Deadpool Diaries: Paying to fallow may not save as much water as we think

Given that we’re about to spend a billion dollars to fallow land to reduce water use in the Colorado River Basin, it’s reasonable to ask how we can be sure we know how much water is actually being saved.

The answer, according to new work by Katharine Wright and colleagues at Arizona State University, may be “not as much as we think.”

Wright and her colleagues looked at the oldest and most well-understood rotational fallowing program in the west – the Metropolitan Water District-Palo Verde Irrigation District deal set up two decades ago to fallow land in PVID and use the water to shore up Met’s Colorado River Aqueduct supplies.

Their key finding: over the time period studied, actual savings were far smaller than MWD’s estimates.

The key difference in their work, as compared to the techniques used to derive official estimates, is an effort to analyze farmer behavior – the choices of which fields to enroll in the fallowing program and farmers’ cropping and irrigation practices on non-fallowed fields. If farmers fallow fields that would have used less water anyway, or use more water on the non-fallowed fields, it could sway the estimates.

In other words, water saved by fallowing is not simply a physical science question. It is a behavioral science question as well.

To be clear, I’m not saying Wright and her colleagues are correct here. The technical details are complicated, and I’m not qualified to judge. I’m saying that this is a serious and importantly, independent effort to estimate how much water is really being saved.

As we prepare to spend a billion dollars to fallow a bunch of land to conserve a bunch of water, we should be attentive to the incentives. Who has an incentive to make the savings we’re getting for our billion dollars look big? Who has an incentive to maximize the amount of water we get in the reservoirs for our billion dollars?

A huge thanks to my colleagues at the University of New Mexico Water Resources Program, who have been looking at rotational fallowing here in New Mexico (we’re all wrestling with these questions) and brought Wright’s work to my attention.

And as always, a big thanks to Inkstain supporters, who make this possible.


  1. I don’t know the source of information you are quoting, but when my PVID committee negotiated the PVID-MWD fallow program we agreed to an estimated recovery of 4.6 AF/acre of consumptive use, with a true-up annually. The actual recovery over the (so far) 19 years of the program has proven to be 4.71 AF/acre. Greater than expected.

  2. I’m curious about your take on ADWR’s post:

    “The results of the numerical basin-scale groundwater flow model projection show that over a period of 100 years, the Phoenix AMA will experience 4.86 million acre-feet (maf) of unmet demand for groundwater supplies, given current conditions. The term “unmet demand” refers to the amount of groundwater usage that is simulated to remain unfulfilled as a result of wells running dry in the model. To show the physical availability of groundwater under the Assured Water Supply (AWS) program, existing and assured water supplies need to be fully met.

    In keeping with these findings of unmet demand, the State will not approve new determinations of Assured Water Supply within the Phoenix AMA based on groundwater supplies. Developments within existing Certificates or Designations of Assured Water Supply may continue, but communities or developers seeking new Assured Water Supply determinations will need to do so based on alternative water sources.”

  3. I sure hope they don’t spend a billion dollars and not verify the water savings.

Comments are closed.