Reducing flood damages is a complex task that requires multidisciplinary understanding of the earth sciences and civil engineering. In addressing this task, the U.S. Army Corps of Engineers employs its expertise in hydrology, hydraulics, and geotechnical and structural engineering. Dams, levees, and other river works must be sized to local conditions; geotechnical theories and applications help ensure that structures will safely withstand potential hydraulic and seismic forces; and economic considerations must be balanced to ensure that reductions in flood damages are commensurate with project costs and associated impacts on social, economic, and environmental values.
Many flood damage reduction projects involve the construction of levees. The Corps's historical approach to coping with hydrologic and hydraulic uncertainties of large floods was based on a best estimate of the levee height required to withstand a given flood, which was then augmented by a standard increment of levee height called “freeboard. ” The best estimate has traditionally been based on the expected height of a design flood (e.g., a 100-year flood, the magnitude of which has a 1 percent chance of being equaled or exceeded in any given year, and which is here called the “1% flood”). Freeboard was then added above the expected height. Many Corps flood damage reduction projects used a standard of 3 feet of freeboard. “Three feet of freeboard” became an engineering tradition within the Corps and was employed in hundreds of Corps flood damage reduction studies and projects.
Challenges to the concept of a standard levee freeboard emerged in the early 1990s. For instance, it was noted that a standard freeboard did
not account for geographic and hydrologic differences at different locations and may thus have provided different levels of flood protection in different localities. Procedures for calculating the economic benefits conferred by levee freeboard were also questioned.
The Corps felt that development and application of risk analysis techniques held great promise in addressing these issues, as these techniques aim to quantify and explicitly incorporate uncertainties in hydrologic, hydraulic, and geotechnical parameters into levee design analysis. It was envisioned that proper application of risk analysis could replace the need for a standard 3 feet of freeboard.
Risk analysis also became part of a federal levee certification procedure jointly conducted by the Corps and the Federal Emergency Management Agency (FEMA). Within the National Flood Insurance Program, FEMA identifies areas subject to varying degrees of flood risk on flood insurance rate maps. One of these areas is the Special Flood Hazard Area (SFHA), defined as the area that is inundated by a flood having a 1 percent chance of being equaled or exceeded in any given year (the 1% flood). Property within a Special Flood Hazard Area is subject to mandatory flood insurance purchase requirements and may be subject to local land use regulations, as well.
Floodplain property can avoid the Special Flood Hazard Area designation, and the mandatory flood insurance requirements that attend it, if it is protected by a levee certified to provide protection against the 1% flood. The Corps of Engineers is responsible for certifying levees as meeting this safety standard. As levee certification could exempt a community from flood insurance purchase requirements (and possible exemptions from local land use requirements), this certification procedure has great local economic and public policy significance.
The historical standard for levee certification had been that levees must provide protection to the average stage (height) of the 1% flood, plus 3 feet of freeboard. With the Corps's adoption of risk analysis techniques in the early 1990s, the freeboard standard for levee certification was abandoned in favor of the new risk analysis standard.
The public, however, was not entirely comfortable with the replacement of a time-tested standard by relatively new techniques. These issues came to a head in a Corps flood damage reduction project planning study in Portage, Wisconsin in the early 1990s. The Corps study recommended a levee of elevation 798.3 feet for the city of Portage. But this recommended levee (the “National Economic Development” levee project alternative) would not have been high enough to be certified as providing protection from the 1% flood. Because the calculations for
levee height were based on the new risk analysis techniques, in 1993 the city of Portage, the Wisconsin Department of Natural Resources, and the Association of State Floodplain Managers challenged the study 's results. An outcome of the ensuing discussions was that the U.S. Congress requested a National Academy of Sciences study of the Corps 's use of risk analysis techniques. (The National Academy of Sciences was subsequently subsumed—in 1999—as part of the National Academies. The National Academies includes the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine.)
The charge to the National Academies was included in the Water Resources Development Act of 1996 (WRDA 96) from the 104th Congress of the United States. Public Law 104-303 of WRDA 96 stated the following (Section 202h):
The Secretary (Army) shall enter into an agreement with the National Academy of Sciences to conduct a study of the Corps of Engineers ' use of risk-based analysis for the evaluation of hydrology, hydraulics, and economics in flood damage reduction studies. The study shall include—
an evaluation of the impact of risk-based analysis on project formulation, project economic justification, and minimum engineering and safety standards; and
a review of studies conducted using risk-based analysis to determine —
the scientific validity of applying risk-based analysis in these studies; and
the impact of using risk-based analysis as it relates to current policy and procedures of the Corps of Engineers.
