Temperature inversions in GFDL-ESM4

[juliettelavoie]

I have found some temperature inversion (tasmin>tasmax) in the raw GFDL-ESM4 files.
I have seen it in at least r1i1p1f1 historical and ssp245, I have not yet look in the other files.
I can't find anything in the Errata about this.
The temperature inversion is carried over in the bias adjusted dataset.

figure of tasmax-tasmin.

[tlogan2000]

@juliettelavoie Do we want to simply mask these occurences as nan in the converted zarr files? Option 2 would be to flip tasmin for tasmax I guess.

[juliettelavoie]

I am leaning towards masking with nan.

[Zeitsperre]

Out of scope for Miranda. Quality checks for multiple variables is not compatible with miranda's design at the moment. Maybe in the future?