ORNL and INL identify 4 gigawatts of untapped hydropower potential at more than 2,600 existing U.S. dams

Only 3 percent of U.S. dams currently generate electricity. Researchers at Oak Ridge National Laboratory and Idaho National Laboratory think that number could go much higher.

A newly published national assessment from both labs identified 4 gigawatts of untapped hydropower capacity spread across more than 2,600 non-powered dams—structures already built but never wired for electricity. If developed, that potential could generate 15.2 terawatt-hours of power each year. That’s enough to supply more than 1.4 million homes.

Assessment finds 4 gigawatts of capacity at non-powered dams

The assessment focuses specifically on non-powered dams—structures already built and serving purposes like water storage, navigation, and flood control, but never retrofitted for electricity generation. Of the roughly 91,000 dams in the United States, only about 3 percent currently produce power.

Flood control schedules, navigation requirements, and other non-power uses can significantly limit how much water is actually available for electricity generation.

Researchers identified more than 2,600 candidate sites with viable hydropower potential. Total capacity across those sites reaches 4 gigawatts, with individual facilities averaging 1.5 megawatts each. That’s a modest per-site figure, but the cumulative impact is real: 15.2 terawatt-hours of annual generation, enough to power more than 1.4 million homes.

The appeal of non-powered dams is pretty straightforward. The civil infrastructure—the dam itself—already exists, which means adding power generation skips the most expensive and time-consuming phase of new hydropower development entirely.

Why previous estimates were less reliable

Earlier national assessments put the theoretical hydropower capacity of U.S. non-powered dams somewhere between 12 and 30 gigawatts. That’s a wide range, and there’s a clear reason: those estimates relied on limited data and didn’t fully account for the technical realities of actually retrofitting a dam for power generation.

The new assessment was designed to close that gap. Researchers used HydroGenerate, a computational tool developed by INL, combined with daily streamflow records from the USGS stream gage network and ORNL’s Dayflow reanalysis dataset. Dayflow simulates how water moves through river networks across the country, providing consistent streamflow estimates even where direct measurements are sparse. That hydrological foundation made a meaningful difference in the reliability of results.

HydroGenerate takes that data and calculates design flow—essentially the optimal turbine size for a given site. It also factors in hydraulic head, which measures the gravitational energy available from water held behind a dam, along with turbine efficiency curves. The result is a realistic estimate of both plant capacity and expected daily electricity output, grounded in long-term operational data rather than theoretical maximums.

Federally owned dams account for the majority of identified capacity

Not all 2,600-plus candidate sites carry equal weight. Federally owned dams dominate the picture, accounting for 86 percent of total identified capacity—a concentration that matters for planning, since federal ownership brings its own regulatory and operational dynamics.

The upper Mississippi River and Great Lakes areas stood out as particularly promising, given the high concentration of large-capacity projects there. Smaller sites also surfaced as real opportunities for utilities and local or state agencies.

One key methodological advance was building operational constraints directly into the model, something previous assessments hadn’t done systematically. Flood control schedules, navigation requirements, and other non-power uses can significantly limit how much water is actually available for electricity generation. By incorporating those constraints into HydroGenerate, researchers produced estimates that reflect what’s achievable rather than what’s physically possible under ideal conditions. HydroGenerate is available as an open-source package, so project developers can run their own site-level feasibility analyses using the same methodology.

New tools give stakeholders actionable site data

The research doesn’t stop at technical findings. It also feeds into NPD Hydro, a resource platform built to help stakeholders evaluate specific non-powered dam opportunities. The platform includes two new tools: NPD Insights, which provides site rankings and updated resource data, and the NPD Toolkit, which helps users estimate payback periods and weigh potential benefits across environmental, community, grid, and industry dimensions.

Updated datasets are available through Oak Ridge National Laboratory‘s HydroSource website, where policymakers, dam owners, and researchers can visualize candidate sites alongside existing energy infrastructure and water management data. The goal is to move from a national-level estimate to site-specific information that can actually drive decisions.

“This refined data will help decision makers focus their efforts on the sites with the greatest possible capacity,” said ORNL’s Carly Hansen, the project’s principal investigator and lead author.

Next steps include future streamflow variability and expanded geographic coverage

The current assessment covers the contiguous United States, but the research team plans to extend it to Alaska and Hawaii in the next phase. Both states have distinct hydrological profiles that require separate treatment.

Climate variability is also on the agenda. Hydropower output depends directly on water availability, which is expected to shift across many regions as climate patterns change—so incorporating future streamflow projections into the model is a logical next step. Hansen noted that in some areas, increased precipitation could actually expand hydropower potential over time.

That long-term perspective matters because hydropower development timelines are long. Permitting, design, and construction for a new facility can easily span decades, which means today’s feasibility decisions need to account for water availability well into the future. The project is supported by DOE’s Hydropower and Hydrokinetic Office within the Office of Critical Minerals and Energy Innovation. Four gigawatts of untapped potential, more than 2,600 candidate sites, better data than ever before, and a set of public tools built to help the people who can actually act on that information do exactly that.

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.