Xarray + rex

[grantbuster]

Why this feature is necessary:
Lots of people use xarray as a more common file handler to spatiotemporal meteorological data. Xarray can work with NREL h5 files but you get a bunch of "phony dims" that aren't very useful. Would be nice to have a way to open NREL data with xarray with meta/time_index used appropriately and data unscaled on read. Also would be nice to allow for remote access to h5 files on s3 with xarray.

A possible solution is:
I think we could add a new backend in rex like this: https://docs.xarray.dev/en/latest/internals/how-to-add-new-backend.html

I have considered the following alternatives:
Currently you can open s3 files remotely like this:

# for backend need to `conda install conda install h5netcdf`
fp = "s3://nrel-pds-nsrdb/current/nsrdb_1998.h5"
s3 = s3fs.S3FileSystem()
s3f = s3.open(fp, "rb")
dset = xr.open_dataset(s3f, phony_dims='sort')

which gives you something ugly like this:

Additional context
This might be required for further collaboration with external orgs like EPRI.

Charge code
Maybe reV maintenance budget?

Urgency / Timeframe
Currently not urgent.