To carry out this assignment, the Water Science and Technology Board (WSTB) of the National Academies's National Research Council (NRC) appointed the Committee on Risk-Based Analyses for Flood Damage Reduction to conduct the study, with sponsorship provided by the U.S. Army Corps of Engineers.
APPLICATION OF RISK ANALYSIS TECHNIQUES
Recognition of engineering and economic uncertainties and their ex-
plicit quantification in flood damage reduction studies leads to projects that are better tailored to local conditions and available data than did the earlier, deterministic levee freeboard standard. The new techniques are a significant step forward and the Corps should be greatly commended for embracing contemporary, but complicated, techniques and for departing from a traditional approach that has been overtaken by modern scientific advances. While some technical issues are not yet fully resolved, the Corps has taken a significant step forward with the development of its risk analysis methods for flood damage reduction studies. The committee also notes that these advances have been made with a relatively modest investment of resources. There should be no turning back from this accomplishment.
The former approach of using standard levee freeboard did not provide consistent levels of flood protection across the nation. A consistent protection standard must properly account for local and regional differences in topography, hydrology, and hydraulics, which the standard freeboard approach did not. In some areas, for instance, as little as 2 feet of freeboard may be required to provide adequate flood protection, while in other areas, as much as 6 feet may be required. The traditional 3 feet of freeboard standard masks a significant degree of variation of risk of levee failure for citizens protected by these levees. This variation in risk of failure can be quantified by the Corps's new risk analysis procedure.
The committee divided its recommendations for improvement into the following areas: (1) refine methods relating to probabilistic and statistical modeling of floods, performance of flood damage reduction systems (e.g., levees), and flood damage assessment, (2) adopt a consistent terminology for communicating risk analysis concepts within the Corps and to the public, (3) simplify and improve the complex and somewhat confusing criteria for certifying levees for inclusion in the National Flood Insurance Program (NFIP), and (4) move toward a more comprehensive decision making approach in flood damage reduction studies.
Risk Measures and Modeling
The committee reviewed a computer program developed by the Corps's Hydrologic Engineering Center (HEC) in Davis, California. This computer program, the Hydrologic Engineering Center Flood Damage Assessment (HEC-FDA) program, is the principal tool used in Corps district offices to calculate flood damage risks. This program implements the Corps's risk analysis and builds upon a deterministic approach
to flood damage estimation that evolved over several decades. Although it benefits from this experience, the Corps's risk analysis method suffers from the difficulty of translating deterministic practice into a risk analysis application. Assessing the risk analysis method 's validity rests on the following questions:
Are the performance measures generated by the risk analysis method useful and complete?
Are all the important uncertainties included in the analysis?
Is the specification of these uncertainties proper?
Are probabilistic and statistical methods used correctly?
Risk analysis is applied to economic performance measures (project net benefits and benefit–cost ratio) and to engineering performance measures (probability of flooding). While the measures of economic performance in the new method are generally practical and informative, there are too many types of engineering performance measures to be clearly understood by most citizens. Standardization to one or two key measures of engineering performance measures would represent an improvement.
The committee recommends that the Corps use annual exceedance probability as the performance measure of engineering risk. This is a measure of the likelihood that people will be flooded (including the probability of failure of flood damage reduction structures, such as levees) in any given year, considering the full range of floods that can occur and all sources of uncertainty.
For engineering purposes, it is useful to calculate other system reliability measures, such as the conditional nonexceedance probability for the 1 percent (100-year) flood. But such measures are difficult to understand and are not as clear as the measure of annual probability of flooding, and they should not be used in communicating flood risks to the public.
Evaluation of the uncertainty in economic benefits that are attributed to knowledge uncertainties represents an important advance for the Corps. Such an evaluation is performed using a Monte Carlo procedure that evaluates expected annual damages using different possible parameter combinations for hydrologic, hydraulic, geotechnical, and economic models. While the Corps's conceptual approach to modeling flood hazard and associated damages—using relationships between flood frequency, stage–discharge, and damage–stage—is consistent with long-
standing scientific understanding, certain improvements to this method are needed. Risk analysis measures for a project must rest upon complete and accurate specification of the uncertainties in each component of an analysis and upon correct probabilistic methods to quantify and combine those uncertainties. As the current method has shortcomings in these areas, the committee recommends that the Corps improve its analysis of hydrologic, hydraulic, geotechnical, and economic uncertainties.
