Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Summary There is no doubt that Sacramento, California, and the surrounding metro- politan area face a significant flood risk from both the Sacramento River and the American River, which converge at the city's doorstep. More than 400,000 people and $37 billion worth of damageable property are vulnerable to flooding in the Sacramento area, including most of the city's central business district and the State Capitol complex. Although there is consensus that action is needed to reduce the level of risk while allowing reasonable use of floodplains, agreement on the appropriate target level and the approach to achieve it has eluded national, state, regional, and local decisionmakers. In 1991 the Sacramento District of the U.S. Army Corps of Engineers completed a study, the American River Watershed Investigation (ARWI), that reviewed the American River's contribution to the area's flood hazard, consid- ered a range of flood control measures, and recommended a preferred flood control strategy (USAGE, Sacramento District, 1991~. The effort met significant criticism, some of it highly technical and some of it political. As a result, Congress directed the Sacramento District to reevaluate its analysis and gather additional input. In response to this congressional directive, additional study and lUSACE (U.S. Army Corps of Engineers) as used herein refers to actions taken by the Washington D.C. headquarters of the Corps of Engineers (e.g., agency-wide policies, procedures, etc.) or com- ments by the headquarters on subordinate activities by subordinate elements such as the Sacramento District. Field activities, reports, work in progress, meetings, etc. by the Sacramento District are identified as the Sacramento District unless and until specifically acted on by USACE.
2 FLOOD RISK MANAGEMENT AND THE AMERICAN RIVER BASIN planning have been done by local, state, and federal interests. These efforts- which continue as this report goes to press-have yielded a more comprehensive picture of the flood risk and a broader array of possible flood risk reduction alternatives. The information available today is more comprehensive and more detailed than that available in 1991. But the fundamental dilemma remains unresolved: how do we balance the potential benefits, impacts, costs, and trade-offs associated with the identified alternatives and select the best management plan for the basin and its residents? This final decision lies not in the realm of science and engineering, but in the arena of public decisionmaking. It requires participants to set aside differences and seek commonalities. It requires weighing competing values. In the end, it will require leadership from local governments, the state of California, the U.S. Army Corps of Engineers, and Congress, as well as a sincere effort by the region's interest groups to agree among themselves about how to respond to the flood hazard. THE COMMITTEE'S CHARGE At the same time that Congress asked for a reevaluation of the potential flood control alternatives available to the Sacramento area, the nature of some of the criticisms caused members of Congress to seek an outside body to review the technical soundness of the analyses and related policy questions. Congress di- rected the Secretary of the Army to ask the National Academy of Engineering to form a special committee, the Committee on Flood Control Alternatives in the American River Basin, to review the 1991 ARWI, with attention to the contin- gency assumptions, hydrologic methods, and other engineering analyses used to support the seven flood control options presented. Significantly, the committee was not asked to recommend a preferred alternative; instead, it was asked to evaluate the scientific and engineering knowledge base on which the selection of a final strategy will ultimately be based. The committee also was asked to take a step back from the often acrimonious debate that has surrounded the American River planning process and provide insights of value to other regions in the nation that face similar problems other areas where cities have grown in flood- prone areas and now face significant flood risks. The massive Midwest floods of 1993 and significant regional floods in 1994 and 1995 are reminders of how serious this issue is for the nation. The committee's charge contains an inherent dilemma. Because there was great controversy surrounding the 1991 ARWI, the committee was asked specifi- cally to review that document. But the controversy surrounding that document was so great that Congress simultaneously asked the Sacramento District to re- vise it. As a result, while the committee was gathering information for its analy- sis, efforts to improve the 1991 report were being made by the Sacramento District, the Sacramento Area Flood Control Agency (SAFCA), the Reclamation
SUMMARY 3 Board of the State of California, and the State Department of Water Resources, among others. Thus in this report the committee comments on the data, analysis, and meth- odologies used in the 1991 ARWI where they are still germane. In addition, where possible, it reviews the new analyses and methodologies being used to reevaluate Sacramento's flood risks and assess alternative flood risk management strategies. This has proven to be a difficult task because the committee's study, the Sacramento District's ongoing efforts, and parallel work through SAFCA continue to move along in near synchrony. Also, at this point there is little written documentation of the new work. The majority of the information concerning the ongoing work was received informally. The committee spoke at length with technical staff from both the federal and the state agencies and tried to understand what methodologies and data were being employed in the current analysis. The committee also heard from a variety of interest groups. In 1994, a new document, Alternatives Report: American River Watershed, California (USAGE, Sacramento District, 1994a), reached