The Missouri River Ecosystem: Exploring the Prospects for Recovery (2002)

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 Over the past century, human activities have caused substantial eco- logical changes to the Missouri River ecosystem. By any measure, the Missouri River ecosystem—the Missouri River’s main channel and its flood- plain—has experienced significant reductions in natural habitat and in the abundance of native species and communities. There have also been sub- stantial reductions in the daily and annual variability of hydrologic and geomorphic processes. Causes of these changes include the removal of snags from the river in the late 1800s; introduction of nonnative fish species beginning in the late 1800s; navigation enhancement beginning in the early 1900s; and damming and flow regulation of the mainstem Missouri River beginning in the 1930s. Land use changes (including urbanization, agricul- ture, transportation infrastructure) and population growth have also af- fected the ecosystem in less direct but important ways. River systems have often proved remarkably resilient, withstanding a variety of human modifications and still responding positively to ecosystem restoration efforts. Strategies for improving ecological conditions in large river systems are relatively new, but some smaller rivers have exhibited rapid and positive ecological responses. In the Kissimmee River in Florida, for example, plant communities, fish, and invertebrates responded favor- ably to water-level manipulation experiments in the early 1980s. More recently, the breaching of Edwards Dam on Maine’s Kennebec River in 1999 resulted in increases in the abundance of select bird and fish species. Nonetheless, there is a point beyond which a large, degraded river system 1 

2 THE MISSOURI RIVER ECOSYSTEM can only be recovered with costly remediation efforts. Some losses—such as species extinction—can never be restored. Given the size and complexity of the Missouri River ecosystem, it is not clear where the point of irreparable environmental change lies, or how close the Missouri River ecosystem might be to passing that point. However, the following changes in the Missouri River ecosystem jeopardize its funda- mental natural processes: the loss of natural flood pulses; the loss of natural low flows; straightening of stream meanders and the elimination of cut- and-fill alluviation; losses of natural riparian vegetation; reductions in wa- ter temperature variation; introduction of nonnative species; and extensive bank stabilization and stream channelization. Specific examples of twenti- eth century changes in the Missouri River ecosystem include the following: • Nearly 3 million acres of natural riverine and floodplain habitat (bluff to bluff along the Missouri River’s mainstem) have been altered through land-use changes, inundation, channelization, and levee building. • Sediment transport, which was the hallmark of the pre-regulation Missouri River (and was thus nicknamed “The Big Muddy”), has been dramatically reduced. Sediment transport and deposition were critical to maintaining the river system’s form and dynamics. For example, before the 1950s, the Missouri River carried an average of roughly 142 million tons of sediment per year past Sioux City, Iowa; after closure of the dams, an average of roughly 4 million tons per year moved past the same location. • Damming and channelization have occurred on most of the Mis- souri River basin’s numerous tributary streams, where at least 75 dams have been constructed. • The amplitude and the frequency of the Missouri River’s natural peak flows have been sharply reduced. With the occasional exception of downstream sections in the state of Missouri, the Missouri River no longer experiences natural spring and summer rises and ecologically-beneficial low flows at other times of the year. • Cropland expansion and reservoir impoundment have caused re- ductions in natural vegetation communities. These vegetation communities continue to shrink with the additional clearing of floodplain lands. The remaining remnant areas will be critical in any efforts to repopulate the floodplain ecosystem. • Reproduction of cottonwoods, historically the most abundant and ecologically important species on the river’s extensive floodplain, has largely ceased along the Missouri River, except in downstream reaches that were flooded in the 1990s and in upstream reaches above the large dams. • Production of benthic invertebrates (e.g., species of caddisfly and mayfly) has been reduced by approximately 70 percent in remnant unchannelized river reaches. Benthic invertebrates are an important food 

