will set the values needed for running the next step
5-Check that in run_daint.sh the "parts" variable contains the steps necessary for the simulation and that the last step is x_chain (if you want to automatically chain your simulation)
6-Execute ./run_daint
7-Enjoy!