the committee in time to be considered, but this interim report lacked detail. For a true reevaluation of the Sacramento District's technical analysis, Congress might wish to request a review of the upcoming Draft Supplemental Information Report and Environmental Documentation, expected to be available in the summer of 1995, because that document will update the 1991 ARWI in detail. THREE PREMISES As the committee conducted this review of the Sacramento District's plan- ning for flood control in the American River basin, it became clear that the members shared certain premises (i.e., assumptions believed to be true on the basis of experience and expertise) that influenced their thinking. These premises are (1) the belief that alleviation of Sacramento's flood risk is critical, (2) the belief that decisionmaking in the American River basin should not stand in isola- tion, but should be seen in light of national policy that stresses the use of multiple strategies to respond to flood hazards, and (3) the belief that technical matters cannot be neatly separated from policy judgments. These three premises are introduced here to provide a context for understanding the scope of the committee's review and the nature of this report's conclusions and recommenda- tions. These introductory ideas are followed by brief overviews of the chapters of the report. Alleviation of Sacramento's Flood Risk is Critical Actions to alleviate Sacramento's ongoing flood risk are urgently needed. The flood-prone development in Sacramento is intense and of high value. This
4 FLOOD RISK MANAGEMENT AND THE AMERICAN RIVER BASIN Sacramento, California, is a city that grew literally at the edge of the American River and it has been plagued by recurring floods as a result. More than 400,000 people and $37 billion worth of damageable property are vulnerable to flooding in the area, including most of the city's central business district and the State Capitol complex. (Robert Childs, U.S. Army Corps of Engineers.) situation occurred in response to historical influences and cannot now be re- versed. Nearly 10 years have elapsed since the city's existing flood defenses were clearly proven to be inadequate in the flood of February 1986. Although the careful analyses necessary to support decisionmaking take time, especially if the process is to allow adequate public participation, there comes a point where talk must turn into action. For a variety of reasons, the public decisionmaking process has been blocked from reaching consensus on a feasible course of action to provide the Sacramento area with a higher level of security. Paradoxically, efforts to enhance protection for the largely undeveloped floodplain of the Natomas Basin have progressed further in Congress and locally than proposals affecting developed areas, including downtown Sacramento and the State Capitol complex. Ultimately, California and the nation need to reexamine their ap- proaches to public decisionmaking. Widespread involvement by stakeholders and careful consideration of all options is of course necessary. But delay per se can be counterproductive, costly, and potentially dangerous.
SUMMARY National Flood Policies Urge Multiple Adjustments to Hazard Flood control in California cannot be treated in isolation but must be treated as a part of a complex system of water control and use that has evolved over a long period under the auspices of many government agencies and in response to significant pressures. As the committee approached the task of assessing flood risk along the lower American River, it was aware of recent laws and policy reviews that reflect a broadening of our nation's response to floods over the past quarter-century. For decades, the predominant response to flood risk was to build large flood control projects dams, reservoirs, levees, diversion channels to store and restrain floodwaters. The adoption of the National Flood Insurance Act of 1968 marked a watershed in national policy on flood hazards because it estab- lished nonstructural measures flood insurance, floodplain management, and selective acquisition as mainstays of national flood policy. Additional nonstructural measures in widespread use today include flood forecasting, evacu- ation planning, public education, and floodproofing of individual commercial and residential structures located in floodplains. More recently, the 1994 Unified National Program for Floodplain Manage- ment (FEMA, 1994) also called for a blend of strategies, from structural ap- proaches to modify flooding to restoration of floodplains. A major evaluation of the Midwest floods of 1993 prepared by the Interagency Floodplain Management Review Committee (IFMRC, 1994) at the direction of the White House, which calls for "shared responsibility" among all levels of government and private interests in responding to flood hazards, also strongly supports the use of nonstructural measures such as relocation of structures out of floodplains and restoration of wetlands, where feasible. As noted by the National Review Committee (1989~: The present status of floodplain management does not encourage complacency. The record is mixed. There are encouraging trends, as with the number of communities having some form of floodplain regulations, but the rising toll of average annual flood losses has not been stopped or reversed. Some activities look more productive on paper than on the ground or in the real vulnerability of people. On balance, progress has been far short of what is desirable or possible, or what was envisaged at times when the current policies and activities were . . . initiated. Thus planners and decisionmakers should proceed with caution. No single technical or institutional "fix" is likely to be an adequate response to the lower American River flood hazard. Responsible federal, state, regional, and local officials must seek to identify a combination of policies and measures that will maximize flood reduction benefits while minimizing economic and environmen- tal costs.
