Infos on Namelist and Forcing:¶
Namelist¶
The namelist is a text file describing the main characteristics of the simulation. There are infos on the soil and vegetation (for example in NAM_DATA_ISBA and NAM_ISBA), there are some choices for physical options (NAM_ISBA_SNOWN), etc.
You should not try to create a namelist from scratch but use and modify an existing namelist.
There are some namelists in snowtools/snowtools/DATA or in snowtools/snowtools/tests/namelists.
It is interesting to sort your namelist in order to compare it to others. You can use namelist_sort.py which is in snowtools/cenutils.
The command is
python namelist_sort.py OPTIONS.nam -o OPTIONS_SORT.nam
The NAM_IO_OFFLINE options in namelist¶
XTSTEP_OUTPUT defines the timestep (in seconds) of the outputs. XTSTEP_OUTPUT=21600 means that you’ll have diagnostics every 6 hours in your PRO file.
It is possible to compress the netcdf PRO file: NOUTPUT_NETCDF = 4 (the output file is a NetCDF4-classic format, the default is NetCDF3 format) and NCOMPRESS_NETCDF = 1 (the level of compression is 1, this is the default).
By default, the FORCING file is read every timestep. In order to decrease the I/O time, you can choose to read all forcing datas in one time by setting NB_READ_FORC = 1.
Forcing files¶
SAFRAN reanalysis is regularly updated.
The directory where you should take SAFRAN meteorological forcing files depend on which slopes you need and on the file system you are using.
At CEN, SAFRAN files are stored on the following directories :
/rd/cenfic3/cenmod/era40/vortex/s2m/DOM_allslopes/reanalysis/meteocontains SAFRAN reanalysis for all massifs; 0°, 20° and 40° slope values, and 8 aspects/rd/cenfic3/cenmod/era40/vortex/s2m/DOM_flat/reanalysis/meteoSAFRAN reanalysis for all massifs over flat area/rd/cenfic3/cenmod/era40/vortex/s2m/postes/reanalysis/meteocontains SAFRAN reanalysis for all stations (include topographic corrections of radiations)
where DOM represents the spatial domain which can be “alp” (Alps), “pyr” (Pyrénées), “cor” (Corsica)
For use at Toulouse, get it from Hendrix:
From hendrix to any informatic resource in Toulouse, note that it is recommended to get SAFRAN files with vortex. See ßcripts/extract/vortex/get_reanalysis.py.
You should not develop your own transfer script.
On supercomputer, use the combination of -m s2m (or -m safran if you want to define your own slopes), -r region and -f reanalysis2020.2@lafaysse to get access to reanalysis with s2m command. Regions have to be chosen in [alp_allslopes, alp_flat, pyr_allslopes, pyr_flat, cor_allslopes, cor_flat, postes].
For information, SAFRAN are stored in the following directories:
/home/lafaysse/vortex/s2m/DOM_allslopes/reanalysis/meteocontains SAFRAN reanalysis for all massifs; 0°, 20° and 40° slope values, and 8 aspects/home/lafaysse/vortex/s2m/DOM_flat/reanalysis/meteocontains SAFRAN reanalysis for all massifs over flat area/home/lafaysse/vortex/s2m/postes/reanalysis/meteocontains SAFRAN reanalysis for all stations (include topographic corrections of radiations)
Creation of FORCING files:¶
If you have to create your own forcing files, there are several possibilities depending on your situation:
You can create one FORCING files from datas from an observation site in .csv or in .txt: look to
snowtools/scripts/create_forcing/Template_creation_FORCING.pyYou can interpolate the SAFRAN reanalysis for point simulations or for 2D simulations: look to Interpolation tools.
- (Not recommanded) you can create a FORCING file from scratch:
you must follow the same structure (same variable names and ranks, same dimensions) as an example file
all variables are mandatory (CO2air and Wind_DIR have no impact in the case of snow and can be fixed to a constant value)
if you do not have distinct data for diffuse and direct solar radiations, you can put everything in the direct radiation for applications not sensitive to this separation (with Crocus, only TARTES and MEB options are sensitive but they are not activated by default).
example for point simulations:
FORCING_test_base.ncfile provided in thesnowtools/DATAfolderexample for 2D simulations:
FORCING_test_2D.ncavailable in Meteo France system in/rd/cenfic3/cenmod/home/viallonl/testbase/FORCINGfolder
NB: If your datas are available only every 3h for example, you have to set the variable FORC_TIME_STP to 10800 instead of 3600 in the FORCING file.
Other files¶
Todo
Document common localizations and access mode for the different files needed by SURFEX simulations and processing around.