By Jay Landers
The risk-management programs used for dam safety by three key federal agencies are “appropriate and sound,” according to the recently released findings of an independent external peer review panel convened to assess the use of risk-informed dam safety practices by the U.S. Army Corps of Engineers, the U.S. Bureau of Reclamation, and the Federal Energy Regulatory Commission.
However, the congressionally mandated review offered several recommendations for improvement regarding the use of risk-informed decision-making in the dam safety arena.
The panel’s final report, while dated Sept. 15, 2020, was publicly released for the first time by the Corps in the spring.
Originating from Oroville
The panel sought to evaluate the agencies’ “use of risk in their management and regulation of dams,” according to the report. The panel also aimed to “provide recommendations to improve the risk policies and methodologies of each agency” and “identify any lessons learned in both the risk informed decision making process and/or design and construction,” the report notes. More broadly, the panel also considered how the practices of the agencies “influence or could influence broader dam safety practices in the United States.”
The review was prompted, in part, by a 2017 incident at Oroville Dam, the 770 ft high structure that was completed on California’s Feather River in 1968. In February 2017, during a period of high flows on the river, the California Department of Water Resources had to halt releases down the dam’s main spillway after it sustained severe damage from the water. A few days later, the DWR released water over the dam’s emergency spillway.
After severe erosion then threatened to undermine the concrete weir at the top of the emergency spillway and potentially cause uncontrolled releases from the dam, downstream communities were forced to evacuate. The DWR resumed releases down the main spillway, which experienced yet more damage, but discontinued releases from the emergency spillway (see “California Moves to Repair Oroville’s Damaged Spillways While Analyzing Causes of Failures,” Civil Engineering, June 2017, pages 18-20).
The near-disaster at Oroville Dam, the nation’s tallest dam, prompted the formation of an independent forensic team to review the incident, assess its causes, and make recommendations (see “Multiple Factors Led to Spillway Problems at Oroville Dam, Report Says,” Civil Engineering, February 2018, pages 29-31).
The incident also spurred other dam owners, including federal agencies, to evaluate their practices.
“Many agencies including USACE took a much closer look at how we have been evaluating safety at our dams,” said Nate Snorteland, P.E., the director of the Risk Management Center within the Corps’ Institute for Water Resources, in a June 7 news release announcing the release of the 2020 dam safety report.
“Congress was also interested in what the Oroville incident meant for the nation and for federal dams specifically,” Snorteland said. “They asked USACE to coordinate an independent review of the risk policies and methods used to assess risk across the three major federal agencies that own, operate, or regulate dams in the U.S. It was Congress’ intent to inform improvements broadly in national dam safety practices.”
A diverse panel
John France, P.E., D.GE, D.WRE, M.ASCE, an independent consultant and sole proprietor of JWF Consulting LLC, chaired the five-member panel. (France also led the team that investigated the Oroville incident.) The other panel members were Gregory Baecher, Ph.D., Dist.M.ASCE, NAE, a professor of civil and environmental engineering at the University of Maryland; Rudolf (Ruben) Jongejan, Ph.D., the statutory director and principal consultant of Jongejan Risk Management Consulting BV, which is based in the Netherlands and specializes in probabilistic design and flood risk analysis; Shane McGrath, CPEng, a board member of Dams Safety NSW and the director of SGM Consulting (Aus) Pty Ltd; and Ali Mosleh, Ph.D., a professor of reliability engineering at the B. John Garrick Institute for the Risk Sciences at UCLA.
The panel members were selected with an eye toward ensuring diverse viewpoints, France says. “We did want some international perspective, particularly from a couple of countries that have been doing a lot with risk and risk analysis for dams and levees,” he says. “Australia certainly is in that camp.” Meanwhile, “the Netherlands, with their dikes and their dike system, have similarly been doing a lot with applying risk analysis,” France notes.
At the same time, the panel also included a mix of academics and practicing engineers, at least one of whom, Mosleh, was familiar with the practice of risk analysis in other sectors. “We wanted the panel to be partly academic, partly practicing engineers, with some international input as well as some input from what's going on in other industries,” France says.