6 FLOOD RISK MANAGEMENT AND THE AMERICAN RIVER BASIN Technical Assessment Includes Policy Judgments The charge to the committee was based on the premise that many of the criticisms of the 1991 ARWI were matters of technical dispute and that a techni- cal judgment could be rendered about the merits of the critics' comments. Rep- resentatives of USACE, SAFCA, environmental groups, and Congress at differ- ent times emphasized that the committee should try to settle the technical debate, in order to let the political process make the public policy choices about the acceptable risk at Sacramento, including Natomas. However, the planning and design of a flood control program, although requiring complex modeling, engi- neering, and data manipulation, do not divide neatly into two parts, technical analysis and policy decisions. For example, even the most apparently technical computational concerns, such as what to assume about the likely coincidence of peak flows at the confluence of two rivers or about use of surcharge space, are based on a policy viewpoint about the acceptable risk of modeling error. IDENTIFICATION AND EVALUATION OF ALTERNATIVES As the committee conducted its review of the American River planning process, it noted that perhaps the most critical step in the development of a flood control strategy is the selection of alternatives for detailed analysis. The 1991 ARWI presented various alternative approaches to providing flood control for the American River basin, addressing level of protection provided, costs, expected benefits, and environmental impacts. The report was controversial, and some criticisms were based on the perceived failure of the Sacramento District to consider and evaluate a full range of effective alternatives, such as modification of the operation of Folsom Dam coupled with improvements in outlet capacity. In considering the issue of alternative flood control plans in both the 1991 ARWI and a more recent document, the 1994 Alternatives Report, the committee fo- cused on four issues: (1) use of Folsom Reservoir, (2) the question of gates should a dam be built at the Auburn site, (3) the viability of the Deer Creek alternative, and (4) the adequacy of the nonstructural measures presented. As detailed in Chapter 3, the committee concludes that the original 1991 ARWI was reasonably complete, especially as supplemented by the 1994 Alter- natives Report. One concern that arose involved the operating policies employed at Folsom Dam. However, ongoing investigations are now exploring the more dynamic use of Folsom storage capacity. Another concern is the fact that the committee was unable to evaluate how Folsom reoperation was actually consid- ered in the 1994 Alternatives Report, particularly what assumptions were used regarding the initial conditions. These concerns are expected to be addressed in upcoming documents; resolution of these questions should not slow the planning process. The committee notes that Folsom Reservoir, despite its limitations, is the
SUMMARY 7 critical component in the flood-control system for Sacramento. Consequently, it is essential that it be operated as efficiently as possible, and thus the soon-to-be released Folsom Flood Management Plan is critical. It is also important that the operation plan for Folsom evolve as necessary in response to changes in the American River system. Regarding possible construction of a dry dam at the Auburn site, the commit- tee notes that, should a dam be built, operational gates are essential for dam safety and to provide flexibility in the dam's operation, allowing operators to coordinate with Folsom and other flood control facilities, and to minimize environmental impacts in the upper American River canyon by regulating drawdown. ENVIRONMENTAL ISSUES A key issue in the controversy of how to provide flood hazard reduction to the American River basin is how to minimize environmental impacts. Environ- mental issues were at the heart of many of the disagreements that resulted from the 1991 ARWI. Among the most contentious were the question of the adequacy of the report in assessing potential environmental damage and the uncertainty surrounding impacts of a detention dam in Auburn canyon. Overall, the committee finds that from an environmental perspective the 1991 ARWI suffered from a lack of scientifically based descriptions of potential impacts and thus did not adequately support the decisionmaking process and help the public weigh the environmental impacts for the range of flood damage reduc- tion alternatives presented. The report understated some environmental impacts, particularly in the upper canyon. The 1994 Alternatives Report, subsequent research, and a report from the Lower American River Task Force (SAFCA, 1994b) show significant improvement in understanding impacts, consideration of options, and minimization of impacts. On the basis of the research to date, the major uncertainty is potential im- pacts on canyon slopes and vegetation from inundation behind an Auburn deten- tion dam. If such a dam is to be seriously considered, the committee recommends the formation of a multidisciplinary research team to design and carry out a program to reduce this scientific uncertainty and recommend a gate design and operating strategy that could be followed to minimize environmental impacts. RISK METHODOLOGY USACE has adopted new risk and uncertainty analysis procedures that are an extension of the traditional paradigm for flood control project planning and com- munity flood protection evaluation. The 1994 Alternatives Report indicates that the Sacramento Diskict's analysis now considers varying degrees of uncertainty in the causes of flooding, such as inflow to Folsom Reservoir, regulated outflow- frequency relationships for Folsom Dam, river stages, and levee stability. The