EXECUTIVE SUMMARY 3 source for the river’s native fishes and an important component of the river’s food web. • Of the 67 native fish species living along the mainstem, 51 are now listed as rare, uncommon, and/or decreasing across all or part of their ranges. One of these fishes (pallid sturgeon) and two avian species (least tern and piping plover) are on the federal Endangered Species List. • In many reaches of the river, nonnative sport fishes exist in greater abundance than native fish species. The nonnative fishes are often more tolerant of altered conditions of temperature, turbidity, and habitat. Al- though some nonnative fish produce substantial economic benefits, nonna- tive species may also contribute to the declining abundance of native fish. These ecosystem changes are not merely abstract, scientific measurements; they also represent the loss of valued goods and services to society. Ex- amples of ecosystem goods and services include outdoor recreation, bio- mass fuels, wild game, timber, clean air and water, medicines, species rich- ness, maintenance of soil fertility, and the natural recharge of groundwater. It is often difficult to recognize the economic values that are lost with declines of these ecosystem benefits, largely because they have historically not been carefully measured. But ecosystem-based activities often provide important economic benefits. For example, thousands of people enjoy canoeing on the Missouri River in Montana each year, which provides an important source of tourism-based income. The values of many of these services historically have not been mon- etized and are not traded in economic markets. Changes in the benefits flowing from these services are thus not easily recognized. With the excep- tion of select outdoor recreation activities, most ecosystem goods and ser- vices tend to be undervalued by decision makers and in resources manage- ment policies. But there is a growing recognition that the replacement costs of these services, assuming their replacement is even possible, would be very high. Degradation of the natural Missouri River ecosystem is clear and is continuing. Large amounts of habitat have been transformed in order to enhance social benefits, and the ecosystem has experienced a substantial reduction in biological productivity as a result. Natural riverine processes, critical to providing ecosystem goods and services, have been greatly altered. The ecosystem has been simplified and its production of goods and services has been greatly compromised. Degradation of the Missouri River ecosystem will continue unless some portion of the hydrologic and geomorphic processes that sustained the pre- regulation Missouri River and floodplain ecosystem are restored—includ- ing flow pulses that emulate the natural hydrograph, and cut-and-fill 

4 THE MISSOURI RIVER ECOSYSTEM alluviation associated with river meandering. The ecosystem also faces the prospect of irreversible extinction of species. STATE OF THE SCIENCE There is a rich, extensive body of scientific research on the Missouri River ecosystem that can provide the foundation for future river manage- ment actions. For example, a 1997 technical report from the U.S. Geologi- cal Survey listed 2,232 studies of the Missouri River ecosystem. These scientific studies date back to Lewis and Clark’s epic 1804–1806 expedi- tion. Since then, many individuals and government agencies have studied the basin’s natural vegetation, fishes, water quality, and impacts of dams. This research has greatly improved scientific understanding of the river’s ecosystem and how it has changed. These studies have provided careful documentation of the ecological changes described in this report. Research on the Missouri River ecosystem provides a sound scientific understanding of ecological structure and the controlling river processes, and how they were impacted by human actions during the twentieth cen- tury. Although knowledge of the ecological intricacies within a system as large as the Missouri River ecosystem will always be limited by scientific uncertainties, the system’s broad ecological parameters and patterns are currently well understood. Nonetheless, existing studies are only a starting point for future man- agement choices because this extensive body of research has not been ad- equately synthesized. Further, the studies have tended to focus on specific species or portions of the river. Only a few studies of Missouri River ecology view the river as a single system from headwater to mouth, or as a single system that considers biological and physical linkages. The lack of synthesis and utilization of these scientific data may be as much a function of institutional and political barriers as it is of the limita- tions of the scientific information itself. Neither discrete scientific disci- plines nor mission agencies have been provided with sufficient incentives to conduct this synthesis and integration. Without this fundamental informa- tion, cast within a system-wide perspective encompassing the entire Mis- souri River ecosystem, truly comprehensive assessments of the ecological state of the Missouri River are not possible. The most significant scientific unknowns in the Missouri River ecosys- tem are how the ecosystem will respond to management actions designed to improve ecological conditions. In addition to improving ecological condi- tions, such actions can also help supplement existing scientific knowledge, especially in understanding how select ecological variables respond to dif- ferent environmental conditions. Management actions, cast as carefully circumscribed and monitored experiments, are necessary in order to ad- 

