Return "base" FSS as part of the score (`FBS_ref`)

[nikeethr]

Is your feature request related to a problem? Please describe.

Interpretation of FSS without the base reference does not give the full picture. We should have an option to return the decomposed score. With FBS_ref being the denominator.

https://journals.ametsoc.org/view/journals/mwre/149/10/MWR-D-18-0106.1.xml

Describe the solution you'd like

The current FSS implementation already computes the final score from decomposed parts of the score. We just need to return the components that make up the score as well, so that they can be used for analysis if required.

[tennlee]

@nikeethr @nicholasloveday I'm not sure whether this needs to be prioritised for version 1. Can you either tag it or leave a comment here?

[nikeethr]

I'll have to handball it to @nicholasloveday. I think its more of a science question.

I feel like the decision would be based on answering:

• is the decomposed score absolutely necessary for usability, based on the original 2008 paper (which is what the current code implements)? Or
• are the author's suggestions in the more recent journal linked above convincing enough; and the FSS would not make sense without the decomposed outputs?

From a software point of view its not super tricky to implement. But we have to account for the testing and validation as well around it, as well as updating the notebooks with examples on how to use the decomposed scores - which I'm not entirely familiar with.

[nicholasloveday]

I think 1.1 is appropriate. The score will work as is, but it is best practice to also display the base rate.