The Corps's conceptual approach of distinguishing between natural variability and knowledge uncertainty is reflected in the engineering modeling components of its risk analysis method, but the conceptual approach needs refinement. Natural variability is variability assumed to be inherent in natural processes, such as flood frequencies or properties of geotechnical materials. Knowledge uncertainty is attributed to limitations of scientific understanding of natural processes. In some cases, such as the hydraulic relationship between a river's stage (height) and its discharge, the risk analysis method appears to include natural variability under knowledge uncertainties; in others, such as geotechnical levee performance, knowledge uncertainties appear to be included under natural variability. This is a critical issue, because knowledge uncertainties and natural variability each affect the calculations of risk in different ways.
The committee recommends that the Corps focus greater attention on the probabilistic issues of identifying, estimating, and combining uncertainties. Better specification of knowledge uncertainties in flood frequencies is needed. The uncertainty in the skewness coefficient for log-Pearson Type III distribution models of flood frequency (which are used by the Corps and other federal agencies for describing return periods of floods) should be explicitly included in the risk analysis. Some measure of the uncertainty inherent in computing flood–frequency curves from rainfall–runoff modeling is also needed.
The committee recommends that the Corps strive to reduce the considerable variation in the estimates of water surface elevation when using different models of river hydraulics. The Corps's experiences in applying alternative methods to estimate flood stage indicate that there can be substantial differences in the results.
The committee recommends that the Corps's risk analysis method evaluate the performance of a levee as a spatially distributed system. Geotechnical evaluation of a levee, which may be many miles long, should account for the potential of failure at any point along the levee during a flood. Such an analysis should consider multiple modes of levee failure (e.g., overtopping, embankment instability), correlation
of embankment and foundation properties, hazards associated with flood stage (e.g., debris, waves, flood duration) and the potential for multiple levee section failures during a flood. The current procedure treats a levee within each damage reach as independent and distinct from one reach to the next. Further, within a reach, the analysis focuses on the portion of each levee that is most likely to fail. This does not provide a sufficient analysis of the performance of the entire levee. This has important implications for not only geotechnical and economic analysis of flood damages, but also for levee certification.
The Corps's new geotechnical reliability model would benefit greatly from field validation. The nation has many years of experience with levee performance and, unfortunately, also with levee failures. Much of this experience is documented and much is accessible to federal agencies. The committee recommends that the Corps undertake statistical ex post studies to compare predictions of geotechnical levee failure probabilities made by the reliability model against frequencies of actual levee failures during floods. In addition, the committee recommends that the Corps conduct statistical ex post studies with respect to the performance of other flood damage reduction structures (e.g., embankments, detention basins, hydraulic facilities). These latter studies should be conducted in order to identify the vulnerabilities (failure modes) of these systems and to verify engineering reliability models.
In the current Corps method (and as mandated by the federal Principles and Guidelines), flood damage is calculated for each set of project alternatives by aggregating over all existing structures (buildings) in the floodplain. Then, reduction in flood damage is calculated by taking the difference between this aggregate number and the corresponding aggregate damage without the project. Correlations among the random variables can introduce serious errors in the analysis. Each structure in a floodplain is modeled as if that structure exists in isolation from all others. The result is that the analysis incorrectly computes uncertainties associated with differences in economic damages that result from different project alternatives.
The committee recommends that the Corps calculate the risks associated with flooding, and the benefits of a flood damage reduc-
tion project, structure by structure, rather than conducting risk analysis on damage aggregated over groups of structures in damage reaches. Furthermore, the practice of summing and subtracting percentile values of probability distributions of flood damage in reaches to obtain risk measures of project economic performance is unsound and produces output measures of unknown accuracy. The outputs of the economic risk analysis using the current procedure are thus of questionable value.
The committee noted that a variety of terms describing aspects of risk and uncertainty are often used interchangeably within and between the Corps's water resources programs. The committee thus recommends that the Corps adopt a consistent vocabulary for describing risk analysis concepts, specifically distinguishing between risk, natural variability, knowledge uncertainty, and measures of system reliability. The Corps should clearly distinguish between natural variability (based on the random nature of physical systems) and knowledge uncertainty (uncertainties attributable to limitations in the current state of knowledge).
