Close 

Dam Safety: Room for Improvement

While dams provide tremendous benefits ranging from flood control to power generation, they also represent a risk to public safety. Some of North America's leading dam safety experts sat down with Hydro Review to discuss efforts to improve dam safety.

By Russell W. Ray

The 2005 breach of Taum Sauk reservoir, a man-made lake that feeds a pumped storage hydro plant in eastern Missouri, and other recent dam failures have led the dam safety community to take a hard look at dam safety programs and the techniques being used to identify deficiencies that could lead to a failure.

Of the more than 80,000 dams in the U.S., about a third pose a "high" or "significant" hazard to life and property if a failure occurs, according to the Federal Emergency Management Agency (FEMA).

Hydro Review magazine Senior Associate Editor Russell W. Ray moderated a roundtable discussion involving some of North America's leading dam safety experts. The discussion centered on new inspection and monitoring methods, the state of financing for dam repairs, the cost of better inspection methods, and the effects on service and product suppliers.

The participants were: Brian Becker, chief of the Bureau of Reclamation's Dam Safety Program; Charles Pearre, dam safety program manager, U.S. Army Corps of Engineers; Dan Mahoney, director of the Federal Energy Regulatory Commission's Division of Dam Safety and Inspections; Constantine "Gus" Tjoumas, technology coordinator for Dam Safety, CEATI International; Paul C. Rizzo, president of Paul C. Rizzo Associates Inc.; and Warren Witt, manager of hydro operations, AmerenUE.

Brian Becker, U.S. Bureau of ReclamationCharles Pearre, U.S. Corps of EngineersDan Mahoney, FERC
Constantine "Gus" Tjoumas, CEATI InternationalPaul C. Rizzo, Paul C. Rizzo Associates Inc.Warren Witt, AmerenUE

What follows is a transcript of that discussion.

Q. How is the dam safety community improving the efficiency and effectiveness of dam safety programs in North America?

Paul: The dam safety community has made a giant leap in its efficiency and effectiveness, with the implementation of Potential Failure Mode Analysis (PFMA), developed originally by the Federal Energy Regulatory Commission (FERC). The program has allowed a large number of people to understand the importance of dam safety, including operators, engineers and managers. It's been a tremendous help in doing dam inspections, dam safety analysis and so forth. I'll go so far as to say that if we had done a PFMA at Taum Sauk in 2003 or 2004, the Taum Sauk failure might have been prevented.

Gus: To echo what Paul is saying, with tools like the PFMAs, we're learning to be more proactive than reactive and understanding what that means. With respect to dam safety, you just don't look at one little piece of the dam. You look at how the dam operates overall, with all of its components. We haven't always done a very good job of that before. We zeroed in on specific things. Now, we understand that we have to look at dam safety on an overall project basis and not just the embankment, the powerhouse or the spillway.

Dan: It's really only been recently - the last couple of years - that the dam safety industry has made the connection between the importance of a good owner's dam safety program and preventing dam failures. All it took was a review of the historical dam failures. Just about every one in recent history had an element of a breakdown in a dam safety program. We really haven't gotten the word out to state dam safety regulators about how much an effective dam safety program can reduce the risk of dam safety incidents.

Brian: I think the dam safety community is doing a better job of communicating by sharing information and common best practices and working together to develop more training opportunities. There's a lot of training available. There's really been an emphasis among dam safety owners to implement and develop emergency action plans. There certainly is a lot of interest in risk management. There's a lot of training and collaboration.

Warren: I'm not a dam safety specialist for our company. I operate and maintain three hydropower plants. So I'm a user of dam safety programs. To me, there appears to be a lot more emphasis being placed on dam safety and a lot more coordination among dam safety entities. As a user, that helps our ability to implement these and to understand what the expectations are. With the increased communications, we all have a better understanding of what the expectations are.

Charles: Five years ago, the Corps moved into more of a risk-informed technology to work our dam safety program. We have completed a portfolio risk analysis and initial screening of all our 650 dams and are developing a set of tools that can be used by different dam owners to review the risks their dams have and the possibilities of failures. The Corps, the Bureau of Reclamation and the Federal Energy Regulatory Commission have been working together to develop these tools and also to improve the Potential Failure Mode Analysis methods. I think all these are bringing us to a point where we are identifying where the greatest risks are.

Q. How do these changes affect service and product suppliers?

Brian: There is enormous potential for service providers to develop expertise in Potential Failure Mode Analysis and risk assessment. I think there is significant demand in the industry to support that.

Paul: All of the dam safety programs of significant quality have increased surveillance and monitoring equipment and services. That part of the industry has definitely increased and improved as a consequence of the upswing in dam safety programs and the results of PFMAs.

Gus: As potential failure modes are identified at each dam, understanding has improved on what needs to be monitored and what needs to be instrumented. Perhaps there are instruments that are no longer necessary. Then again, it might lead to the design of new types of instruments. Hopefully, they'll come up with improvements to instruments that are available to measure the things we need to measure. As examples, zones of seepage and piping are always difficult to ascertain.

