Add albedo function for inland water bodies

[IoannisSifnaios]

• Part of Google Summer of Code: Floating PV #2068
• I am familiar with the contributing guidelines
• Tests added
• Updates entries in docs/sphinx/source/reference for API changes.
• Adds description and name entries in the appropriate "what's new" file in docs/sphinx/source/whatsnew for all changes. Includes link to the GitHub Issue with :issue:`num` or this Pull Request with :pull:`num`. Includes contributor name and/or GitHub username (link with :ghuser:`user`).
• New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
• Pull request is nearly complete and ready for detailed review.
• Maintainer: Appropriate GitHub Labels (including remote-data) and Milestone are assigned to the Pull Request and linked Issue.

This PR adds a function for calculating the albedo of inland water bodies as described in the paper: Dvoracek M.J., Hannabas B. (1990). "Prediction of albedo for use in evapotranspiration and irrigation scheduling." IN: Visions of the Future, American Society of Agricultural Engineers 04-90: 692-699.

You can contact me if you would like me to send you the paper.

[echedey-ls]

An initial review. I like the docs redaction.

[kandersolar]

A few comments. Nice and clean code!

[cwhanse]

I have no comments beyond those made by other reviewers. Approving pending resolution of other comments.

[kandersolar]

A few last comments :)

[kandersolar]

It's possible to get an array out despite putting a Series in:

In [399]: inland_water_dvoracek(pd.Series([45]), "muddy_water_no_waves")
Out[399]: array([0.13516185])

This is not an uncommon problem when using numpy tricks like np.where. Our usual workaround is to fix up the types at the end of the function like this:

    if isinstance(solar_elevation, pd.Series):
        albedo = pd.Series(albedo, index=solar_elevation.index)

    return albedo

A little inelegant, but it gets the job done.

[AdamRJensen]

Wouldn't the type check of int/floats I suggested earlier do the job?

[kandersolar]

Type checking has some tricky edge cases (e.g. isinstance(np.array([45.])[0], (int, float)) is False because it is an np.float64, not a float). A test based on np.isscalar might be ok.

But I think it is generally better to fix up types at the end instead of following different calculation paths based on type. Less chance of accidentally calculating different output values for different types.

[IoannisSifnaios]

@kandersolar the problem is that np.where returns an array (even if you pass a pd.Series for the solar_elevation) - so the if statement is never true. I did a fix, but it might not be the prettiest one... can you approve it (or suggest an alternative)?

[echedey-ls]

Minor comments. Merging feels close 🚀

[echedey-ls]

LGTM with the changes down below 🚀

[kandersolar]

LGTM. Thanks @IoannisSifnaios and everyone that reviewed!