Download instructions for the MSWEP data

[remi-kazeroni]

Describe the bug
The MSWEP data needed for the recipe hydrology/recipe_hydro_forcing.yml are a bit difficult to download. The current instructions can be found in the ESMValCore documentation but the link there (http://www.gloh2o.org/mswep/) only gives access to the latest version of the data (V280). For that the user needs to apply to get access to a Google Drive folder where the data are stored. On that page, it is said:

MSWEP V1 lacks cumulative distribution function (CDF) corrections, and therefore exhibits significant drizzle and less reliable trends due to changes in the data sources through time. Conversely, MSWEP V2 matches the CDF of each dataset combination to a single reference CDF, resulting in less drizzle and a more homogeneous record. See the technical documentation for the full list of changes.

Currently the hydrology/recipe_hydro_forcing.yml uses MSWEP V220 data but these can't be found on the official website. I could only download them using a link in the description of the PR #969 which introduced the on-the-fly cmorization for MSWEP data. I'm not sure that link follows the requirement for Tier3 data and that V220 is still distributed.

It would be good to clarify the following points:

• Which version of the data shall I place in the RAWOBS directory of Mistral (and maybe Jasmin)? We may need to update the documentation accordlingly.
• Depending on version of the MSWEP data recommended, the recipe recipe_hydro_forcing.yml may need to be updated.

[Peter9192]

@jeromaerts perhaps you can comment on the version and (public) availability of MSWEP?

[remi-kazeroni]

I ran into the same issue while we ran all the recipes for the release of v2.5 (#2535). So I downloaded the old version (v220) of the MSWEP data to be able to run the recipe hydrology/recipe_hydro_forcing.yml.

Getting the data MSWEP_V220/global_daily_050deg/2015*.nc and MSWEP_V220/global_daily_050deg/2016*.nc from the unofficial source given in ESMValCore/pull/969 was enough to run the recipe successfully, see output.

But the inconsistency between the version of the MSWEP data used in the recipe (v220) and the officially available version (v280) remains. I think we should either update the recipe_hydro_forcing.yml or our instructions on how to get the data from.

[Peter9192]

I can see this turning into an endless chase of the most recent version, so I'd prefer to update the instructions.

[remi-kazeroni]

Thanks, I agree. I will transfer this to ESMValCore since info on natively supported non-CMIP datasets lives there