Despite heavy rainfall in recent months, Lake Mead could return to near-historic lows by 2025. As the Bureau of Reclamation looks to reach an agreement by the end of the year — before a potential change in administration — the agency, for the first time, And putting climate change at the heart of how we plan for the future.
Those who depend on the river are now testing water-sharing strategies using the agency's new online tool that harnesses more than 8,000 potential futures for the river to see how policies can confront the wild fluctuations and uncertainties caused by a warming climate.
“We ultimately have a very broad range of conditions that can occur under climate change,” reclamation official Rebecca Smith said during a symposium in November. “Scientists do not expect this number to shrink any time soon.”
The driving force for this innovation is frightening: the realization by policymakers that the way they predicted the river's future is no longer working.
By relying on records of river flow over the past century, the federal government and Western states have repeatedly downplayed the importance of drought and failed to prevent major reservoirs from approaching dangerous levels that could threaten the water supply of millions of people. Over the years, Reclamation has looked to climate science to solve the real-world problem of the shrinking Colorado River, according to river experts and people involved in the effort. However, deals have been made with states over the past two decades To deal with drought not incorporated Climate change models In river simulations used to develop long-term policy.
This is changing now. The new approach to reclamation will test policies against a future informed by climate models that project continued warming, as well as tree-ring records of ancient droughts far worse than anything in the recent past — which include a much wider range of potential river flows than is available in the world. Modern historical record.
“We're teetering here on the edge of being able to say, yes, climate change models are part of our decision-making process,” said Terry Volpe, who spent three decades in reclamation projects, including eight years on the Lower Colorado River Project. Regional director of the basin until his retirement in 2020. “It's been a long time coming.”
To do this, a scrappy group of government officials and academics dives into a new world of unpredictability.
Their approach is called “decision making under deep uncertainty.” This highlights the idea that anyone can predict what the river's future flow might be — a strategy that has left the reclamation process in “crisis management mode,” said one official who was not authorized to speak publicly.
The severity of the latest crisis eased about a year ago. The country's second-largest reservoir, Lake Powell, has nearly declined to the point where the hydroelectric dam can no longer produce power. They were saved by an unusually wet winter and a short-lived agreement with states to conserve water in exchange for billions of dollars in federal money. But scientists warn that renewed drought could quickly threaten the region again.
“Last year was probably the anomaly,” said Tom Bushatzky, director of the Arizona Department of Water Resources.
Reconsider previous assumptions
When Volpe joined Reclamation in 1989, the future of the Colorado River was planned on a single computer in an office building in suburban Denver.
The lumbering mainframe computer in the Lakewood office took up an entire room and slowly spun out rows of numbers on perforated paper — a simulation of how the West's most important river would be distributed among its 10 major reservoirs for years in the future.
The only people who could use this model — known as the Colorado River Simulation System, or CRSS — were federal government water managers, like Volpe. To make modifications, his code was rewritten in an old computer language called Fortran. It was “really stressful,” Volpe recalls.
In the early 1990s, Volpe collaborated with academics at CU Boulder's Center for Advanced Decision Support for Water and Environmental Systems to develop a program known as RiverWare and a version of the CRSS model that could be used by others who depend on the river, from states to the United States. Cities for farmers and tribes.
In the Colorado River Basin, states were competing more for water; New federal environmental laws require comparing alternatives before making resource decisions; Many of them do not trust the federal government.
“They all hated reclamation,” recalls Edith Zagona, a former Reclamation official who runs the UC Boulder center. “We had to prove to them that this new program could work. They didn't want someone to attack them so quickly.”
Over the years, the software and the models that power it have become an essential tool for resolving disputes, avoiding litigation between competing parties and reaching agreements about how water is stored and distributed. It is used on many major rivers in the United States, including Columbia River and Big river.
The CRSS model simulates the flow of water through canals, pipelines and reservoirs as the Colorado River crosses 1,450 miles, seven states and two countries – from the snow-capped peaks of the Rocky Mountains to the Sea of Cortez. That law includes the rules governing who gets priority access to those waters — the complex web of regulations, laws, treaties and court cases dating back more than a century, known as the law of the river.
It can reveal how changes in water supply, demand, or the policy environment will affect issues including reservoir levels, hydropower production, flood control, or protecting endangered fish that swim through the Grand Canyon.
To operate it, water managers must estimate the downstream flow entering the river system. Since the model was first developed, they have relied on it What was observed on the Colorado River dates back to 1906.
Scientists knew this period to include the very wet years of the 1920s, when the Western Water Sharing Agreement was struck, and the 1980s, when Lake Powell nearly overflowed at Glen Canyon Dam. But the twentieth century record was seen as the best way to understand what the river was capable of.
