The United States is not using nearly as much hydropower as it could be, new research has found.
Researchers at the Department of Energy’s (DOE) Oak Ridge National Laboratory (ORNL) and Idaho National Laboratory (INL) issued an updated assessment of hydropower technical potential across the U.S., finding that as much as 15.2 terawatt-hours of electricity could be generated annually at non-powered dams.
Only three percent of dams in the U.S. have been used for hydropower generation, with the rest serving other functions like water storage, navigation, and flood control. The remaining non-powered dams (NPD) “present an opportunity for domestic energy production,” ORNL said. ORNL’s previous assessment found potential capacity of four gigawatts (GW) across more than 2,600 NPDs, with individual facilities averaging 1.5 megawatts (MW).
Federally owned dams account for 86% of this total potential capacity, and the upper Mississippi River and Great Lakes are “particularly promising” for hydropower development given the large number of projects with high potential capacity, ORNL said.
“The potential is heavily concentrated in large, federally owned dams, but we also see many opportunities for utilities and local or state agencies,” said ORNL’s Carly Hansen, the project’s principal investigator and lead author of the report. “This refined data will help decision makers focus their efforts on the sites with the greatest possible capacity.”
Past assessments estimated the nation’s capacity of NPDs between 12 and 30 GW, but ORNL argues these estimates were “hindered by a lack of detailed data and an incomplete representation of technical factors relevant to hydropower retrofits.” The recent assessment, on the other hand, aims to address this gap by including “more sophisticated methodologies and detailed datasets” to offer a clearer picture, ORNL said.
“This project was really motivated by the need to improve confidence in what’s technically possible,” Hansen said.
For the latest assessment, researchers utilized a tool developed by INL called HydroGenerate, combined with daily records of streamflow. After analyzing long-term hydrological data, HydroGenerate calculates the design flow, hydraulic head, and information on turbine efficiency to calculate information like expected plant capacity and daily electricity generation. These tools have allowed for more accurate projections of operational constraints and seasonal variations in water availability and movement, ORNL argues.
“By using HydroGenerate, we are considering different types of turbines suitable to the conditions of the flow and hydraulic head at each site,” said INL researcher Juan Gallego-Calderon. “The tool is also offered as an open-source package so stakeholders such as researchers and project developers can use it to verify their own feasibility assessments.”
ORNL and INL researchers plan to add future streamflow variability into the model to capture long-term impacts of changing water availability on these potential resources, and the next phase will extend to regions like Alaska and Hawaii.
“Hydropower development takes time, often decades, so we need to consider what future water availability might look like,” Hansen said. “In some regions, increasing precipitation could even create greater potential for hydropower in the future.”
Facts Only
* 15.2 terawatt-hours of electricity could be generated annually at non-powered dams.
* Only three percent of U.S. dams have been used for hydropower generation.
* Remaining dams serve functions like water storage, navigation, and flood control.
* Non-powered dams present an opportunity for domestic energy production.
* ORNL’s previous assessment found potential capacity of four gigawatts (GW) across more than 2,600 non-powered dams.
* Individual facilities in the previous assessment averaged 1.5 megawatts (MW).
* Federally owned dams account for 86% of the total potential capacity.
* The upper Mississippi River and Great Lakes are identified as particularly promising for hydropower development.
* Researchers utilized the HydroGenerate tool combined with daily streamflow records for analysis.
* HydroGenerate calculates design flow, hydraulic head, and turbine efficiency to estimate plant capacity and generation.
* ORNL and INL plan to add future streamflow variability into the model.
Executive Summary
New research from the Department of Energy’s Oak Ridge National Laboratory and Idaho National Laboratory assesses the untapped hydropower potential across the United States. The analysis indicates that up to 15.2 terawatt-hours of electricity could be generated annually from non-powered dams. Currently, only three percent of U.S. dams have been used for hydropower generation; the remainder serve other functions such as water storage, navigation, and flood control. Researchers found that the remaining non-powered dams present an opportunity for domestic energy production. Federally owned dams account for 86% of this total potential capacity, with the upper Mississippi River and Great Lakes identified as particularly promising areas for development.
The study addresses previous estimates, which suggested national NPD capacity ranged between 12 and 30 gigawatts (GW), arguing that those earlier figures were limited by incomplete data regarding technical factors for retrofitting. To improve accuracy, researchers employed the HydroGenerate tool, which analyzes long-term streamflow data to calculate design flow, hydraulic head, and turbine efficiency, allowing for more accurate projections of capacity and generation considering operational constraints and seasonal variations. Future work plans involve incorporating projected streamflow variability and expanding the modeling to regions like Alaska and Hawaii.
Full Take
The narrative pivots on the gap between existing infrastructure reality and theoretical energy potential. The shift from previous estimates (12-30 GW) to the current assessment suggests a struggle against data limitations, positioning the new methodology as a necessary corrective force for building public confidence in technical feasibility. The concentration of potential capacity in large, federally owned facilities highlights an existing structural bottleneck, where opportunities are heavily channeled through existing governmental structures rather than dispersed across all available assets.
The reliance on complex modeling tools like HydroGenerate and the explicit plan to incorporate future streamflow variability reveal an underlying tension between current engineering assessments and long-term environmental uncertainty. The implication is that perceived energy potential is often constrained by insufficient data—a common theme in large-scale infrastructure planning where unknown variables introduce systemic risk. Furthermore, the focus on "improving confidence" suggests a process where knowledge generation is intentionally used to drive action, necessitating a critical examination of which forms of certainty are prioritized and who benefits from those refined projections.
What assumptions underpin the prioritization of capacity in specific regions like the Great Lakes? What governance structures facilitate or impede the dispersion of these opportunities among local and state agencies versus federal entities? How does the incorporation of future variability change the risk calculus for long-term infrastructure investment versus immediate operational realities?
Sentinel — Human
The text reads like a summary of a research report, demonstrating high internal coherence and specific sourcing typical of institutional reporting rather than pure generative output.