search

ESMValGroup / ESMValCore

ESMValCore: A community tool for pre-processing data from Earth system models in CMIP and running analysis scripts.
https://www.esmvaltool.org
Apache License 2.0
42 stars 38 forks source link

extract_levels does not preserve bounds when using interpolation #414

Open bouweandela opened 4 years ago

bouweandela commented 4 years ago

Preprocessor function extract_levels does not preserve bounds when using interpolation. We could add the bounds in some cases, e.g. when the source of the pressure levels is another dataset we can simply copy the bounds from there. Also for the CMIP6 CMOR coordinates bounds are available.

mattiarighi commented 4 years ago

Doesn't this belong to ESMValCore?

valeriupredoi commented 4 years ago

so I wonder if the new routine (which doesn't seem to be a repackaging of the old stratify) will fix this: https://scitools.org.uk/iris/docs/latest/iris/iris/experimental/stratify.html

zklaus commented 2 years ago

Also for the CMIP6 CMOR coordinates bounds are available.

This is not always true. The pressure level coordinates have no bounds associated.

Generally speaking, there are three ways vertical coordinates for atmospheric data are provided. Those are model levels (has bounds), pressure levels (doesn't have bounds), and altitudes (has bounds). Model levels are very high resolution and it is unusual to use the data on model levels for general analysis. From the other two types, I think pressure levels are far more common than altitudes*. So in summary, I think most atmospheric CMIP6 data that we see, will come with no clear bounds.

*I was trying to get some numbers for their relative prevalence, but in a pinch couldn't come up with something. If we want to have that information, I think it is possible to figure something out.