EXECUTIVE SUMMARY 5 vance our understanding of how regulated rivers respond to changes. It is important that ecosystem monitoring programs be designed specifically to produce results that serve as input into river ecosystem recovery programs. The emerging paradigm of adaptive management provides a useful conceptual basis for framing such management actions. The concept has been and is currently being used to guide ecosystem restoration efforts in the Colorado River, the Columbia River, and the Florida Everglades. Adap- tive management is also being initiated by Missouri River management agencies. The U.S. Army Corps of Engineers, for example, in its August 2001 revised draft environmental impact statement for the Missouri River Master Water Control Manual, acknowledges the importance of adaptive management. This committee was requested to comment on “policies and institu- tional arrangements . . . that could promote an adaptive management ap- proach to Missouri River and floodplain ecosystem management.” Adaptive management recognizes that scientific uncertainties and un- foreseen environmental changes are inevitable. It thus seeks to design organizations and policies that can adapt to and benefit from those changes. Adaptive management is not merely an elaborate “trial and error” ap- proach. Rather, it emphasizes the use of carefully designed and monitored experiments, based on input from scientists, managers, and citizens, as opportunities to maintain or restore ecological resilience and to learn more about ecosystems. These actions are monitored for scientific findings to help improve understanding of how policy decisions affect ecosystems. Findings from ecosystem monitoring are then to be used to appropriately adjust management policies. Adaptive management requires that clear goals and desired outcomes be established so that progress toward desired future conditions can be assured. Although adaptive management is a powerful approach that holds great potential, it should not be viewed as a panacea for Missouri River basin management and ecosystem improvements. The committee was keenly aware that the practice of adaptive management is “a work in progress” and that there is inadequate experience with successful or unsuccessful experiments to comprehensively evaluate the underlying theory. Adaptive management is not necessarily easy to implement and execute and, like the Missouri River basin itself, presents many complexities. In those eco- systems where it has been implemented, it has proven useful in many ways. However, endangered species are still listed, stakeholders still disagree with one another, and key management agencies are constrained by resources, legal mandates, and political realities. Nonetheless, there can be little disagreement that a new management paradigm is needed if further declines in the Missouri River ecosystem are to be halted and reversed. Adaptive management represents a framework for promoting stakeholder discussion 

6 THE MISSOURI RIVER ECOSYSTEM and for strengthening the links between the Missouri River ecosystem and the region’s economies and societies. Just as adaptive management encour- ages experiments, implementation of adaptive management will in itself represent an experiment. But no other alternative restoration strategy holds the promise that adaptive management does, and federal, state, and local governments, as well as several other National Research Council com- mittees, have embraced the concept as an important instrument to promote biodiversity conservation and restoration. The U.S. Army Corps of Engineers, in cooperation with the U.S. Fish and Wildlife Service and several state agencies, has completed and is imple- menting several habitat preservation and restoration projects along the Missouri River. These projects represent useful steps toward recovering the Missouri River ecosystem. However, they are limited in scope, are insuffi- ciently coordinated among agencies and among various reaches of the river, receive inconsistent funding, and lack adequate support for monitoring. These programs also are not framed within an overarching plan for recov- ering key elements of the Missouri River’s pre-regulation hydrologic and geological processes. The sum of these efforts is insufficient to noticeably recover ecological communities and fundamental physical processes in the Missouri River ecosystem. To substantially improve the ecosystem, a more systematic and better-coordinated approach that considers ecological con- ditions on par with other management goals in the entire Missouri River system will be required. MISSOURI RIVER NAVIGATION AND BANK STABILIZATION No Missouri River management issue has polarized the river’s stake- holders as much as the debate over how the provision of flows and channel depths for navigation has affected the Corps’ ability and willingness to meet ecosystem needs. Improved navigation was a major feature of the mid- twentieth century vision of the 1944 Pick–Sloan Plan, as navigation’s future economic benefits were assumed to be substantial. However, the 1950 projections for commercial waterway traffic were overly optimistic; com- mercial towboat traffic on the Missouri River peaked in 1977 (below pro- jected levels) and has fallen slowly and steadily since then. Missouri River navigation, conducted on the river’s 735-mile channelized stretch between Sioux City, Iowa and St. Louis, Missouri, is controversial for both economic and environmental reasons. The current dam and reservoir operation schedules reduce the river’s natural hydrologic variability in order to provide a steady and reliable 9-foot deep navigation channel. Such operations run counter to established river science, in which a large degree of natural hydrological variability is essential to biological 

