[Feature]: Add vertical regridding

[tomvothecoder]

Blocked by #71

Describe the solution you'd like

Vertical regridder that can remap hybrid vertical coordinates to standard pressure levels

Acceptance criteria:

• Remapping to z-levels
• Add more....

Additional context

cdms has reconstructPressureFromHybrid, logLinearInterpolation, and linearInterpolation

[chengzhuzhang]

I came across this script that we may be able to refer to..

[durack1]

@tomvothecoder @chengzhuzhang the xgcm code base (also xarray) could be a useful reference, see https://xgcm.readthedocs.io/en/latest/transform.html

[chengzhuzhang]

@durack1 Thank you for the reference.

[tomvothecoder]

Meeting Notes (7/20):

Alternative solutions:

1. Ryan Abernathey and Paul both mentioned vertical regridding is available in xgcm
2. Jill found this script from GeoCAT: https://github.com/NCAR/geocat-comp/blob/0263d2d77ab22a6be4d1d3f16d5bfcfc68057364/src/geocat/comp/interp_hybrid_to_pressure.py
3. CDAT's vertical regridding code: https://github.com/CDAT/cdutil/blob/master/cdutil/vertical.py

Action items:

• Check the similarities and differences between the existing solutions
• Check the existing solution(s) meets our requirements

[chengzhuzhang]

@tomvothecoder Thank you for summarizing and updating this issue. I added CDAT's vertical regridding code as a reference.

Just a note here that the GeoCAT solution actually calls the interpolate from metpy package to actually do the vertical remapping. By a brief examination, I think both solutions should meet our requirements in terms of functionality. And perhaps performance and use APIs will be two aspects we should focus on.

[chengzhuzhang]

@jasonb5 Hey, Jason. I spent some time to experiment with the existing vertical regridding methods. Please see the jupyter notebook here: https://github.com/xCDAT/xcdat_test/blob/main/validation/v1.0.0/vertical_interp/Vertical_interp.ipynb

Notice that only xgcm method (the third one in the notebook) is working. And I have hard time to trouble shooting the first two methods. When you have time, would you please also take a look for me. Let me know if you'd like any clarification on this. Thank you!

[jasonb5]

@chengzhuzhang Thanks, I'll give it a look over.

[chengzhuzhang]

Hey @jasonb5 , as discussed I added a 4th method, which is to use the cdms2/cdutil approach. The note book now is updated. https://github.com/xCDAT/xcdat_test/blob/main/validation/v1.0.0/vertical_interp/Vertical_interp.ipynb Thank you for looking over!