Closed charliepascoe closed 2 years ago
Here's a snippet from David Plummer on the topic:
I took a quick peek at the CMIP6 CV document (http://goo.gl/v1drZl) and under footnote #14 describing source_type they have the following bit:
All types that apply to the model (i.e., are active in a given experiment) should appear in a list with each type separated by a single space. Sometimes source_type for CMIP6 simulations will be described by a single label, as in "AOGCM" or "AGCM" or "OGCM", but when additional interactive components are included, then multiple labels should be found in source_type (e..g., "AOGCM BGC", "AGCM CHEM AER", "AOGCM AER", "AOGCM ISM", "AGCM SLAB", "OGCM BGC"). Single labels will be used for "offline" models such as "ISM" and "RAD"
This was the guidance I followed in cooking up 'AGCM CHEM'. Let me know if you are okay with accepting this. Although I haven't heard of this error for other data submissions, so I am guessing not many people added more than one component? Is it possible to accept single components and multiple components?
CMIP6_global_attributes_filenames_CVs CMIP6 Global Attributes, DRS, Filenames, Directory Structure, and CV's 10 September 2018 (v6.2.7) Document short URL: https://goo.gl/v1drZl Karl E. Taylor, Martin Juckes, V. Balaji, Luca Cinquini, Sébastien Denvil, Paul J. Durack, Mark Elkington, Eric Guilyardi, Slava Kharin, Michael Lautenschl... goo.gl
example data can be found in /datacentre/arrivals/users/dplummer/ccmi-2022a/v20210311/
@martinjuckes I forgot to mention this in our ceda-cc meeting on Monday. ceda-cc should allow data files to specify more than one source_type attribute
example data can be found in /datacentre/arrivals/users/dplummer/ccmi-2022a/v20210311/
Now /badc/ccmi/data/post-cmip6/ccmi-2022/ECCC/CMAM/refD1/r1i1p1f1/Amon/o3/gn/v20210311/o3_Amon_CMAM_refD1_r1i1p1f1_gn_196001-201812.nc
Fixed.
For CCMI-2022 the ceda-cc checker should allow multiple entries against the source_type attribute, each entry should be separated by a single space.