There is a geospatial package for duckdb though which is pretty slick. It is actually how I first learned of duckdb. We were dealing with nationwide parcel datasets and need to apply transformations nationwide and save out to more parquet files. It was easier and cheaper to replace all of the pandas workloads with duckdb.
> In a similar vein I made a site to check local water quality and swimming conditions in the Bay Area.
Not to rain on anyone's parade here, but if we have to check the results on the official agency postings anyway instead of relying on what you put on the website, what's the point of the website in the first place? Kind of feels like either own up to that you get the data import right, or don't bother at all. Maybe I miss the point though?