Warren: If you are a service or product supplier, you will have more certainty in what will be accepted. That will allow you as a supplier to go out to other customers and say this has been an accepted process.

Charles: We've been able to put more money on some dam safety fixes. We realized that there's a maximum capability of the construction contractors and the equipment they own in order to do foundation fixes. We're keeping them busy.

Q. What new tools are available for diagnosing dam behavior and preventing problems?

Dan: The Potential Failure Mode Analysis is a big step forward in diagnosing dam behaviors and preventing problems.

Charles: A number of the things that we're finding today are problems that occurred during design and construction 40 and 50 years ago. We're able to take more instrumentation readings. There's automation of the instrumentation where you can get readings automatically and set programs up to tell you if a trend is occurring. This is especially useful for owners of multiple dams.

Warren: Automation allows you to do real-time, continuous monitoring instead of spot monitoring.

Q. What are the expected financial and staffing implications of the applications for these new tools and methods?

Dan: I think the PFMA can have a beneficial impact on financial and staffing implications because of what the PFMA does. In general, there are 100 ways dams can fail. But for a specific dam, a lot of those ways can be ruled out. That's what the PFMA does. It looks at the specific dam and it really determines how this particular dam can fail. It can make the monitoring program more efficient and less costly. It also could reveal a potential failure mode that you didn't realize, which would require additional resources and additional monitoring.

Brian: I think there are significant financial and staffing implications. I think we're right on the crest of the wave of moving forward with PFMA and risk management in the dam safety community. The potential extends beyond dam safety to any other infrastructure. It behooves anybody to develop a good understanding of PFMA and risk management.

Paul: Before the PFMA protocol became a requirement of the FERC and other nonregulated dam owners, the traditional dam inspection had moved toward becoming a commodity. Who could do it the cheapest and on the fastest schedule? It wound up being something that would be done by a junior engineer at the lowest cost. Inserting the PFMA into the process returned it back to its original intent: A thorough inspection thoroughly thought out by experienced people. Cost became less of a factor, although it became a more expensive process in the end.

Gus: Someone said earlier that it's going to cost more in the beginning, but in succeeding years, it's going to be less. Well, it could be, and in certain cases, it may not be. That still hasn't been borne out. Certainly the cost and efforts required for those dams identified to have potential problems identified through new tools and advanced methods will increase, but these costs will undoubtedly be less than costs associated with a dam failure.

Warren: It takes financing and staffing to implement these tools and methods. But once implemented, because you've automated some things, your staffing may be able to decrease or certainly not increase because you've got computers helping you do some of what you used to do manually. I haven't talked to many end users that have actually been able to reduce staff because of this. I think most people are still in the stage of increased engineering and expenditures, trying to put a lot of this stuff in.

Charles: It's been very hard to maintain a growing operations and maintenance budget that meets the requirements of inflation. Therefore, we have been looking at reducing staff and automation and instrumentation are ways we are looking at to do this. It does cost a little bit to put it in. In the end, you get the information from the instruments much easier, quicker and the computers are helping you analyze the data. Therefore, the cost of analyzing the data goes down. However, with instrumentation and automation, you've put something on the structure and it's subject to vandalism and wear and tear. So you're going to have maintenance that you might not have had in the past. Your financial costs are probably going to be about the same.

Q. How are risk assessment methods being applied at dams in North America?

Brian: We do risk assessments at all levels, at the very rudimentary screening level and at the comprehensive facility review level. As we're taking a more detailed look, we perform another risk assessment to ensure the decisions we are making are the right decisions. We're using them significantly and at various levels. We do risk assessments to verify that proposed modifications are going to satisfy the intent. Even after we've modified a project, we continue to perform risk assessments on a periodic basis.

Paul: I think risk assessments, at least in North America, are moving more and more toward probabilistic approaches to risk assessment. In the earlier days, we began looking at probabilistic hydrology issues. Nowadays, our seismic hazards are determined by probabilistic seismic hazard analyses. One of these days, we may put all of this together in some sort of logic tree and come up with probabilistic failure of the dam by overtopping, by seismic, by liquefaction and a number of other different factors.

Brian: I think that's one of the intents. To be able to compare, as Paul was saying, those specific failure modes relative to one another on a rational basis. You're putting the potential of seepage failure on the same relative scale as you would a seismic failure or a hydrologic failure.

Gus: There is much more work ahead of us with respect to the probabilistic approach. We need better data and information to do a better job at that. There is still data to be collected and case histories to be reviewed to help in this. That's why it is so important to have a good database of case histories. The Bureau of Reclamation is using risk assessments and analyses for a good risk-informed decision making process. But it's not done alone. We still do our deterministic approaches. We still do our traditional analyses. It's another tool we use to make a better risk-informed decision.

