Hmm, my understanding was that fabs needed an initial supply of water, but then could cycle that same water for a long time.
A quick Google[0] suggests that they've only started doing this heavily pretty recently, but I'm guessing new fabs in Arizona will implement state of the art water-recycling.
That said, the numbers in this article suggest that 98% recycling would drop usage to the equivalent of ~6000 homes, which still feels significant.
Low seismic activity is the benefit of Arizona that I've seen cited in the past, which I'd guess outweighs the water sourcing issues.
That article is somewhat wrong. 10 million gallons of water is not enough for 300k households. It's maybe enough for 300k people. 10 million divided by 300k is 33 gallons. The average use per person even in urban areas is around 40 gallons per day - so that's generally ignoring outdoor irrigation. It's more like enough water for 200k-250k people.
So 2% of that is maybe around enough water for 5k people. Arizona has 7 million people. And residential doesn't even use most of Arizona's water.
A quick Google[0] suggests that they've only started doing this heavily pretty recently, but I'm guessing new fabs in Arizona will implement state of the art water-recycling.
That said, the numbers in this article suggest that 98% recycling would drop usage to the equivalent of ~6000 homes, which still feels significant.
Low seismic activity is the benefit of Arizona that I've seen cited in the past, which I'd guess outweighs the water sourcing issues.
[0] https://spectrum.ieee.org/fabs-cut-back-water-use