EXECUTIVE SUMMARY 7 productivity and species richness of large floodplain rivers. A resolution of the differences between managing flows for navigation or for more natural hydrology is constrained by the fact that the benefits of navigation are expressed in dollars, while the benefits of ecosystem improvements from operational changes have yet to be monetized. The ultimate decision regarding the proper balance between these uses is a public policy issue and, as such, was beyond this committee’s charge. Nevertheless, this issue is so crucial to the river’s future that this committee could not ignore it. Differences of opinion may be artificially magnified by framing Missouri River navigation as an “all or nothing” issue. Coopera- tive dialogue might be easier if incremental changes in navigation and river management were considered. Because net national navigation benefits are relatively small in total, and because waterway traffic volumes decrease moving upstream, an incre- mental analysis of the economics of retaining segments of the navigable waterway would be useful. Relaxing the responsibility to maintain naviga- tion flows would make it demonstrably easier to introduce flows for im- proving river ecology in that segment. As an example, if the segment from Sioux City, Iowa downstream to Omaha, Nebraska proved to be uneco- nomic when comparing its incremental benefits with its incremental costs— factoring in the values of all potential ecosystem goods and services—then that segment would be a candidate for enhancing river ecology through operational changes. Ecological enhancement, however, would not neces- sarily proceed rapidly. The banks along the river’s navigable channel are stabilized and contain communities and other important infrastructure in many areas. If it is decided to enact management actions to improve the state of the ecosystem, and if those management actions are to be effective, some degree of Missouri River meandering must be restored. Allowing the Missouri River to meander would require a significantly wider public corri- dor in some portions of the channel than currently exists. This would require close coordination with those who live and work along the river. In some cases, significant improvements in river ecology may require reloca- tions. In proceeding segment by segment, the analysis should discover the point at which it is beneficial to retain navigation to the mouth of the river. The case for retaining some navigation might be stronger if navigation were discontinued or less fully supported in those segments where it is economi- cally inefficient. Congress should give the Corps of Engineers authority to provide navigation services on an incremental basis along the channelized portion of the Missouri River, to be exercised on the basis of analysis and stakeholder input. 

8 THE MISSOURI RIVER ECOSYSTEM POLICIES, INSTITUTIONS, AND ADAPTIVE MANAGEMENT The Corps of Engineers constructed and operates six of the seven mainstem dams on the Missouri River; the U.S. Bureau of Reclamation operates the seventh, Canyon Ferry Dam, east of Helena, Montana. When the Corps of Engineers constructed five of the Missouri River mainstem dams in the 1950s and 1960s after passage of the Pick–Sloan Plan, goals for dam and reservoir operations were to reduce flood damages, enhance navi- gation, generate hydroelectric power, and store water for irrigation. But changes in social preferences have resulted in a new mix of uses and stake- holders on the Missouri River today. Many of these new uses revolve around recreational and environmental considerations, such as boating and sport fishing. Some Missouri River stakeholders, such as environmental and recreational groups, call for revised operations and a redistribution of the river’s benefits. Other stakeholders, such as the navigation industry, the hydropower industry, and floodplain farmers generally prefer the status quo. Scientific knowledge, economies, and social preferences have clearly changed across the Missouri River basin since the mainstem dams were planned and constructed. However, the institutional and policymaking framework for Missouri River management has not changed accordingly. The decision-making context for the Missouri and its tributaries is charac- terized by prolonged disputes, disaffected stakeholders, and degrading eco- logical conditions. Barriers to resolving this policy gridlock on the Mis- souri River include a lack of clearly stated, consensus-based, measurable management objectives, powerful stakeholders’ expectations of a steady delivery of entitlements, and sharply differing opinions and perspectives among some Missouri River basin states. Current management protocols for operating the Missouri River sys- tem represent an accretion of federal laws, congressional committee lan- guage, appropriations instructions, and organizational interpretations that have been enacted or developed over the past century. This guidance has generally not been updated to reflect changing economic and social condi- tions, new needs in the basin or the nation, or advances in scientific knowl- edge. The Corps of Engineers and some basin stakeholders view the collective statutes, committee reports, and agency interpretations as barriers to prospective management changes that would seek to balance ecological values and services with current realities and values of navigation, recre- ation, and sound floodplain management. Although this committee believes the Corps of Engineers may have greater legal authority to manage the Missouri River system than it has exercised, the Corps’ ability to do so has been constrained by sharp differences of opinion among stakeholders. If the condition of the Missouri River ecosystem is to improve, agencies 

