Quantify completeness of historical FERC 714 planning area geometries

[zaneselvans]

Develop routines for checking how complete a set of generated historical geometries are, with some visual outputs. Should allow filtering based on range of years, and the option to only look at respondents that reported demand in each year.

• [ ] In each year, what fraction of respondents have a geometry associated with them?
• [ ] In each year, what fraction of overall reported demand has a geometry associated with it?
• [ ] Break each of the above down depending on whether respondent was is a BA or a Utility.
• [ ] What's the total area of all geometries in each year (including overlapping areas)?
• [ ] What's the total area covered by all geometries in each year (excluding overlapping areas)?
• [ ] What fraction of the continental US is covered by the geometries in each year?

[zaneselvans]

It may be necessary to make the classification of respondent_type into balancing_authority vs. utility something that varies by year -- unclear whether respondents that show up as BAs in one year necessarily show up as BAs in other years.

[zaneselvans]

It may also be preferable to use the annual list of utility_id values that appear in the service_territory_eia861 table as an indicator of whether a given ID is associated with a "utility" rather than using the utilities_eia860 table.