BGCval not adapted to NEMO4.2
Running BGCval to compare NEMO4.0 and NEMO4.2 runs, it appears that the AMOC is weaker in NEMO4.2 runs.
Dan Hodson calculated the same AMOC indeendantly "using the NEMO specific CDFTOOLS package. This required a bit of tweaking as dq101 doesn't seem to have the ocean halo grid points that do495 does. I think I have done this correctly, the results seem sensible."
And indeed, both NEMO4.2 and NEMO4.0 AMOCS are very similar calculated that way.
This suggest that something is not done correctly in BGCval, with NEMO4.2 outputs.
Main difference is that NEMO4.2 does not include the halos. grid size is different, and mesh-mask is different as well (all Area_weighted operation don't work as the UKESM-mesh mask is not adapted anymore.
maybe adding a UKESM2 model with associated mesh-mask would help ?
BGCval not adapted to NEMO4.2
Running BGCval to compare NEMO4.0 and NEMO4.2 runs, it appears that the AMOC is weaker in NEMO4.2 runs.
Dan Hodson calculated the same AMOC indeendantly "using the NEMO specific CDFTOOLS package. This required a bit of tweaking as dq101 doesn't seem to have the ocean halo grid points that do495 does. I think I have done this correctly, the results seem sensible."
And indeed, both NEMO4.2 and NEMO4.0 AMOCS are very similar calculated that way.
This suggest that something is not done correctly in BGCval, with NEMO4.2 outputs.
Main difference is that NEMO4.2 does not include the halos. grid size is different, and mesh-mask is different as well (all Area_weighted operation don't work as the UKESM-mesh mask is not adapted anymore.
maybe adding a UKESM2 model with associated mesh-mask would help ?