Cornell study finds Colorado West Slope basins face up to 50% drop in Lake Powell inflows by midcentury

A Cornell University study published July 13 in Earth’s Future warns that Colorado’s six West Slope river basins—which supply roughly 70% of Lake Powell’s water—could see inflows to the reservoir drop by as much as 50% by midcentury. Lake Powell serves as a critical source of water and energy for 40 million people.

Lead author Sai Veena Sunkara and senior author Patrick Reed set out to identify the key forces driving water shortage risk across the region. What they found was more complicated than either expected.

Study scope and methodology

Cornell’s study is being called the most extensive analysis of its kind. It builds on a 2024 modeling effort that found Colorado’s West Slope basins may be approaching a tipping point—where traditional water delivery levels to Lake Powell become unsustainable.

Department of Energy’s Office of Science, specifically through the Multisector Dynamics area of the Earth and Environmental System Modeling program.

To push further, Sunkara expanded the range of conditions under examination. She modeled everything from extreme dryness to extreme wetness, layering in variables like streamflow changes, snowmelt timing, drought persistence, and water demands from agricultural, municipal, and industrial users. The result: 20,000 possible midcentury scenarios representing 2.1 million simulated years.

Sunkara pulled demand projections directly from Colorado state planning documents for each of the six basins: the Upper Colorado, Gunnison, Yampa, White, San Juan, and Dolores. Grounding the model in official planning data gave it a realistic baseline for how human water use might evolve — not just how the climate might shift.

Key drivers of water shortage risk

Despite real differences across the six basins, researchers found a consistent midcentury signal. Major reservoirs across the region are projected to sit 40% to 55% below their historical median storage levels—a significant structural deficit that strips the system of its main buffer against dry spells.

One driver stands out above the rest: snowmelt timing. Reed put it simply—snow is a form of storage. When it melts too early in the season, less water is available later to offset dry periods, and that shift in timing emerged as a dominant, system-wide factor cutting across all six basins.

Demand-side drivers are more complicated. Agricultural water use can rise or fall depending on land conversion, irrigation efficiency improvements, or changing crop patterns. Sunkara noted that all of these factors were incorporated into the model as increases or decreases in demand, depending on the basin and the scenario.

Water rights add another layer. By folding in the legal and institutional framework governing who gets water and when, the model captures what Reed calls the “human institutional management” side of the equation. That combination of natural and human variables makes the picture considerably more complex—and more realistic—than earlier approaches.

The researchers were direct about one conclusion: no single management strategy can address all future water shortages. Drivers differ by basin, by sector, and by user. A solution that works in the Yampa may not translate to the San Juan.

Implications for water management and policy

The study doesn’t offer a fixed forecast. It’s a map of plausible futures—a way of identifying where the pressure points are and which interventions are most likely to help.

“It changes by basin, changes by sector, changes by user,” Reed said. “You have to account for water rights. You have to account for the natural drivers as well as the human drivers. It was more complex than we initially thought.”

That complexity has direct policy implications. Reed emphasizes that effective responses will need to be basin-specific and sector-specific rather than broad regional fixes. The model is designed to highlight where actions like demand management are most likely to be effective—and where they’re not.

The urgency behind this work is real. Lake Powell, the nation’s second-largest reservoir, has been under severe drought conditions for roughly 20 years, supplying water and hydroelectric power to tens of millions of people across the Southwest. Coordinated, targeted planning isn’t a future consideration—it’s already overdue.

The study also points to a fundamental challenge in modeling this region: decadal and multidecadal droughts are rare events, and historical streamflow data doesn’t provide enough statistical samples to capture them reliably. That’s precisely why the researchers generated millions of simulated years—to account for the kinds of extreme events that records alone can’t anticipate.

Funding and research team

The study was funded by the U.S. Department of Energy’s Office of Science, specifically through the Multisector Dynamics area of the Earth and Environmental System Modeling program.

Co-author David Gold, Ph.D. ’22, contributed to the research in the Cornell Chronicle during his time as a postdoctoral researcher at Cornell. He’s now an assistant professor at Utrecht University. The findings were published July 13, 2026, in the journal Earth’s Future.

The core takeaways are worth keeping in focus. Colorado’s six West Slope River basins face a potential 50% reduction in Lake Powell inflows by midcentury. Reservoir storage across the region may already be running 40% to 55% below historical medians, and early snowmelt is emerging as a dominant, system-wide risk factor. Demand pressures—shaped by water rights, land use, and agricultural shifts—vary enough by basin that no single policy response will cut it. The research doesn’t predict one fixed future. It maps the range of what’s possible so planners and policymakers can start making better decisions now.

Kelly Lippke

Kelly is an experienced writer with 15 years of experience exploring the big stories that shape our world, from tech breakthroughs and space exploration to climate, energy, and the fascinating quirks of science. She has a talent for turning complex ideas into sharp, memorable insights that stay with readers long after they’ve finished reading.