Is it worthwhile standardizing the experiment names with the ones specified in the CCMI2022_CV.json table? In the case of 'REF-D1' it would be 'refD1'. I am not sure why the experiment names evolved this way and I guess it is not terribly important unless writing netCDF metadata, but I wondering if this would help reduce confusion?
e.g.
https://catalogue.ceda.ac.uk/uuid/031313dc6bfc4a0baabd58a51629bc21