Interestingly, the source report calculates this, though the article doesn't mention the distinction: https://pubs.usgs.gov/circ/1441/circ1441.pdf#page=50 https

Interestingly, the source report calculates this, though the article doesn't mention the distinction: https://pubs.usgs.gov/circ/1441/circ1441.pdf#page=50 https://pubs.usgs.gov/circ/1441/circ1441.pdf#page=50 95100 Mgal/day fresh water withdrawn, 3760 Mgal/day consumed. Even worse, the "pipe and tank" diagram from OP shows total water consumption, which includes seawater! (37800 million gallons per day "withdrawn", though we are not exactly worried about running out of seawater right now) Additionally, the article doesn't mention hydroelectric at all, and the report says it's out of scope. (Almost all the water coming out of a dam is either consumed or allowed to discharge into the sea)