It seems like we could get some good estimates based on non-census data. For example cellular data or ad data, rf emissions, CO2 or other chemical emissions, s

It seems like we could get some good estimates based on non-census data. For example cellular data or ad data, rf emissions, CO2 or other chemical emissions, satellite photography of human activity (mowing, cropping, logging, building, etc), food or fuel shipments, etc. A census may miss people, but the rest of the data wouldn't.