8 FLOOD RISK MANAGEMENT AND THE AMERICAN RIVER BASIN methodology computes the risk of flooding due to combinations of hydrologic events, hydrologic parameter uncertainty, uncertainty in reservoir operations, stage-discharge relations, and levee performance. USACE traditionally has in- cluded safety factors in its design of facilities and the specification of operating policies to address important hydraulic and operational uncertainties in flood control planning calculations; with its new risk and uncertainty analysis method- ology, one can investigate the extent to which such safety factors are economi- cally justified. The committee concludes that the USACE risk and uncertainty procedures are an important initiative. The explicit recognition of modeling uncertainty should result in a better understanding of the uncertainty of flood risk and damage reduction estimates. This change in methodology is important to the American River planning process because the ongoing evaluation of flood control alterna- tives for the basin is one of the first applications of the methodology. It is almost certainly the most complex application yet attempted by USACE. As discussed in Chapter 4, the new risk and uncertainty procedures, which directly include hydrologic uncertainties in the calculation of average flood risk and the average annual flood damages, tend to inflate those estimates. This tendency can yield benefit-cost calculations more favorable to project justifica- tion. The chapter suggests how risk, variability, and uncertainty in hydrologic, hydraulic, and economic processes should be conceptualized and how the calcu- lations can be organized to avoid introducing such biases while still communicat- ing residual risks and associated uncertainty. The committee also questions the value of the system reliability index com- puted by the Sacramento District in its American River study. The 1994 Alterna- tives Report was found to be particularly confusing because no distinction was made between estimates of flood risk calculated with the traditional level of protection and those calculated with the new risk and uncertainty procedures. Such distinctions are important. USACE needs to develop a consistent scientific methodology and an effective vocabulary for communication of residual flood risks and uncertainties to technical and public audiences. FLOOD RISK MANAGEMENT BEHIND LEVEES History shows a close relationship between flood protection and develop- ment in flood-prone areas. From the mid-1930s to the late 1960s federally subsidized flood control projects such as levees and upstream storage were the prevalent form of national response, but nevertheless flood losses continued to rise because of continuing development on floodplains. The reasons are many and complex: floodplains can appear to be desirable building locations, and the hazards sometimes are not seen or are unavoidable. Sometimes, development actually is encouraged by federal protection. Once a levee is built to protect development in a floodplain, for instance, it opens the way for additional devel
SUMMARY 9 opment, which in turn prompts demands for higher levels of protection. Such development can impose heavy burdens on society. Thus in this era of tightened budgets the question of who pays to support this "flood protection-development spiral" is becoming increasingly important. One question in the American River basin is whether this flood protection- development spiral is the fate of the Natomas Basin. The Natomas Basin is a flat, marshy lowland of about 55,000 acres near Sacramento that lies entirely within the 100-year floodplains of the American River and the Sacramento River. To- day the basin is surrounded by a 41-mile ring of levees and is devoted primarily to agriculture. The basin is now home to 35,000 people, but because of its prime location, it is projected to be a major growth area for new housing and commer- cial development. Although the existing levees lessen the flood risk to some degree, the Natomas Basin faces significant residual risk. The basin lies below the levels of the American and Sacramento rivers at flood stage and could fill like a bathtub in the event of a flood that breaches or overtops the levees. According to plans prepared by local authorities, large portions of the basin are poised for development despite