Allow calculations to wrap around the globe

[ColemanTom]

What should we add?

I was looking at the code for vorticity and divergence (and I'm sure this applies to other calculations too), but if you run these on a global model, you should be able to have the calculation wrap around the globe as it a contiguous domain. So the derivative around longitude 0 may be from 355 to 5 if a 5 degree spacing was used. This would improve the results at the edges, and mean for global datasets that you don't have to lose edge points for longitude.

Reference

No response

[dopplershift]

Implementation challenges, and their performance implications, aside (this would mean new derivative operations that are only used for periodic cases), any idea how to determine this robustly using netCDF-CF metadata? How would you tell the difference between a truly global run and a subset with lat/lon coordinates?

I'm also a little skeptical since even getting proper coordinate reference system information (i.e. grid_mapping, yes I'm looking at you HRRR) has been problematic.

As a work-around for now you could concatenate the first "column" to the end of the array, and the last "column" to the beginning; then subset the result. It would get you an identical result if the accuracy were important to you.

[dopplershift]

This on-going work in xarray might at least possibly pave the way for simplicity of implementation, and just leave good ol' metadata as the challenge.

[kgoebber]

What does the behavior look like if you add a cyclic point using a tool like Cartopy? Don’t know if that would work directly on an xarray DataArray, but it should (?) be able to be done.