EXECUTIVE SUMMARY 9 responsible for adaptive management must have clear lines of authority and the necessary resources to work toward this goal. The Corps of Engineers has always set the water release schedules for the Missouri River mainstem dams. Guidance for mainstem dam water release priorities is established in the Corps’ Missouri River Main Stem Reservoir System Reservoir Regulation Manual, also known as the “Master Manual.” Decisions regarding water release schedules from the Missouri River mainstem reservoirs ultimately determine the distribution of the river’s benefits. As mentioned, these decisions have become increasingly contro- versial and pose challenges to the Corps. In the late 1980s, the Corps of Engineers began a revision to its 1979 Master Manual that today—thirteen years later—is not yet complete be- cause of competing demands for the river’s resources and sometimes strong differences of opinion. In working toward this revision, the Corps has consulted with numerous stakeholders and the public at large across the Missouri basin, including environmental groups, the navigation industry, farmers, and other floodplain residents and communities. Any agency would be challenged to find a solution amenable to all users in the current context of Missouri River management, and a consensus on how the dams and reservoirs are to be operated has remained elusive. A moratorium on further revision of the Master Manual should thus be implemented until such revisions reflect a collaborative, science-based approach based upon adaptive management to improve the condition of the Missouri River eco- system. Adaptive management should be adopted as an ecosystem management paradigm and decision-making framework for modifying water resources and reservoir management for the Missouri River ecosystem. As part of this management strategy, the goal of improving ecological conditions should be considered on par with other management goals. Specific Mis- souri River adaptive management experiments and activities—involving a broad spectrum of river system stakeholders in a collaborative process to establish goals and guidelines for such experiments—should be implemented as soon as possible. Adaptive management actions for improving ecologi- cal conditions should be examined and conducted within a systems frame- work that considers the entire Missouri River ecosystem from headwaters to mouth, as well as the effects of tributary streams on the mainstem. Determining specific goals and objectives for Missouri River manage- ment that society desires will require the participation of a wide spectrum of groups with stakes in Missouri River management. Missouri River mainstem reservoir operations objectives and means, including adaptive management actions, should be set by a formal multiple-stakeholder group that includes, but is not necessarily limited to, the U.S. Army Corps of Engineers, the U.S. Department of Energy, the U.S. Environmental Protec- 

10 THE MISSOURI RIVER ECOSYSTEM tion Agency, the U.S. Fish and Wildlife Service, the U.S. National Park Service, Indian tribes, the Missouri River basin states, floodplain farmers, navigation groups, municipalities, and environmental and recreational groups. The stakeholder group should review other adaptive management efforts to learn about successes, failures, and potential management actions that could be usefully implemented on the Missouri. To help resolve scien- tific uncertainties and to assure progress in considering some level of eco- system recovery, a scientific peer-review process that includes an indepen- dent, interdisciplinary scientific panel should provide solicited input to the stakeholder group. Support of the U.S. Congress is ultimately needed to help establish acceptable goals for the use and management of the Missouri River system. Congress must also identify the necessary authorities to do so. The stake- holder group should help frame Missouri River management decisions. But if the trends of ecosystem decline are to be halted and reversed, that stake- holder group must define ecosystem improvements as one of its key goals. Federal legislation mandating ecosystem protection and enhancement is one means to help stakeholders focus on Missouri River ecology. Sustained stakeholder participation in a system the size of the Missouri River ecosys- tem, and in which there are sharp differences of opinion over appropriate management goals, will require sustained commitments of time and re- sources. Some of the participants may possess inadequate resources and will require assistance to ensure their participation. Successful implementa- tion of adaptive management will also require administrative and facilita- tion resources. Congress should provide the necessary legislative authorities and the fiscal resources to implement and sustain an adaptive management ap- proach to Missouri River management. Resources should include adminis- trative and facilitation services for a multiple-stakeholder group to develop consensus positions on river management objectives and reservoir opera- tions policies. To ensure support of the adaptive management program and management actions that balance contemporary social, economic, and en- vironmental needs in the Missouri basin, Congress should enact a federal Missouri River Protection and Recovery Act designed to improve ecological conditions in the Missouri River ecosystem. This act should include a requirement for periodic, independent review of progress toward imple- menting adaptive management of the Missouri River ecosystem. 

