Open rbeucher opened 1 year ago
The rationale behind having a daily cmoriser for ERA5-Land hourly data is that quite a few recipes use daily averages from ERA5. Some HPC systems have chosen to store variables from ERA5-Land instead because they are better resolved. pr is one of them. It would be good to be able to run the recipes with either of them.
The CMOR checks for valid values for the longitude. I don't understand why it is not picking the same checks as ERA5 hourly data.
Hi @rbeucher I never encountered this for monthly data (unfortunately I'm on holidays next two weeks and don't really have time to look at it before). It should be easy to include further variables for era5_land if the fix exists in ESMValCore/esmvalcore/cmor/_fixes/native6/era5.py you can just include them as pr in ESMValCore/esmvalcore/cmor/_fixes/native6/era5_land.py. But they need to be tested if further adjustments are necessary, therefore they are not in main.
@rbeucher Which tests are used for ERA5 is determined in ESMValCore/tests/integration/cmor/_fixes/native6/test_era5.py Maybe an analogue ESMValCore/tests/integration/cmor/_fixes/native6/test_era5_land.py is needed to switch this test off? Unfortunately I have not much experience with the tests.
that's probably on a -180 -> 180 lon grid
Hi All,
I have been trying to use the
recipe_daily_era5.yml
with ERA5-Land hourly data. For now, only thepr
variable is supported via the native6 project era5_land.pySo I tried to run the following recipe using hourly data for
pr
Here is what I get:
Longitude has negative values of course (-180 to 180) but that is the same for ERA5.
Is this related to the grid size that is different in ERA5-Land (9km)?
@katjaweigel I am pinging you here as I have found you worked on the monthly CMORiser #879
Thanks for your help