A shrinking river.
Reservoir storage in the Colorado River basin is now approximately equal to two year’s average annual consumptive use. In the three months since reservoir storage peaked in July 2024, drawdown of those reservoirs lost more than 80% of the increase accomplished by the 2024 snowmelt inflow season, which had increased basin reservoir storage by only 2.5 million acre feet despite the Upper Basin snowpack having peaked at a snow water equivalent that was 13.5% greater than the long-term average1. If this rate-of-use continues for the next six months, there will be a net loss in basin reservoir storage. Water supply reliability and security for Colorado River water users can only be accomplished if we replenish the amount of water stored in reservoirs and not further deplete the declining supply.
On 15 October 2024, total contents of the reservoirs of the Colorado River Basin upstream from the Gila River were 27.8 million af (acre feet). This amount of reservoir storage would support two years of consumptive use of the Colorado River2, assuming that basin consumptive uses remain approximately 13 million af/yr, the average between 2021 and 2023. Reservoir storage today is comparable to conditions in mid-June 2021 (Fig. 1) when there was increasing concern among the basin’s water managers about the security and reliability of water supplies provided by the Colorado River. Today, we should be just as concerned as we were in 2021.
Figure 1. Graph showing total basin reservoir storage (blue line), and storage in different parts of the Colorado River watershed between 1 January 2021 and 15 October 2024. CRSP reservoirs are those authorized by the Colorado River Storage Project Act.
The only way to increase the security and reliability of the water supply is to increase reservoir storage, and we are not doing a very good job of achieving that goal. There is no doubt that the large reservoir inflows of 2023 benefitted the basin water supply, allowing us to take a step back from the edge of the cliff of crisis. Basin reservoirs in mid-March 2023 were the lowest they had been (21.3 million af) since late May 1965, when the Colorado River Storage Project’s reservoirs were just beginning to fill and other reservoirs had yet to be built. Snowmelt runoff in 2023 recovered 8.4 million af of reservoir storage, nearly a 40% increase from the March 2023 low point (Fig. 2).
Figure 2. Graph showing reservoir storage between 1 January 2023 and 15 October 2024, highlighting the amount of reservoir recovery during each snowmelt season and the amount of reservoir drawdown during intervening periods.
However, little additional progress in reservoir recovery was made in 2024. We were encouraged that reservoir drawdown during the nine months immediately following the 2023 inflow season was remarkably small, only 2.15 million af and only 26% of the preceding gain in storage. However, snowmelt inflow only resulted in 2.5 million af of gain in reservoir storage in 2024 (Fig. 2).
In contrast to last year, basin uses and losses are much greater this year. In the first three months following the 2024 inflow season that ended in mid-July, reservoir drawdown has been 2.14 million af, more than 80% of the gain of the preceding inflow season (Table 1). Slightly more than half of the drawdown during the last three months has been from the 42 reservoirs upstream from Lake Powell. Those releases supported the needs of mid- and late summer irrigated agriculture, were exported out of the basin, or flowed into Lake Powell. It is likely that the drawdown from these reservoirs will decrease during winter. Slightly more than 30% of the drawdown has been from the combined contents of Lake Mead and Lake Powell. Recent agreements to decrease diversions in the Lower Basin hopefully will reduce drawdown from Mead-Powell combined storage during the next six months. The continued drawdown from Mead-Powell storage will be a robust test of the effectiveness of recent drought management measures.
Table 1. Reservoir drawdown during the first three months following the 2024 snowmelt compared to the total drawdown during the nine months following the 2023 snowmelt season.
Basin water use between now and April 2025 is on track to exceed the inflows of the 2024 snowmelt season, resulting in a net loss of reservoir storage since the bounty of 2023. The persistent decrease in runoff in the 21st century is severely challenging the quest to rebuild reservoir storage. We desperately need to accomplish that goal to avoid another water supply crisis such as occurred between 2020 and 2022.
The only way to replenish the amount of water stored in reservoirs is to decrease reservoir drawdown to match or exceed each year’s gains that occur during the inflow season. For the next six months, that is our goal.
A federal judge this week criticized the federal government for failing to consider the risk of a Colorado River Compact call in its environmental review of the planning for Denver Water’s expansion of Gross Reservoir in Boulder County.
Wrangling over the risk of a compact call – which the judge said could force water use reductions in the Upper Basin if the Upper Basin states fail to deliver enough water past Lee Ferry to the Lower Basin – has been a key point in current negotiations between the two basins over future Colorado River operations.
The ruling, in a lawsuit against Gross Reservoir expansion by Save the Colorado River and others, allows construction to proceed, but criticizes the project’s planners for not considering the fact that the risk of a compact call means there might not be enough water to fill it. (Here’s Michael Booth’s Colorado Sun story on the decision.)
