[v2.12.0] recipe_check_obs.yml broken due to missing data

[sloosvel]

During the recipe testing for version 2.12.0 (#3916), recipe_check_obs failed due to the following missing data:

ERROR   [1376260] Could not create all tasks
ERROR   [1376260] Missing data for preprocessor CMAP/pr:
- Missing data for Dataset: pr, Amon, OBS6, CMAP, v1
ERROR   [1376260] Missing data for preprocessor JRA-55/cli:
- Missing data for Dataset: cli, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/clivi:
- Missing data for Dataset: clivi, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/clw:
- Missing data for Dataset: clw, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/clwvi:
- Missing data for Dataset: clwvi, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/clt:
- Missing data for Dataset: clt, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/prw:
- Missing data for Dataset: prw, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rlus:
- Missing data for Dataset: rlus, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rlut:
- Missing data for Dataset: rlut, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rlutcs:
- Missing data for Dataset: rlutcs, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rsus:
- Missing data for Dataset: rsus, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rsuscs:
- Missing data for Dataset: rsuscs, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rsut:
- Missing data for Dataset: rsut, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/rsutcs:
- Missing data for Dataset: rsutcs, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/ta:
- Missing data for Dataset: ta, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/tas:
- Missing data for Dataset: tas, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor JRA-55/wap:
- Missing data for Dataset: wap, Amon, OBS6, JRA-55, 1
ERROR   [1376260] Missing data for preprocessor NCEP-DOE-R2/pr:
- Missing data for Dataset: pr, Amon, OBS6, NCEP-DOE-R2, 2
ERROR   [1376260] Missing data for preprocessor NCEP-DOE-R2/tauu:
- Missing data for Dataset: tauu, Amon, OBS6, NCEP-DOE-R2, 2
ERROR   [1376260] Missing data for preprocessor NCEP-DOE-R2/tauv:
- Missing data for Dataset: tauv, Amon, OBS6, NCEP-DOE-R2, 2
ERROR   [1376260] Missing data for preprocessor NCEP-DOE-R2/tos:
- Missing data for Dataset: tos, Amon, OBS6, NCEP-DOE-R2, 2
ERROR   [1376260] Missing data for preprocessor TROPFLUX/ts:
- Missing data for Dataset: ts, Amon, OBS6, TROPFLUX, v1
ERROR   [1376260] Missing data for preprocessor TROPFLUX/tos:
- Missing data for Dataset: tos, Omon, OBS6, TROPFLUX, v1
ERROR   [1376260] Missing data for preprocessor TROPFLUX/hfls:
- Missing data for Dataset: hfls, Amon, OBS6, TROPFLUX, v1
ERROR   [1376260] Missing data for preprocessor TROPFLUX/rlut:
- Missing data for Dataset: rlut, Amon, OBS6, TROPFLUX, v1
ERROR   [1376260] Missing data for preprocessor TROPFLUX/tauu:
- Missing data for Dataset: tauu, Amon, OBS6, TROPFLUX, v1
ERROR   [1376260] Missing data for preprocessor TROPFLUX/tauv:
- Missing data for Dataset: tauv, Amon, OBS6, TROPFLUX, v1
ERROR   [1376260] Missing data for preprocessor TROPFLUX/rsds:
- Missing data for Dataset: rsds, Amon, OBS6, TROPFLUX, v1

• See the full logs at: https://esmvaltool.dkrz.de/shared/esmvaltool/v2.12.0/recipe_check_obs_20250224_095226/run/

[bouweandela]

@ESMValGroup/obs-maintainers Is the data above available on Levante?

[rbeucher]

TROPFLUX is definitely not on Levante. I am questioning the relevance of that recipe considering that we are operating multiple centres with potentially different needs in terms of observation datasets.

[bouweandela]

We could split the recipe up into multiple recipes and put them all in a single folder, that would also prevent the entire recipe from being marked as broken if one or two datasets are not correct.

To facilitate the (regression) testing the CMORizers (#3946), I think there is definitely value in having all the CMORized data available at one or more sites.