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.
Executive Summary In response to a request by the Assistant Secretary of the Interior for Water and Science and the Assistant Secretary of the Army for Civil Works, the National Research Council established an ad hoc Committee on Safety Cri- teria for Dams to prepare an inventory of currently used criteria for dams relating to safety from hazards of extreme floods and earthquakes, and to identify and evaluate alternative criteria for safety of federal dams. The committee was macle up of recognized experts in risk assessment, regulation of dams in the interest of public safety, law, and science and engineering, most of them involved with consideration of safety of dams from hazards of extreme floods and earthquakes. The committee was assisted in its analysis of the problems of protecting dams and those persons and properties down- stream as well as upstream from dams from flood and earthquake hazards by several federal agency representatives and a great many organizations and individuals who provided information on current practices. The committee considered a wide variety of issues ranging from the sciences of meteorology and earth tectonics to the role of government and principles of common law. This report presents the results of the committee's work and its recommenda- tions. The prime focus of the committee's considerations was the protection of dams from failures caused by floods and earthquakes and, thereby, the protection of people and properties from the hazards of dam failures. The committee's principal findings and conclusions are summarized as follows: 1. Extreme floods and earthquakes are random events that can but may not occur during the life of a dam. 1
2 SAFETY OF DAMS 2. Current dam design practices reflect great diversity in the standards used for classifying dams relative to hazards and in the criteria for evaluation of safety from extreme floods and earthquakes. Of the organizations re- sponding to the committee's request for information, the U. S. Army Corps of Engineers, the Bureau of Reclamation, and some others apply state-of-the- art methods in evaluating safety of dams from floods and earthquakes. 3. While simple hazard rating categories based on downstream develop- ment may be useful for identifying dams for high-priority safety evaluation and study, they do not reflect the potential for incremental loss of life and damage caused by the failure of a dam due to an inadequate spillway when a river is already in flood. 4. More uniformity is needed among the several federal and state agen- cies establishing size and hazard definitions and correlative design stan- cIards. (The committee has not recommended size and hazard definitions in this report, but reference to hazard classifications in the report correspond in a general way to those used in the National Program of Inspection of Non- Federal Dams conducted by the U.S. Army Corps of Engineers.) 5. Many of the basic concepts for estimating extreme flood-producing capabilities of watersheds have been used for several decades. However, new concepts and improved methods for estimating floods, changes in patterns of land use and an expanding data base have resulted in generally larger flood estimates. For these same reasons, in general future estimates of magnitudes of extreme floods can be expected to continue to increase. However, unless the runoff characteristic of the watershed were to change, increments in future estimates for a specific basin resulting from extension of the data base should, in general, be less than those noted in the past. Also, it is noted that there have been instances of later, more intensive hydrometeorological stud- ies that resulted in reductions in estimates of probable maximum precipita- tion potentials from those established by earlier investigations. 6. A clam designed for the estimated probable maximum flood, based on the estimated probable maximum precipitation as determined by current practices, does not necessarily provide absolute assurance that the dam is safe for every possible flood. 7. During the past 20 years, there have been major advances in our knowledge of earthquake ground motions, in our abilities to predict resul- tant structural response, and in the methods of analyzing the safety of dams against earthquakes. 8. Methods of estimating potential earthquake ground motions are be- coming more reliable in regions with well-identified fault systems (such as the western United States) and also in regions where fault systems are not well identified (such as the eastern United States). 9. Currently, three basic approaches are used to determine the magni-
Executive Summary tude of floods for which the safety of dams should be evaluated (i.e., the deterministic, probabilistic, and risk analysis approaches). Each approach has advantages and disadvantages that must be evaluated in selecting an appropriate method for any specific project. 10. Currently, three basic approaches are used to determine the magni- tude of earthquake motions for which a dam will be evaluated (i.e., the deterministic-statistical, seismotectonic province, and probabilistic-risk analysis approaches). At the present time, the deterministic-statistical and the seismotectonic approaches are the most widely used. 11. Risks are present in every human activity. The federal government has become increasingly involved in risk management issues in the last two decades. A number of federal agencies have developed various risk manage- ment standards for specific types of risk, but these are not applicable to risk management for dams. Adaptation of these concepts to dam safety requires research that has not been done by this committee. 12. Court decisions relating to dam failures in general have held the owner liable for the damages resulting from a failure. The committee recommends the following: 1. To the extent practicable, reservoir safety evaluations should strike a balance among such considerations as project benefits, construction costs, social costs, and public safety, including the possible consequences of dam failure due to major earthquakes and floods. (While achieving such balance is the ideal, currently available technology does not permit this balancing with full confidence in the results. ~ 2. Safety evaluation standards for existing dams ant] proposed dams neec! not necessarily be the same. 3. The use of probable maximum floods (PMF), based on estimated prob- able maximum precipitation (PMP), as the general design standard (safety evaluation flood (SEF)) for proposed high-hazard dams should be contin- ued. However, instances sometimes will be encountered where a lower stan- dard may be justified if failure of a dam during floods of PMF magnitude would cause no significant increase in potential for loss of life or property damage. 4. For existing high-hazard dams, the adopted safety evaluation flood (SEF) should take into account estimated flood probabilities, expected proj- ect performance, and incremental damages that would result from dam failure for a range of floods up to and including the probable maximum floods. 5. In the design of new dams and spillways when design alternatives of approximately equal cost are available, a selection among these alternatives
4 SAFETY OF DAMS should give consideration to potential future needs for increaser] safety against extreme floods and earthquakes. 6. The maximum credible earthquake (MCE) ground motions developed by the cieterministic-statistical method applied to known causative faults provide an acceptable level of conservatism for safety analyses of high- hazard dams; this approach shouIcl be used whenever possible. However, research should be continued to develop probabilistic-risk analysis methods. Where earthquake sources are not well identified, the seismotectonic prov- ince method for determining MCE should be used. 7. The maximum credible earthquake (MCE) should be adopted as the safety evaluation earthquake (SEE) for a high-hazard dam; but the SEE for a lower-hazard dam may, in some cases, be less severe. 8. Response of a high-hazard dam to earthquakes in regions of significant seismicity should be analyzed utilizing dynamic analysis techniques, com- bined with appropriate judgment, with the objective of assuring that the safety evaluation earthquake will not cause catastrophic release of water from a reservoir. 9. The safety evaluation of dams need not consider the simultaneous occurrence of the safety evaluation flood and the safety evaluation earth- quake because of the extremely low probability of such occurrence. 10. Periodic reviews of hazard determinations and safety decisions for all dams should be required, especially when safety evaluations are based on criteria less conservative than the probable maximum flood or the maximum credible earthquake. 11. Research efforts designed to provide better bases for estimating mag- nitudes and frequencies of extreme floods ant] earthquakes, for estimating reactions of dams to such natural phenomena, and for establishing accept- able levels of risks should be continuecl. 12. As advances occur in seismology, hydrology, meteorology, and the relevant data bases, and as changes are noted in public attitudes toward risk, the federal agencies should periodically undertake a review of dam safety practices and standards by an independent body representing the professions involved in engineering for clams and experts from other relevant clisciplines.
