Add a test case for extrapolating WOA 2023 data

[xylar]

This merge adds a new utility test group mostly for provenance and ease of updating data sets in the future.

For now, the only test case is ExtrapWoa, used to extrapolate WOA 2023 data into ice-shelf cavities and ice shelves, then into grounded ice, land and below bathymetry.

To convert from in situ to potential temperature, the gsw package has been added as a compass dependency.

Checklist

• User's Guide has been updated
• Developer's Guide has been updated
• API documentation in the Developer's Guide (api.rst) has any new or modified class, method and/or functions listed
• Documentation has been built locally and changes look as expected
• Document (in a comment titled Testing in this PR) any testing that was used to verify the changes

[xylar]

Testing

I ran the test case here:

/lcrc/group/e3sm/ac.xylar/compass_1.2/chrysalis/test_20230414/extrap_woa23_pt/ocean/utility/extrap_woa

The resulting dataset is here:

/lcrc/group/e3sm/ac.xylar/compass_1.2/chrysalis/test_20230414/extrap_woa23_pt/ocean/utility/extrap_woa/extrap/woa23_decav_0.25_extrap.nc

I am also putting it in the initial-condition database for now, so it can be found here:

/lcrc/group/e3sm/public_html/mpas_standalonedata/mpas-ocean/initial_condition_database/woa23_decav_0.25_extrap.20230414.nc

or here:
https://web.lcrc.anl.gov/public/e3sm/mpas_standalonedata/mpas-ocean/initial_condition_database/woa23_decav_0.25_extrap.20230414.nc

[xylar]

@milenaveneziani, @maltrud, @proteanplanet, @vanroekel do any of you have the time and interest in looking at the extrapolated WOA 2023 initial condition that I have created?

I think it's in good shape in terms of how it treats ice-shelf cavities. I think the extrapolation approach is reasonable for grounded ice, land and below bathymetry but I'm open to suggestion there. See:
https://github.com/xylar/compass/blob/add-woa23-extrap/docs/developers_guide/ocean/test_groups/utility.rst#extrap_step
for a description of what I'm doing.

The WOA 2023 data itself certainly has its idiosyncrasies (visible discontinuities and noise in both T and S) that you will no doubt see. In my experience, other datasets are not typically better. If you would like to work with the data from WOA, it is also in the database:

/lcrc/group/e3sm/public/e3sm/mpas_standalonedata/mpas-ocean/initial_condition_database/woa23_decav91C0_*

[xylar]

I need to rerun since I found I used the wrong sign for z (positive down depth rather than positive up altitude) when computing the pressure used to convert from in situ temperature to potential temperature.

[xylar]

@milenaveneziani, @maltrud, @proteanplanet and @vanroekel, there is a new version of the extrapolate initial condition here:

/lcrc/group/e3sm/public_html/mpas_standalonedata/mpas-ocean/initial_condition_database/woa23_decav_0.25_extrap.20230414.nc

It has be bug fix mentioned above.

[xylar]

I merged this because it seems like the basic approach is working. I'm find with revisiting either the input dataset or the algorithm later on but this will do as at least a first draft.