The management of dams has emerged as a critical area of focus due to their dual role in flood mitigation and the risks of catastrophic failure associated with them. Recent studies indicate alarming trends regarding dam overtopping probabilities across the United States, as evidenced by an analysis covering 33 major dams from 1973 to 2022.
Dams are essential infrastructures that help in regulating water flow and storing excess water during heavy precipitation. However, when faced with hydrologic extreme events, such as floods exacerbated by climate change, these structures can pose significant threats. Notable instances of dam failures, such as the 2017 spillway failure of Oroville Dam, which prompted evacuations of over 180,000 people, and the more recent failures in Michigan and North Carolina, underscore the potential catastrophic consequences.
The American Society of Civil Engineers has expressed concerns regarding the state of U.S. dams, assigning them a D+ grade, which marks a decline from previous evaluations. There is an estimated investment requirement of 185 billion U.S. dollars to address the existing issues confronting these structures. Given that the threat of dam failures is a global concern, the 2023 tragedies in Libya caused by dam failures—resulting in 11,000 deaths and 19 billion U.S. dollars in economic losses—serve as a dire reminder of this reality.
Many dams in the U.S. are aging and often do not comply with modern safety standards. Among the nearly 16,000 classified as “high-hazard,” their failure could result in significant loss of life and property damages. These historical and ongoing challenges have raised concerns about how best to manage and upgrade aging dam infrastructure safely and effectively.
Recent research has transitioned from basic risk and reliability analyses to more complex, comprehensive systems that integrate a wider array of factors affecting dam safety. For example, investigations in Taiwan utilized annual maximum series data for risk assessments, while studies in China evaluated overtopping risks by factoring in various uncertainties, including storm impacts and reservoir characteristics.
Despite advancements, research still faces limitations; many studies focus on a limited number of dams, and the nonstationarity of hazards due to climate change and evolving hydrologic conditions often remains unaddressed. A notable study highlighted the elevated risks of dam failures across the United States due to compound rainfall clusters but did not directly assess the impact of water levels on overtopping probabilities.
The recent study aimed to analyze overtopping probabilities at 33 major dams in the United States over a 50-year span, specifically targeting those with long-term public domain data. These dams were not only representative due to their longevity but also high-hazard potential. By defining overtopping as uncontrolled water levels exceeding dam crests, the study assessed how overtopping probabilities increase the risks of catastrophic failures.
The analysis utilized updated stationary frequency analysis by fitting Generalized Extreme Value (GEV) distributions to rolling 30-year windows of annual maximum water level data. This approach enabled the identification of overtopping probabilities that may not be apparent through simple trend analyses. Dams were classified into four probability levels—very low, low, moderate, and high—based on their likelihood of overtopping, providing a clearer picture of risk.
Observations revealed significant changes in overtopping probability levels over time, particularly at the very low and low levels. Notably, the number of dams classified as very low decreased while those at low probability levels increased, highlighting a concerning shift towards higher overtopping risk over the studied period.
Analyzing temporal changes suggested that the number of critical dams experiencing heightened risk now exceeds those classified as non-critical. Furthermore, an assessment focusing on critical and non-critical dams made it evident that the proportions of very low probability dams have declined, indicating an upward trend in overtopping probabilities.
The study identified specific regions with alarming exposure; for instance, dams in the Southern Great Plains often faced higher overtopping probabilities due to the climate’s propensity for intense storms and extreme weather events. Among the 33 dams, 11 remained classified with critical probabilities for over half of the study period, emphasizing their potential hazards, especially all of which are classified as high-hazard.
Statistical analyses of overtopping probabilities yielded significant findings, with twelve dams exhibiting notable trends—some showing increasing risks while others displayed decreases. The alarming trends were particularly pronounced among dams in heavily impacted regions, necessitating immediate attention to dam safety management practices.
The study further identified the dams with the greatest overtopping probabilities, emphasizing their proximity to large population centers, which heightens the risks associated with potential dam failures. For instance, Whitney Dam has a downstream population of approximately 140,000 residents, while others like Somerville Dam are much closer to residential communities, emphasizing a critical need for preventive actions and structural evaluations.
The authors highlighted that dams operated by the U.S. Army Corps of Engineers and the Bureau of Reclamation face unique challenges regarding data collection and risk assessment. Extensive filtering processes established the final assessment of 33 independent dams suitable for frequency analyses.
Future assessments of dam overtopping probabilities must consider the evolving impacts of climate variability and watershed characteristics. As evidenced by the findings, there is a pressing need to revisit current operational strategies to address the increasing challenges posed by extreme weather events.
The increasing risks associated with dam failures and overtopping necessitate not just immediate preventive and management strategies but a fundamental reevaluation of long-term hydrologic safety protocols to protect communities that rely on these critical infrastructures.
image source from：nature