Dan: I think that's an important point to make. In the short time that risk assessment has been used in the U.S., it has gone from risk assessment answers to risk-informed decision making. The risk assessment is used as another piece of information along side the deterministic analyses that the industry has traditionally done.

Warren: It really helps in prioritizing. There's only so much expertise and contractors available out there. So you've really got to prioritize. It's helping the industry prioritize what are the most risk-significant dams and dam problems and putting your efforts toward those first.

Charles: We've been able to use risk assessment methods to level the playing field for our dams so that we are getting to the most critical ones first. It's a national priority and it's getting us additional budgeting priority from Congress. We are able to tell them what risks we are reducing. When you can define the risk and explain what the risk is, they become more willing to provide funding for you.

Q. What are the chief causes of dam failures in North America?

Paul: Overtopping.

Brian: Overtopping, seepage.

Dan: Statistically, I thought it was seepage.

Brian: I think overtopping is the highest.

Gus: I think overtopping is the high and seepage and piping is probably second.

Paul: Overtopping is something that's avoidable. You can perform the necessary calculations and develop a "fix" to prevent it.

Gus: Overtopping might occur because the spillway was inoperable. That's the type of thing that causes dams to overtop in general, other than extreme floods.

Warren: The chief causes are more programmatic in that whatever technically caused it, it probably had something to do with a lack of maintenance or lack of an oversight program. Some of that gets back into funding.

Charles: I agree that lack of maintenance has been one of the main causes of failures. Most of the failures on privately owned and publicly owned dams have been due to foundation problems. Lack of maintenance and lack of continual monitoring to catch things at an early stage allowed things to proceed to the point that the dam failed.

Warren: We had a dam failure a few years ago, a big dam, Taum Sauk reservoir. One of my concerns in looking at the root causes and the responses to that was that a lot of dam safety experts are really focused on the civil engineering aspects of how to build or maintain the dam. A lot of our oversight really focuses on the civil engineering, technical aspect. I think our dam safety programs need to look much more programmatic at what programs do you have in place to train your staff. What finances do they have available? Do they have adequate staff to do their jobs? Do they have a formal formatted surveillance monitoring program?

Q. Can you describe the state of financing for maintenance, upgrade and repair at non-federal dams?

Dan: Of the 75,000 dams that are privately owned, there is a lack of resources. Right now, the law is that the dam owner has to pay to fix it, and the dam owners, a lot of times, don't have the financing.

Gus: Several states do not have funds available for this. New Jersey has a low-interest loan program. As an example, I think there is even some grant money given out under certain circumstances.

Warren: Because we had a dam failure in the not too distant past, our state of financing has been extremely good. We have done a lot of work, a lot of maintenance, a lot of upgrade at our hydro facilities. As we come to the end of doing those upgrades and things become more standard, our financing will get a little more difficult to maintain because there will be other things that become more critical and we'll have to fight to keep our financing and manpower resources available.

Charles: I'd say the state is terrible. The locals do not seem to be willing to put the funds into repairing the dams. Only when they get to near failure do they start looking for funding. For the most part, most of them do not have the funds to fix the dam.

Q. The National Dam Safety and Security Act was passed in 2002. The measure was designed to help states improve their dam safety programs, increase training for dam safety engineers, and boost funding for dam safety research. How would you assess the law's impact on dam safety thus far in the U.S.?

Brian: It seems like the act has had a positive effect to improve dam safety programs and to increase and provide training opportunities. I think a potential shortfall is related to the governance of the National Dam Safety Review Board. Could it be reconsidered, where would it be most effective and more appropriate to lead that organization from, and how that program can be more effective. I think the National Dam Safety and Security Act was well intended and successful. But I think with a good introspective review and some strategic thinking, that group can become more effective.

Dan: I would add, though, that with respect to the resources that the state dam safety programs need to really have adequate programs, the funds that the National Dam Safety program is providing doesn't come close to filling the gap. There is some improvement. There's increased training and a lot of coordination and collaboration, but the resource deficit is still very much there.

Paul: I think Brian is being too polite. I think the program was well intended, but I don't see much in the way of benefits. There's just not enough money.

Gus: There's been some help to the states, especially those that have little money. It has been a benefit to the states with respect to being able to get out and do a few more inspections. The big thing is the training aspect that it provides to state dam safety engineers.

Charles: The Corps is a member of the National Dam Safety Review Board. This program has provided the states more money and I believe they are moving forward on increasing their training and boosting the amount of money that's going into dam safety research. Only in the last two years has the Federal Emergency Management Agency been able to get an increase in the amount of funding to the states. I think it is starting to help and it will help in the future.


Russell Ray is senior associate editor of Hydro Review

 

More Hydro Review Current Issue Articles
More Hydro Review Archives Issue Articles


To access this Article, go to:
http://www.hydroworld.com/content/hydro/en/articles/hr/print/volume-29/issue-6/cover-story/dam-safety-room-for-improvement.html