In the early 1990s the Corps and FEMA adopted a risk analysis approach to replace the practice of certifying levees that had 3 feet of freeboard above the 1% flood level. This risk analysis approach and the levels of flood protection it provided were controversial. Negotiations between the Corps and FEMA led to the current practice of certifying a levee based on a three-tiered decision rule, using: (1) 3 feet of freeboard, (2) a conditional nonexceedance probability of 90 percent of passing a 1% flood, or (3) a conditional nonexceedance probability of 95 percent of passing a 1% flood. Although this three-tiered criterion represents a reasonable transition from the former certification criterion into the risk analysis framework, it has the following deficiencies: (1) it still leads to different levels of flood protection for different projects, (2) the three-tiered decision rule is unnecessarily complicated, (3) the method evaluates levees individually rather than as a levee system that is intended to provide flood protection for a community, and (4) certification is incomplete in that it considers only the 100-year flood, not the full range of
The committee recommends that the federal levee certification program focus not upon some level of assurance of passing the 100-year flood, but rather upon “annual exceedance probability”—the probability that an area protected by a levee system will be flooded by any potential flood. This annual exceedance probability of flooding should include uncertainties derived from both natural variability and knowledge uncertainty.
The criterion for certifying a levee should be that it provides satisfactory protection against failure of the flood damage reduction system, expressed as an annual probability of flooding. This new criterion should promote better communication among the Corps, FEMA, other regulatory and expert groups, and communities and local cosponsors.
Substantial resources and time may be required to implement the annual exceedance probability approach for certifying a levee. Until the measure of annual exceedance probability is adopted as the key criterion for levee certification, the committee recommends that the Corps and FEMA set a single conditional nonexceedance probability for levee certification.
The former certification criterion was flawed in that it produced vastly different levels of flood protection for different communities. The committee recommends that the certification criterion provide a uniform level of flood protection. Which level of protection to choose is not obvious. Insisting on the highest level of protection would mean that only a small proportion of levees would be certified. In the committee's judgment, the certification criterion should be the level of protection provided to most people in the past—the median level historically provided. Based upon a small sample of all Corps flood damage reduction projects, the committee found that the median annual exceedance probability of Corps flood damage reduction projects is approximately 1/230.
This is the committee's best estimate of the median annual exceedance probability. To obtain a more reliable measure of the median annual exceedance probability of approved projects, the committee recommends that the Corps examine a larger number of flood damage reduction projects and audit the process of estimating the annual exceedance probability for these projects.
The committee recommends that the Corps develop a table showing percentiles of variability in the annual exceedance probability of its flood damage reduction projects. By choosing an appro-
priate percentile value in this range, a corresponding level of assurance can be obtained that the expected level of protection is at least 100 years, as required. It was the lack of allowance for this variability that led to the abandonment of the annual exceedance probability criterion during the 1990s.
Neither the U.S. Congress nor the Corps of Engineers have defined an explicit goal for management of the nation's floodplains. In the committee's opinion, the goal of floodplain management should be to use the land for the greatest social benefit. Broadening the scope of the Corps's risk analysis and expanding the types of alternatives considered would provide more useful insight about how best to achieve this goal.
As currently specified by the federal Economic and Environmental Principles and Guidelines for Water and Related Land Resources Implementation Studies, flood damage reduction studies emphasize direct economic damage reductions and the costs of alternatives; these are quantified in the Corps's risk analysis methodology. To ensure that the Corps's flood damage reduction projects provide adequate social and environmental benefits, the committee recommends that the Corps explicitly address potential loss of life, other social consequences, and environmental consequences in its risk analysis. Furthermore, the Corps's risk analysis should not be limited to structural alternatives such as levees, dikes, and dams. Nonstructural alternatives such as warning systems and zoning regulations should also be considered, both separately and in conjunction with structural alternatives.
Given the breadth of federal agencies and programs devoted to U.S. floodplain and flood hazard management, the Corps clearly cannot implement these recommendations alone. Further, it is not likely that such a broadening of the Corps's risk analysis methods will occur over a short period of time. To include a broader range of social and environmental implications in the benefit–cost calculations of flood damage reduction studies, appropriate revisions of existing legislation and planning guidance, consistent with these recommendations, may have to be enacted by the U.S. Congress.
To appropriately include such consequences and their relative importance, the committee recommends that the ecological, health, and other social effects of Corps flood damage reduction studies, and the tradeoffs between them, be quantified to the extent possible and
included in the National Economic Development Plan. More explicit efforts at including these types of consequences and values in the Corps's benefit–cost calculations should result in increased social benefits of the Corps's flood damage reduction studies. The Corps should seek guidance from the Office of Management and Budget and seek consistency with other federal agencies on the use of alternative metrics for incorporating potential loss of life, environmental impacts, and other effects of floods.