Remove monotonicity coordinate check for unstructured grids

[jvegreg]

Description

We currently can not work with unstructured grids because they fail the monotonicity check. This PR fixes this by skipping the check when we detect an unstructured grid, as we do for the irregular ones when ndim of lat and lon is greather than one.

I opted for a fairly easy way to detect those grids: look for ndim == 1 and latand lon sharing the same dimension

• Closes Working with unstructured ocean AWI-CM data #751

---

Before you get started

• ☝ Create an issue to discuss what you are going to do

Checklist

• PR has a descriptive title for the changelog
• Code follows the style guide
• Documentation is available for new functionality
• YAML files pass pre-commit or yamllint checks
• Circle/CI tests pass
• Codacy code quality checks pass
• Documentation builds successfully on readthedocs
• Unit tests are available

If you make backwards incompatible changes to the recipe format:

• Update ESMValTool and link the pull request(s) in the description

---

To help with the number pull requests:

• 🙏 We kindly ask you to review two other open pull requests in this repository

[koldunovn]

Thanks a lot for doing this, @jvegasbsc ! Looks good to me. The assumption of the same sized 1d coordinate is not necessary mean unstructured mesh, one can have a setup on regular grid with same number of lons and lats (some crazy regional model), but better to be practical :)

[jvegreg]

Thanks a lot for doing this, @jvegasbsc ! Looks good to me. The assumption of the same sized 1d coordinate is not necessary mean unstructured mesh, one can have a setup on regular grid with same number of lons and lats (some crazy regional model), but better to be practical :)

Is not checking is the same size, it is checking is the same dimension in the cube (see https://scitools-iris.readthedocs.io/en/latest/generated/api/iris/cube.html?highlight=coord_dims#iris.cube.Cube.coord_dims) so it should be safer

[Peter9192]

Checking in to offload Bouwe. Code make sense to me. Just one suggestion, a question, and a typo.
If you can point me to a nice example cube with unstructured data I could do a more thorough check, but maybe that's not necessary if one of the other reviewers did it.
Cheers! @jvegasbsc

[bouweandela]

If you can point me to a nice example cube with unstructured data I could do a more thorough check

You can find the name of the model and variable in the linked issue, it would be great if you could try running a recipe to make sure it really works.

[jvegreg]

This set of HighRESMIP experiments, for example. Be aware that you may need some extra tags, as I run this in our local copies of the data

     - {dataset: CESM1-CAM5-SE-HR, project: CMIP6, exp: hist-1950, start_year: 1960, end_year: 2014, ensemble: r1i1p1f1, grid: gn, mip: Amon}
     - {dataset: CESM1-CAM5-SE-LR, project: CMIP6, exp: hist-1950, start_year: 1960, end_year: 2014, ensemble: r1i1p1f1, grid: gn, mip: Amon}
     - {dataset: CESM1-CAM5-SE-HR, project: CMIP6, exp: highres-future, start_year: 2015, end_year: 2050, ensemble: r1i1p1f1, grid: gn, mip: Amon}
     - {dataset: CESM1-CAM5-SE-LR, project: CMIP6, exp: highres-future, start_year: 2015, end_year: 2050, ensemble: r1i1p1f1, grid: gn, mip: Amon }

[Peter9192]

This set of HighRESMIP experiments, for example.

Thanks, that seems to work like a charm 👍

Now we just need to make sure that the checks pass. The docs build is fixed here: #964.

[jvegreg]

Checks work, codacy down.

[nielsdrost]

Seems like Codacy is spitting out another warning. Can you see if this is fixable @jvegasbsc ?

Thanks, that seems to work like a charm 👍 Now we just need to make sure that the checks pass.

[jvegreg]

Sorry for dismissing your review @Peter9192 , but the check pass and the codacy issue is solved now, and we want this in 2.2

[Peter9192]

Sorry for dismissing your review @Peter9192 , but the check pass and the codacy issue is solved now, and we want this in 2.2

No problem! Sorry I didn't get back to it soon enough.