> "There are companies out there that are endeavoring to use satellites and computers, that are artificial intelligence and machine learning to replace the boot

> "There are companies out there that are endeavoring to use satellites and computers, that are artificial intelligence and machine learning to replace the boots on the ground foresters that have to go out and do these very precise measurements of trees by using satellite imagery to size up the trees and tell you how many trees there are, how big they are, how much they've grown over time. And the thinking is that by doing that and removing a lot of the upfront costs, you can start doing smaller and smaller tracts of woods." It seems like having a national or a global database of trees would be a useful and interesting public data set to have. For instance, you could have the location, approximate age, approximate mass and height, and probable species for all the known trees in your region of interest. I wonder what the technical limitations on collecting this data are? I mean, the imagery and metadata Google Maps uses probably could be used to find most of the full-grown trees, as long as they aren't clustered too tight to each other to distinguish. Maybe you'd need radar scans or close-up imagery for dense forests. Most young trees would probably be missed unless you had some very precise way to detect them. (Drones flying through forests, maybe? Or miniaturized Google-streetview-type cameras affixed to wild deer?)