the unresolved and perhaps unresolvable issue of its flood hazard. Clearly, the Natomas Basin is well situated in terms of proximity to Sacramento, but it is poorly situated in terms of chronic flood risk. Improvements in the existing flood protection system, including the reoperation of Folsom Dam, levee expansion, and other improvements that are in progress or are foreseeable, can help reduce the risk, but significant residual risk will remain. Development within the Natomas Basin thus should be subject to prudent flood- plain management requirements under federal, state, and local authority. In addition, the public should be informed of the residual flood risks despite the presence of the levee system. WATER RESOURCES PLANNING AND DECISIONMAKING The application of the USACE planning process to the search for acceptable flood control for the American River basin has illustrated the need for reforms in how such decisions are made, and the committee believes that the lessons of the American River can be transferred to other areas of the nation. Early decisions, such as Congress's direction to limit the project purpose to flood control, that were made ostensibly to lessen controversy and speed the process instead pro- longed the debate because public interests desired a wider view. Indeed, many who commented on the 1991 ARWI were critical of its failure to consider any purpose other than flood control as a planning purpose and noted that this single- purpose approach precluded a more integrated approach to planning. In particu- lar, the dispute over the proposed dry dam alternative has stalled the study pro- cess. Although progress is being made to mate environmental restoration concerns with improved levee stability and conveyance in the lower American River, in large part because of the work of the Lower American River Task Force
10 FLOOD RISK MANAGEMENT AND THE AMERICAN RIVER BASIN facilitated by SAFCA, efforts to resolve disputes over alternatives and impacts in the upper American River have met with little success. The current decisionmaking situation in the American River basin can be described as a diffusion of separate interests having access to numerous political and legal veto points, making it far easier to stop an activity than to move one forward. Despite some errors and problems with the planning process as imple- mented in the American River basin, the committee recognizes that USACE to date has been embroiled in larger California water controversies and at times technical complaints have been used as weapons in a policy dispute. The com- mittee believes that in the American River context and similar situations USACE must make its work part of a shared planning process where the local sponsor, other agencies of the federal and state governments, and nongovernmental inter- ests can cooperatively develop the data and models, understandings of risks and tradeoffs, formulation of alternatives, and consensus on the most appropriate alternative. The American River situation is not unusual; USACE has frequently seen its recommendations challenged in recent years and thus needs to find ways to improve the planning process so it works more effectively in the future. Areas open to reform include (1) acceptable damages and the flood insurance program, (2) water project cost sharing, (3) communication of flood risk, (4J water project planning, and (5) water policy and management at the national level. FINDINGS AND RECOMMENDATIONS This committee's task was to evaluate the scientific and engineering knowl- edge on which the selection of a flood hazard reduction strategy for the lower American River will ultimately be based. The committee also endeavored to provide insights on public policies concerning flood hazard management that are of concern to the nation. In line with that dual charge, the committee offers findings and recommendations specific to the USACE planning process as ap- plied to the American River basin, as well as some broader comments on the nature of flood risk assessment and its application nationwide. The findings and recommendations presented in detail in Chapter 7 relate to (1) the identification and evaluation of alternatives, (2) environmental issues, (3) risk methodology, (4) flood risk management behind levees, (5) risk communica- tion, and (6) water resources planning and decisionmaking. Some of the key issues are summarized here, but Chapter 7 provides a fuller treatment of the findings and recommendations. · Overall, the committee finds that the 1991 American River Watershed Investigation, as supplemented by the 1994 Alternatives Report, was reasonably complete in its consideration of structural flood protection measures. Alternative
