Update Coupled Model and Delete adf_quick_run.ipynb now that link_to_ADF.ipynb exists

[TeaganKing]

This PR removes the old adf_quick_run notebook since we now prefer supporting link_to_ADF.ipynb and the helper scripts that create the ADF configuration files

All PRs Checklist:

• Have you followed the guidelines in our Contributor's Guide?
• Have you checked to ensure there aren't other open Pull Requests for the same update/change?
• Have you made sure that the pre-commit checks passed (#8 in Adding Notebooks Guide)?
• Have you successfully tested your changes locally?

[mnlevy1981]

adf_quick_run.ipynb is still run from examples/coupled_model/config.yml:

    atm:
      adf_quick_run:
        parameter_groups:
          none:
            adf_path: ../../../externals/ADF
            config_path: .
            config_fil_str: "config_f.cam6_3_119.FLTHIST_ne30.r328_gamma0.33_soae.001.yaml"

We've talked about encouraging folks to use key_metrics instead of coupled_model, but until we get rid of the latter (which should be done, since it doesn't follow the same variable name conventions as key_metrics and that is confusing) I think we should keep the ADF quick run notebook.

Another concern is that maybe @justin-richling, @brianpm, or @nusbaume want to provide an example of invoking ADF from a notebook instead of the command line. If that's not the case, maybe we could remove the quick run notebook but run a different atmosphere notebook in coupled_model?

[mnlevy1981]

Should've marked "request changes" with #198 (comment)

[nusbaume]

Hi @mnlevy1981 @TeaganKing, after discussing it in an ADF meeting we decided that there isn't really a need to have an example of the ADF being called from a notebook (calling it directly in python or from the command line is fine with us), so feel free to move ahead with this particular PR. Thanks for checking!

[TeaganKing]

Thanks @nusbaume for this feedback. @mnlevy1981 I removed this from the coupled model config file, too. That said, I'm not in any rush to get this in, and it could certainly be done as a larger removal (or updating) of the coupled model example.

[TeaganKing]

The past few commits address these changes:

• The register_cmap error is coming from mom6-tools. There are changes in mom6-tools that already address this deprecation, so this PR also now pulls in a more recent mom6-tools external.
• The robust argument of plot() has also been deprecated and hence removed from the Global_Land_Compare_Two_Cases notebook.
• Notebook names have been updated to adhere to our suggested format.
• The 'coupled model' (now 'additional metrics') example configuration file has been updated to be a similar format as the key metrics configuration file.
• Parameter cells have been updated for the ocean and land notebooks such that the proper parameters are used from the configuration file.

Remaining issues:

• The land notebook fails on data that does not include 'lon' in line 4: ds[var].sel(lon=300, lat=-10, method="nearest").plot(hue="case", y="levsoi")

[mnlevy1981]

The land notebook fails on data that does not include 'lon' in line 4: ds[var].sel(lon=300, lat=-10, method="nearest").plot(hue="case", y="levsoi")

Is this a case that was run on the CAM-SE grid instead of the FV grid? I bet something like

if "lndgrid" in ds[var].dims:
    # find index of lndgrid close to 300 E, 10 S
    ds[var].isel(lndgrid=lndgrid_ind).plot(hue="case", y="levsoi")
else:
    ds[var].sel(lon=300, lat=-10, method="nearest").plot(hue="case", y="levsoi")

would work (where the tricky part is finding the lndgrid index closest to (300,-10): I would do it offline once, and then hard-code the index with a comment about where it came from and how it's only useful for the ne30 grid); alternatively, we could require this notebook to use regridded output if it was run on the SE grid.

[TeaganKing]

I found the closest land indices by finding a minimum non-NaN distance between the landgrid and the lat/lon values.