“It's been almost universal in water planning: the assumption that the future will look like the past,” said Jeff Lucas, a climate scientist and Colorado River researcher.
But after the drought began in 1999 — and Lake Mead and Lake Powell began their long decline — there were growing doubts about the utility of the 20th-century record. One major warning came in the form of tree rings.
Rings found inside logs, logs, and dead trees tell a much older story about how much water is available in the Colorado River Basin. Researchers in 2007 created a record dating back to 762 AD, which revealed punishing medieval droughts, including a 62-year drought in the 12th century that produced streamflows lower than anything in the 20th century.
As levels in Lake Mead and Lake Powell dropped from almost completely full in 2000 to about half five years later, Reclamation and the states negotiated an agreement on how to operate the reservoirs as they declined — and how the shortfall would be divided.
In 2007, modeling predicted that Lake Powell had a zero percent chance of rising so low that it would no longer be able to help produce energy by 2026. Last year, states struck an emergency agreement to make unprecedented reductions in water use to avoid that very fate. .
Drawing the future of the river
In 2012, after a decade of drought, Reclamation published a major study assessing the health of the Colorado River and what a warming climate means for its future.
It was not a simple answer. Leading global climate models needed to adapt to the Colorado River region — where the terrain in the Rocky Mountains varies widely — and then combine them with other models to estimate streamflows. While the results predicted continued warming in the region, lower snowpack in the Rocky Mountains and more frequent droughts, it was – and still is – less clear how much rain and snow will fall each year.
“The precipitation portion of the climate model output is not good,” said Brad Udall, a hydrologist and climatologist at Colorado State University. “We still have problems, especially at the regional level, in terms of getting things right.”
A RAND Corporation scientist named David Groves, who was hired to help with the 2012 study, saw this uncertainty as an opportunity to rethink how river modeling works. Rather than trying to make incorrect predictions about what the Colorado River will look like in the future, he and his colleagues advocate testing different river management strategies against thousands of potential river futures drawn from a wide range of tree-ring sources. The record, the climate models, the historical record – to see how well it holds up.
At the time, such intensive computer modeling was slow. In November 2014, Groves convened a group of Colorado River stakeholders at Lawrence Livermore National Laboratory in California to show how the lab's supercomputers could speed up the process. He tested the policies against 12,000 river scenarios, something that would typically take six weeks.
“While we were feeding them lunch, we turned on the supercomputer and it did the entire analysis in an hour,” Groves said.
He said that at the time there were a lot of potentially ominous outcomes in his analysis, including the disappearance of Lake Mead if consumption patterns continued.
“People saw it and it worried them,” Groves recalls. “But even if someone says that’s what could happen, you can still say: ‘Well, that’s far-fetched.’”
By the fall of 2022, Lake Mead and Lake Powell were only a quarter full. Reclamation had been anticipating a greater than 50% chance that these lakes would fall below critical thresholds without further action, although the outlook is now less dire.
The new web tool, created by Virga Labs, a water consultancy, in partnership with the Zagona Center and the Land Reclamation Foundation, is the culmination of years of work on the science of “decision making under deep uncertainty.” Virga Labs CEO Season Martin said the company is using cloud computing to run the CRSS model with thousands of variations in ways that were not possible in the past.
But the tool also represents a more fundamental change in how Reclamation approaches the problem of the dwindling Colorado River. Instead of trying to determine probabilities about future risks, the agency aims to test different approaches to river management against the widest possible range of river scenarios — and see what works best to avoid disaster.
“We want to know that things are going to be okay even under really tough conditions,” Smith, the reclamation official, said in November.
Bushatzky, Arizona's top water official, called the new modeling tool “a very positive step forward,” because it allows people or entities without technical expertise to explore changes in river management. But he expects his agency will continue to run numbers and test strategies the old-fashioned way, too.
“We spent enough time and effort learning how to do this,” he said.
Reclamation said it will also use other analytical methods it has relied on in the past as it evaluates proposals states are expected to submit in March on how to share a river strained by climate change.
Not everyone is confident that this new approach will yield better results, or prevent reservoirs from drying up in the future.
Udall, the Colorado State University scientist, said sophisticated new tools may give the illusion of being able to find a safe path forward for a river whose natural flow has declined by 20 percent over the past two decades. Simply reducing the amount of water states can consume by a significant amount might be wiser, he said.
“I think we've become half as smart with all this technology,” he said. “I think you can make this process a lot easier. Which is what we need to plan for much lower flows, and let's not hide behind or obscure some of these difficult details with really complicated models.”