In the decision, federal judge Christine Arguello noted that the Army Corps of Engineers environmental review of the project “rests on the assumption that there will be no compact call…. However, considering the American West’s last few decades of severe aridity, such an assumption warrants considerable scrutiny.”
Here’s the full language from Arguello’s ruling. I’ve bolded the key bits:
The American West’s viability rests in large part on the Colorado River, which is an undeniably overutilized water source. The Colorado River’s over-allocation is due to a fatal flaw baked into the 1922 Colorado River Compact—the bedrock-level agreement that forms the basis of the various overlapping management systems dictating the River’s death by a thousand cuts. The Colorado River Compact rests on a politically unpalatable truth—the Compact promised the basin states water that simply does not exist. According to water experts, the Compact’s apportionment scheme draws from deliberately misleading hydrological modeling of the Colorado River’s flow rates. See generally Eric Kuhn & John Fleck, Science Be Dammed: How Ignoring Inconvenient Science Drained the Colorado River 1 (Univ. Ariz. Press 2019). That fatal error now sits enshrined within the Compact, and all of the basin states’ various agreements with each other (and with the Department of the Interior) are forced to overcorrect for the Compact’s flawed hydrology because revisiting the agreement would require a virtually unthinkable amount of time, energy, money, and political will. However, should Colorado River diversions exceed the river’s available water supply—which appears inevitable, at this rate—the 1922 Compact provides a “compact call” mechanism whereby the upper basin states are bound to curtail their water users’ diversions to ensure enough water reaches the lower basin states and Mexico to satisfy the artificially inflated water delivery obligations set forth in the Compact. This Court emphasizes this context for good reason: the cracked foundation of the Colorado River’s management system all but demands skepticism over any proposal that will affect the hydrology of the Colorado River basin. In the instant case, the Corps noted in its FEIS that its reasoning rests on the assumption that there will be no compact call. AR 159092; but see AR 0014667. However, considering the American West’s last few decades of severe aridity, such an assumption warrants considerable scrutiny. See, e.g., Anne Castle & John Fleck, The Risk of Curtailment under the Colorado River Compact 5, 34–35, 40 (2019) (emphasis added) (“Even if the risk of curtailment of Colorado River rights were assumed to be low, the consequences are not. Cities, and farmers and ranchers on the West Slope, would lose economic activity, jobs, income, and community benefits.”); Cherokee Water Dist. v. City of Colo. Springs, 519 P.2d 339, 340–41 (1974) (“Water is essential to the existence of a community.” (emphasis added)). In light of this context, it is perplexing to this Court that the Corps dismissed the possibility of a compact call in its analysis of a proposed water management project.
Further reading of the judge’s sources:
The Middle Rio Grande Conservancy District Board this afternoon (Monday Oct. 14, 2024) approved an agreement among the District, the Albuquerque Bernalillo County Water Utility Authority, and the Bureau of Reclamation for temporary storage of District water in Abiquiu Reservoir on the Rio Chama while El Vado Reservoir is out of comission.
This is a huge deal. Without it, we could expect low river flows and lousy irrigation supplies through New Mexico’s Middle Rio Grande Valley in late summer for years to come. Even with it, there are still obstacles to good late summer flows – most notably New Mexico’s current Rio Grande Compact debt to Texas. But it’s a huge step.
All of this is necessary because El Vado Dam, where the 1930s relic built to store water during high spring flows for late summer use, is sorta broken. “Temporary” here seems to be as much as ten years.
It’s a hairball of an agreement. (You can read it here, it starts on page 44 of the pdf of today’s MRGCD board packet.)
Abiquiu, built by the U.S. Army Corps of Engineers in the 1950s, began its life as a flood control dam. But its role as a water storage facility has been expanding ever since, with Congress in the early 1980s approving the storage of San Juan-Chama Project water imported from the Colorado River Basin. It is that San Juan-Chama space, under the control of the Albuquerque Bernalillo County Water Utility Authority, that can be used under the new agreement to store native Rio Grande water for MRGCD.
You also get – Heron storage waivers!
The accounting will be tricky. The first water stored will be the top-priority water for the Six Middle Rio Grande Pueblos – a priority granted in the 1920s as part of a deal that provided the necessary up-front funding, via the federal Indian Irrigation Service, to kick-start the creation of the MRGCD. (Bob Berrens and I tell that story in our new book Ribbons of Green, which you’ll be able to read at some point once we finish the revisions currently underway.) The Pueblo water goes by the name “prior and paramount”. Once the P&P water is in, the next layer of stored water has to be equivalent to our New Mexico’s total Rio Grande Compact debt to southern New Mexico and Texas, currently 121,500 acre feet. (But maybe shrinking this year, based on early numbers?)
But wait, there’s more!