Thinking shifts on risk
The past few decades have witnessed a transition in the approaches used by the Corps, Reclamation, and FERC to assess dam safety. “Prior to the 1970s, dam safety was managed almost exclusively using defined standards,” according to the report. “This led to binary conclusions regarding the safety of a dam as either safe or unsafe. The conclusion was based on a comparison of engineering calculations relative to a defined standard.”
“You'd come up with some design event,” France explains. “Then you would analyze the structure for that design event, and you would apply factors of safety to it.” This approach resulted in a “very deterministic” finding based on whether a dam met or did not meet the factor of safety, France says.
At the same time, little attention was given to the likelihood of different outcomes regarding loss of life in the event of a dam failure, France says. “Whether the failure would likely cause one fatality or cause 1,000 or 10,000 fatalities, they'd all be considered equal,” he says. “There really was no explicit consideration of what the actual consequences might be in terms of loss of life.”
Since then, the Corps, Reclamation, and FERC, to varying degrees, have gradually shifted to the RIDM approach for evaluating dam safety. Under this approach, a dam’s risk is assessed in terms of a wide range of loads, the likelihood of those or greater loads occurring in any given year, the structure’s response to the loads, and the consequences of failure. The resulting computations then are compared to various risk guidelines, depending on the agency involved.
The agencies then use the findings determined by the RIDM process to help decide which dams are at the greatest risk and where money should be first used to take care of them, France says.
Ultimately, the RIDM approach helps ensure that dam safety funding is used to address the highest risks. Before risk analysis, a lot of the money spent for dam safety was going to build things for events that might be extremely unlikely to happen; these included massive spillways for unlikely floods and dam reinforcement for unlikely earthquakes, France says. However, much less funding was dedicated to addressing potential failure modes such as seepage or internal erosion “because the criteria for them weren't as easily quantified,” he notes.
After applying risk analysis, the agencies realized that many of their dams were much more at risk of failure from such everyday concerns as seepage or internal erosion, as compared with low-probability events such as extremely large floods or earthquakes, France says. “They started putting more money into those higher risks,” he notes. In this way, RIDM “helped them to prioritize where the (dam safety) money was going to get used.”
Generally supportive
Overall, the members of the independent panel “generally were supportive of what the agencies were doing” regarding assessing dam safety risks, France says. The Corps, Reclamation, and FERC “were using reasonable approaches,” he says. “They were consistent with the federal guidelines that were out there and how to apply risk analysis to dams.”
That said, the “levels of maturity in the application of RIDM differ among the three agencies and moving forward with the risk programs there are issues that warrant attention in all three agencies,” according to the report.
Among its recommendations, the panel encouraged FERC to “continue its movement toward the application of RIDM in regulating the dams under its purview,” according to the report. “FERC will need to further develop RIDM staff capability and staff resources within its organization and develop or support development of increased RIDM capacity in the private sector.” Similarly, the panel recommended that FERC “explore opportunities to provide enhanced risk assessment competency among dam safety practitioners through” collaboration with dam industry groups.
The three agencies would do well to learn more about how other industries assess and reduce risk, the panel says. The agencies should “evaluate more formalized safety case approaches using ‘prevention, mitigation, and control’ methods employed in other hazardous industries, to help ensure all reasonably practicable measures have been taken to reduce risk for all potential failure mechanisms, including surveillance, maintenance and other activities that control dam safety,” according to the report.
The panel’s assessment largely confirms that the three agencies are moving in the right direction regarding risk assessment of dams, according to a Corps official. “Overall, the review validated our efforts with some room for improvement,” said Travis Tutka, the chief of the Corps’ Headquarters Dam and Levee Branch, in the June 7 release. “By maintaining openness and transparency of our programs with the public through reviews such as this, we hope to increase trust in our management of USACE dams on behalf of the nation.”
This article is published by Civil Engineering Online.
