There was a bug in Iris, SciTools/iris#3678, which caused problems with parametric vertical coordinates. This issue has been solved in Iris, but that solution has not yet been adopted in ESMValCore. This leads to problems with datasets for which no fix has been implemented that enforces the 'wrong' interpretation of the CF conventions that Iris previously expected.
I made a start with fixing this in #2999.
There was a bug in Iris, SciTools/iris#3678, which caused problems with parametric vertical coordinates. This issue has been solved in Iris, but that solution has not yet been adopted in ESMValCore. This leads to problems with datasets for which no fix has been implemented that enforces the 'wrong' interpretation of the CF conventions that Iris previously expected.
I made a start with fixing this in #2999.