Model configuration ... consistency with CVs.

[martinjuckes]

The CMIP6 experiment_id CV has required_model_components and additional_allowed_model_components which provide information about model configuration(s) needed for each experiment. This should, I think, be consistent with the ES-DOC model_configuration attribute. Since the information is presented differently, it is not obvious what the mapping should be, but the most widely used configuration ia AOGCM configuration which matches well with required:AOGCM + additional:AER BGC CHEM, with 100 experiments matching correctly. There are 9 that don't match (listed below). Beyond that it gets confusing. AGCM configuration and Atmosphere-Land configuration match required:AOGCM + additional:AER BGC CHEM and required:AOGCM + additional:AER CHEM --- but there is some cross-over between the two ES-DOC categories and the two pairs of categories, so it is not clear to me whether AGCM configuration should match required:AOGCM + additional:AER BGC CHEM or required:AOGCM + additional:AER CHEM (or perhaps it is a different categorisation and is not intended to match either).

There is some additional discussion here: https://github.com/WCRP-CMIP/CMIP6_CVs/issues/604

Here are some details:

-- AOGCM Configuration & AOGCM required, AER, BGC, CHEM allowed -- Experiments matching: 100 AOGCM ... esdoc only: hist-1950 control-1950 highres-future:: required: AOGCM; allowed: AER hist-stratO3 ssp245-stratO3:: required: AOGCM; allowed: AER BGC hist-spAer-aer hist-spAer-all:: required: AOGCM; allowed: AOGCM ... CV only: hist-resAMO hist-resIPO:: CGCM Configuration

-- AGCM Configuration & AGCM required, AER, CHEM allowed -- Experiments matching: 31 AGCM ... esdoc only: highresSST-p4K highresSST-present highresSST-smoothed highresSST-4xCO2 highresSST-LAI highresSST-future:: required: AGCM; allowed: AER spinup-1950:: required: AOGCM; allowed: AER amip-climSST amip-lfmip-pObs amip-lfmip-pdLC pdSST-piAntSIC pdSST-futAntSIC modelSST-futArcSIC pdSST-futArcSIC pdSST-futArcSICSIT pdSST-pdSIC pdSST-futBKSeasSIC pdSST-piArcSIC pdSST-futOkhotskSIC piSST-piSIC piSST-pdSIC amip-lfmip-rmLC futSST-pdSIC pdSST-pdSICSIT modelSST-pdSIC amip-climSIC amip:: required: AGCM; allowed: AER BGC CHEM AGCM ... CV only: piClim-histall piClim-4xCO2 piClim-ghg piClim-anthro piClim-histnat piClim-lu piClim-histghg piClim-histaer:: Atmosphere-Land Configuration

[charliepascoe]

I have reviewed the ESDOC CMIP6 experiments with a view to achieving consistency between the ES-DOC and PCMDI model configuration descriptions or at least clear mappings. These are correct as of version 97 of the ES-DOC experiment spreadsheet. As a general rule the ES-DOC "model configuration" maps to the PCMDI "required model components". I am recording the mappings in this google document https://docs.google.com/spreadsheets/d/1ECfvY4RNAzmtrW1TKgMjHRq2-DFQ2htOSG5yXdCrfEs/edit#gid=0

[martinjuckes]

Hi Charlotte, that is a nice piece of work .. it is good to see a high level of consistency. Have you raised an issue with PCMDI about the "OMIP2 spinup" and "atmosphere land-surface" configurations?

[charliepascoe]

Not yet but I intend to later today.