Consider whether to create sidecar application for final outputs and graphs

[bryanmgee]

It would be easiest to use the existing script for all data retrieval to avoid pseudo-duplication, but that script returns a lot more outputs than are necessary. Consider pros and cons of sidecar script versus embedding final clean-up, outputs, and graphs in that script.

[bryanmgee]

Benefit of sidecar is that any file could be easily plugged in as the input df; could include historic manually-compiled ones for example.