A UC Irvine study estimates groundwater overdraft at 40 million acre-feet in the past four years. If that is right, it is astonishing.

40 million acre-feet in four years = 10 million acre-feet a year groundwater overdraft.

(10 million acre-feet a year)/ (3 acre-feet/acre of land irrigated) = 3.3 million acres of ag land in the Central Valley was supported by groundwater overdraft for the past four years.

I hadn’t thought it was that much, although those were drought years. Eventually, groundwater will get prohibitively expensive to pump (and the quality gets worse as it concentrates and you reach older water, so there are potentially even treatment costs for very deep groundwater). Before this report, my rough feel was that about one million acres of irrigated lands in CA are supported by unsustainable groundwater withdrawals. If overdraft is 10 MAF/year in the Central Valley, it makes me think my prediction of a loss of 3 million acres of irrigated ag is conservative, rather than the wild-eyed provocation one finds on blogs these days.

You know, I never get a piece of information that challenges my rough guess that we’re going to go from 9 million irrigated acres in CA to about 6 million irrigated acres over the next few decades. The Delta flows report (which of course won’t get implemented fully) was in the same ballpark. Even when we see approaches from entirely different aspects of the water conflict, a back-of-the-envelope calculation of three acre-feet/irrigated acre gets me to the same conclusion. Of course, I could be subject to confirmation bias and pushing my pre-existing policy beliefs. Or I could be right.

A loss of three million acres of irrigated ag in California forces its own conclusions. A shift away from grains for animals to grains directly for human consumption, especially as Russian and Chinese grain crops fail. Meat and dairy will get pricier. (The end result of that is that the increased dairy production in the Valley of the last decade will be stranded capital. These bubbles are expensive.) I find it hard to believe that we would still grow the better part of a million acres of almonds. The remaining 6 million acres of farmland could well be stable and profitable as food costs go up. Pot is sort of a wildcard, but I bet the Great Plains could produce plenty of pot.

I should have some balls, and say which three million acres will no longer be irrigated. Half a million in the Delta. Eight hundred irrigated acres in Imperial and Coachella will drop out as the Colorado dries; there’s no way what is left of the Colorado will go to farmland instead of urban uses. A million acres on the west side of the San Joaquin. The other 700,000 irrigated acres? Maybe they’ll dry up with the Kern and Tulare rivers.

***

On a slightly different note, this level of groundwater overdraft is why I do not believe that there are large quantities of water to be gained from agricultural water conservation. When you include groundwater in your boundaries of a basin, the basins are clearly losing water. If there were substantial slop, on the order of millions of acre-feet, it would have to go somewhere, and we would see it in rising groundwater levels. I fully believe that individual growers can do better, and then the growers one farm over who were using their tailwater will have to back to the original source for more. There are good reasons to do that (keeps water in rivers longer, better water application may improve yields). But I do not expect to see big yields from conserving ag water that could be used for urban or the environment. Rather, I think we’ll see farmland retired and cities get what they need.