SUMMARY 11 assumptions could have been selected, but nothing of a degree that should call the overall results into question. . The committee does not and can not judge whether construction of a dry dam at the Auburn site is the best approach to reducing Sacramento's flood risk. However, the committee strongly believes that if a dry dam is built it must contain operational gates to ensure management flexibility, protect public safety, and minimize environmental impacts. Environmental concerns are significant, and additional research is needed to understand the potential impacts of a dam on the canyon environment, particularly plant communities and slope stability. Such information could be used to help set operational guidelines so impacts of such a dam could be minimized. In addition, if a dam is built, the committee believes it should be used as a last line of defense to contain peak flows from extreme floods, thus reducing the frequency of impacts on the canyon. · The new USACE risk and uncertainty procedures are an innovative and timely development. The explicit recognition of modeling uncertainty should result in a better understanding of the uncertainty of flood risk and damage reduction estimates. However, the committee is concerned about the specific ways in which uncertainty is currently represented and included in the calculation of average flood damages and the residual risk of flooding, and about USAGE's ability to communicate information about flood risk and community vulnerabil- ity. USACE leadership is encouraged to convene an intra-agency workshop, including outside experts, to review the new risk and uncertainty procedures. · The determination of the federal interest in construction of water manage- ment facilities has always been a complex process affected by many factors, such as societal goals, the nature of the problem to be addressed, and financial con- straints. The rationale for federal interest in flood control in the American River basin should be reviewed, and Congress should explicitly address whether fed- eral involvement is warranted on the basis of the presence of widespread national benefits from flood protection or a limited ability of the community to provide its own flood protection. If a federal interest is clear, project construction should be delayed until SAFCA, working with FEMA and private insurers, has a program to require that new development at Natomas and in the city purchase flood insurance at actuarially sound rates for the residual risk. Also, SAFCA should implement a flood hazard mitigation plan, to be part of the area's land use plans, that includes flood risk communication, flood warning systems, evacuation plans to reduce loss of life, highway and other infrastructure designs to facilitate evacu- ation, and floodproofing and elevation requirements. The fundamental question in the American River planning process is how to reduce flood risk in the lower American River basin given a decisionmaking arena that includes significant scientific uncertainty and organized opposition to some of the possible risk reduction alternatives. This report discusses the uncer- tainties that confront floodplain managers and offers recommendations in many
2 FLOOD RISK MANAGEMENT AND THE AMERICAN RIVER BASIN areas, including the need for additional research in some areas. But decision- makers, agency officials, and interest groups reading this report should not use calls for additional research as an excuse for not taking action. It is time to select and implement a flood risk reduction strategy for the American River basin. There are still areas where data and information are incomplete, particularly in our understanding of environmental impacts, but that should not forestall the decisionmaking process. The recommendations offered in this report are in- tended to improve the process, not delay it further. THE ROLE OF SCIENCE IN THE DECISIONMAKING PROCESS The issue decisionmakers face is how best to determine and then implement an acceptable flood risk management program for the American River basin. Beyond all the complexities and subtleties, the ultimate question is whether the flood damage reduction offered through a combination of measures not including a dam is acceptable, or whether a new upstream dam is judged to be necessary to reduce risk to an acceptable level. The committee cannot answer that question, in part because detailed technical analyses comparing the alternatives are still being developed (this information is expected in the Sacramento District's forthcoming Supplemental Information Report, scheduled to be available in the summer of 1995) and, importantly, because that judgment is beyond the committee's appro- priate role. The public should be forewarned that even when the technical analy- ses are available, there will be no simple technical answer. Scientists and engi- neers can and should provide careful analyses and interpret the information so it is available to support decisionmakers, and they should be frank about uncertain- ties and risks. But the decision to be made should ultimately reflect more than technical factors; it should reflect economic considerations and value judgments pertaining to the appropriate use of natural resources, public monies, acceptable levels of risk, and willingness to accept constraints on land use. The final deci- sion on these issues rests with the public and the political officials who represent them.
