Comment by scythe

2 days ago

Potter's analysis of the various consumptive uses of water relies on the USGS survey data of the uses of water, generally a good source. However, there is a small flaw when we try to turn consumptive use into consumption, which is alluded to but not quantified in the USGS report: water losses to evaporation during storage (in reservoirs) and transportation. This is discussed in e.g.:

https://journals.ametsoc.org/view/journals/bams/99/1/bams-d-...

https://www.sciencedirect.com/science/article/abs/pii/S00344... (translation: 33.73e9 m^3/yr ≈ 24400 Mgal/day, roughly corn + alfalfa + steel)

Much of the literature is preliminary and recommends further study, but the initial estimates indicate that the amount of water that is simply lost from reservoirs is surprisingly large. So I like to yak about covering reservoirs (possibly with solar panels), which won't solve everything, but it has a far larger impact than data centers.