Add more desciption for forcing files generation

[jbp54]

Is it possible to add more documentation for the generation of forcing files ?

I am currently looking at the generation of the forcing era5 files and I already obtained zipfiles with netcdfs inside.

However during the execution of the extract_ERA5_meteocloud_wrapper.job I have some questions with the cdo commands :

https://github.com/HPSCTerrSys/eCLM_static-file-generator/blob/f1c177e2ffbbec81a06b03b610b107a9984cbdac/mkforcing/extract_ERA5_meteocloud.sh#L78

cdo sellonlatbox,${area[0]},${area[1]},${area[2]},${area[3]} /p/data1/slmet/met_data/ecmwf/era5/grib/${year}/${month}/${year}${month}${day}${hour}ml.grb cut_domain${year}${month}${day}${hour}.grb

• Could you please indicate me where can I find the /p/data1/slmet/met_data/ecmwf/era5/grib/${year}/${month}/${year}${month}${day}${hour}_ml.grb file in the first cdo command ? The cdsapi download only contains netcdf files but no grb.

Same question for prepare_ERA5_input.sh

https://github.com/HPSCTerrSys/eCLM_static-file-generator/blob/f1c177e2ffbbec81a06b03b610b107a9984cbdac/mkforcing/prepare_ERA5_input.sh#L11

wgtcaf=/p/scratch/cslts/poll1/sim/euro-cordex/tsmp2_wfe_eur-11u/dta/rmp_gridwgts/wgtdis_era5caf_to_eur11u-189976.nc
wgtmeteo=/p/scratch/cslts/poll1/sim/euro-cordex/tsmp2_wfe_eur-11u/dta/rmp_gridwgts/wgtdis_era5meteo_to_eur11u-189976.nc
griddesfile=/p/scratch/cslts/poll1/sim/euro-cordex/tsmp2_wfe_eur-11u/dta/rmp_gridwgts/griddes_eur-11u_189976.txt
clm3grid=""

What are these files and how can they be generated ?

[jbp54]

@jjokella

[jbp54]

use cdo gendis commands to solve prepare_era5_input.sh blocking point

[jjokella]

@jbp54 I had a closer look and added some more documentation in HPSCTerrSys/eCLM_static-file-generator#46

The grib files used as input in extract_ERA5_meteocloud.sh can only be accessed on JSC machines.

A local README provides the following link:

Further information on ERA5 can be found at:
https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5

However, I cannot open that particular link, it may be an outdated version of https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5

I will try to get information on how to download these GRIB files.

[jbp54]

@jjokella, thanks a lot for the update. Indeed, accessing the grib files is a priority for us.

[jjokella]

Clarification from internal meeting (actions will follow)

extract_ERA5_meteocloud.sh loads data (temperature, wind, humidity) from the JSC-local GRIB-files that can potentially also be downloaded via download_ERA5_input.py

This would need an adaption of the downloaded variables in the request:

eCLM_atm-forcing-generator/mkforcing/download_ERA5_input.py

Lines 71 to 74 in 5587e36

	"surface_pressure",
	"mean_surface_downward_long_wave_radiation_flux",
	"mean_surface_downward_short_wave_radiation_flux",
	"mean_total_precipitation_rate"

Also, prepare_ERA5_input.sh has to be adapted, such that all files can read from the downloaded data:

eCLM_atm-forcing-generator/mkforcing/prepare_ERA5_input.sh

Lines 76 to 105 in 5587e36

	if $lrmp; then
	unzip ${pathdata}/download_era5_${year}_${month}.zip -d ${tmpdir}
	cdo -P ${ompthd} remap,${griddesfile},${wgtcaf} ${tmpdir}/data_stream-oper_stepType-instant.nc ${tmpdir}/rmp_era5_${year}_${month}_ins.nc
	cdo -P ${ompthd} remap,${griddesfile},${wgtcaf} ${tmpdir}/data_stream-oper_stepType-avg.nc ${tmpdir}/rmp_era5_${year}_${month}_avg.nc
	cdo -P ${ompthd} remap,${griddesfile},${wgtmeteo} ${pathdata}/meteocloud_${year}_${month}.nc ${tmpdir}/rmp_meteocloud_${year}_${month}.nc
	fi
	if $lmerge; then
	cdo -P ${ompthd} expr,'WIND=sqrt(u^2+v^2)' ${tmpdir}/rmp_meteocloud_${year}_${month}.nc ${tmpdir}/${year}_${month}_temp.nc
	cdo -f nc4c const,10,${tmpdir}/rmp_era5_${year}_${month}_avg.nc ${tmpdir}/${year}_${month}_const.nc
	ncpdq -U ${tmpdir}/rmp_era5_${year}_${month}_avg.nc ${tmpdir}/${year}_${month}_temp2.nc
	ncpdq -U ${tmpdir}/rmp_era5_${year}_${month}_ins.nc ${tmpdir}/${year}_${month}_temp7.nc
	cdo selvar,t,q ${tmpdir}/rmp_meteocloud_${year}_${month}.nc ${tmpdir}/${year}_${month}_temp3.nc
	cdo merge ${tmpdir}/${year}_${month}_const.nc ${tmpdir}/${year}_${month}_temp3.nc ${tmpdir}/${year}_${month}_temp2.nc \
	${tmpdir}/${year}_${month}_temp.nc ${tmpdir}/${year}_${month}_temp7.nc ${tmpdir}/${year}_${month}_temp4.nc
	ncks -C -x -v hyai,hyam,hybi,hybm ${tmpdir}/${year}_${month}_temp4.nc ${tmpdir}/${year}_${month}_temp5.nc
	ncwa -O -a lev ${tmpdir}/${year}_${month}_temp5.nc ${year}-${month}.nc
	ncrename -v sp,PSRF -v avg_sdswrf,FSDS -v avg_sdlwrf,FLDS -v avg_tprate,PRECTmms -v const,ZBOT -v t,TBOT -v q,QBOT ${year}-${month}.nc
	# ncap2 -O -s 'where(FSDS<0.) FSDS=0' ${year}_${month}.nc
	ncatted -O -a units,ZBOT,m,c,"m" ${year}-${month}.nc
	ncks -O -h --glb author="${author}" ${year}-${month}.nc
	ncks -O -h --glb contact="${email}" ${year}-${month}.nc
	rm ${tmpdir}/${year}_${month}_temp*nc ${tmpdir}/${year}_${month}_const.nc
	fi