After those two accounts are filled, MRGCD will be able to begin storing water for farmers and environmental flows. But there’s a neat twist here to the agreement. Once this last account rises above 10,000 acre feet (it’s called “MRGCD usable water”), the District agrees to keep river flows high enough through the Middle Rio Grande Valley for ABCWUA to operate its drinking water plant.
But wait, there’s more!
In addition to all the above, if you act now (the ABCWUA board will take this up in a couple of weeks), you also get, at no additional charge, storage waivers in Heron (I’m sorry, this is already too complicated for more background, Heron’s another reservoir, it’s on Willow Creek, JFGI, it’s late and I’m tired.) for San Juan-Chama water that would have otherwise been parked in Abiquiu!
Update: Data here is from USDA, the “USGS” in the graphs is a typo in the code I used to generate them, which I’m too lazy to fix.
New Mexico alfalfa acreage is the lowest it’s been since the 1950s.
The U.S. Department of Agriculture’s 2024 estimate of alfalfa acreage in New Mexico, 130,000 acres, is the lowest it’s been since the 1950s. Acreage is down 55 percent since the peak at the turn of the century.
Here’s Arizona:
Different story for Arizona.
California:
California, growing less alfalfa
Colorado
The Arizona Department of Water Resources has published a thoughtful and also delightfully testy response to Enduring Solutions on the Colorado River, the white paper Kathryn Sorensen et al. (I’m one of the et alias) published in August.
First a reminder of our core premise:
As we work to reduce water use on the post-2026 Colorado River, two paths lie open before us.
One is to incentivize conservation by giving water users the chance to bank saved water for later use. Known most commonly as Intentionally Created Surplus (ICS), and more broadly in a series of increasingly creative implementations as “Assigned Water,” this creates short term savings. But in the long run, the approach entitles the users to take the water back out of the bank.
The other involves permanent reductions – “System Water.” Water use is reduced for the benefit of the Colorado River as a whole.
Investment in Assigned Water, attractive to water managers because of the allure that they can get their water back, has crowded out investment in the more durable System Water reductions that will be needed to bring the Colorado River into balance.
ADWR agrees with a lot of what we said (yay!):
ADWR agrees wholeheartedly with the general premise that System Water is preferable to any category of Assigned Water. Increased volumes of System Water will improve outcomes for water users across the entire Basin, as well as the environment. ADWR also agrees with the need to divorce decisions regarding system operations from any Assigned Water stored in the system.
ADWR takes issue with a lot of what we said (also yay!), and also was kinda mean – “Baseless Accusations and Little Substance” (also yay! because we need to be having these frank discussions – I worry being too polite isn’t helping right now).
I’m delighted this discussion is underway in a public forum. Go read the whole thing.
Water resources must be governed before they can be managed, and the crafting of governance institutions is an ongoing challenge of the human condition. Institutional artisanship requires the availability of tools for institutional design and creation, and a reflective understanding of the use of those tools. Yet neither the tools nor the understanding exist in their entirety before institutional artisans attempt their craft: the tools at hand must be adapted to new purposes, and skill in their use must be learned along the way. Errors are inevitable, but successes are possible.
Blomquist, William. “Crafting water constitutions in California.” The practice of constitutional development: Vincent Ostrom’s quest to understand human affairs (2006): 99-134.
Bouncing along the bottom.
Sept. 30 marks the end of the “water year,” an accounting milestone that gives us an opportunity to take stock.
The change in total water storage year-over-year is one way to do this, to help understand if we took more water out of the reservoirs than the climate put in. The graph above is actually based on Sept. 20 year-over-year (the Reclamation data updates lag a bit), but it’s enough to give us a feel for two things.
First, we’ve seen no real reversal of the long term pattern – a huge reduction in storage in the early 21st century, and then basically dragged the bottom of the reservoirs ever since.
Second, on a shorter one- or two-year time scale, total storage is down ~350,000 acre feet at the end of water year 2024 compared to the end of 2023. Over a two-year time scale, we basically burned through the bonus water from a wet 2023 and are back where we were at the end of 2022.
Rio Grande flow this year at Otowi in north-central New Mexico has been 63 percent of the period of record mean, going back to the late 1800s.
When John Oliver used some of my work in a piece about water a couple of years ago, I spent a long time with one of the show’s producers as they developed it. Then, right before it aired, I spent more time in an intense round of fact-checking.
It was marvelous. They were fact-checking the jokes.
In a recent interview, Oliver explained the process:
But in terms of the responsibility of journalism, we do have intense fact-checking because we want it to be right. Those big stories are aggregations of incredible journalism. So it cannot function without journalism. Now, we recheck it to make sure it’s accurate or that it hasn’t changed, but we’re building this to make jokes. It’s just we want the foundations to be solid or those jokes fall apart. Those jokes have no structural integrity if the facts underneath them are bullshit.
