Thanks @Michael_Sumner , I need to try more with masking approaches.
Just another follow-up here that I’m still very much interested in working out some more robust python solutions here. while the rasterstats with brute-force parallelization was working very nicely with the ~ 400,000 small continental us polygons there, I’ve found it doesn’t handle larger/more polar polygons (looking at you, protected areas in Alaska) very well at all, and I seem to be crashing out of RAM with even a single polygon.
I’ll keep poking at this for a robust workflow. If anyone wants to explore, just take a look at this copy of the polygons from Source Cooperative, which is just a geopaquet version of the non-continuous US polygons from the USGS gdb file Protected Areas Database of the United States (PAD-US) 3.0 (ver. 2.0, March 2023) - ScienceBase-Catalog )
Happy Earth Day! Let’s play with some protected area data!