1 Introduction Rivers, watersheds and aquatic ecosystems are the biological engines of the planet. World Commission on Dams, 2000 The Missouri River basin (Figure 1.1) extends over 530,000 square miles and covers approximately one-sixth of the continental United States. The one-hundredth meridian, the boundary between the arid western states and the more humid states in the eastern United States, crosses the middle of the basin. The Missouri River’s source streams are in the Bitterroot Mountains of northwestern Wyoming and southwestern Montana. The Missouri River begins at Three Forks, Montana, where the Gallatin, Jefferson, and Madison rivers merge on a low, alluvial plain. From there, the river flows to the east and southeast to its confluence with the Missis- sippi River just above St. Louis. Near the end of the nineteenth century, the Missouri River’s length was measured at 2,546 miles (MRC, 1895). Large, looping meanders of the main channel, some of which were nearly circular and that measured tens of miles in circumference, were then prominent features of the river. Much of the river has since been dammed, straight- ened, and channelized, and these large meanders have been virtually elimi- nated. As a result, the Missouri River’s length today is 2,341 miles—a shortening of roughly 200 miles (USACE, 2001). Between 1804 and 1806, Meriwether Lewis and William Clark led the first recorded upstream expedition from the river’s mouth at St. Louis to the Three Forks of the Missouri, and eventually reached the Pacific coast via the Columbia River. The Missouri River subsequently became a corri- dor for exploration, settlement, and commerce in the nineteenth and early twentieth centuries, as navigation extended upstream from St. Louis to Fort Benton, Montana. Social values and goals in the Missouri River basin in 11 

12 THE MISSOURI RIVER ECOSYSTEM FIGURE 1.1 The Missouri River Basin. SOURCE: USACE, undated. this period reflected national trends and the preferences of basin inhabit- ants. Statehood, federalism, and regional demands to develop and control the river produced a physical and institutional setting that generated de- mands from a wide range of interests. Over time, demands for the benefits from the Missouri’s control and management resulted in significant physical and hydrologic modifications to the river. These modifications led to substantial changes in the river and floodplain ecosystem. Numerous reservoirs are scattered across the basin, with seven large dams and reservoirs located on the river’s mainstem. Six of these dams were constructed pursuant to a 1944 agreement between the U.S. Army Corps of Engineers and the Department of the Interior’s Bureau of Reclamation. The agreement, ratified by the U.S. Congress, is known as the Pick–Sloan Plan and is the effective existing management regime for the Missouri River. The Pick–Sloan Plan represented a merger of Missouri River basin development plans that were formulated independently in the early 1940s by the Corps of Engineers (the Corps’ “Pick Plan” was headed by Colonel Lewis A. Pick) and the Bureau of Reclamation (the Bureau’s “Sloan Plan” was headed by Regional Director William G. Sloan). The separate plans were coordinated in Senate Document 247 (S.D. 247), which 

INTRODUCTION 13 was part of the Flood Control Act passed by Congress on December 22, 1944. The final paragraph of S.D. 247 states that the plan “will secure the maximum benefits for flood control, irrigation, navigation, power, domes- tic, industrial and sanitary water supply, wildlife, and recreation.” The first public mainstem dam on the Missouri River pre-dated Pick– Sloan. The Fort Peck Dam was built in Montana in the 1930s as a Works Progress Administration project promoted by President Franklin D. Roosevelt. The five mainstem dams downstream of Fort Peck and dozens of tributary dams were constructed as part of Pick–Sloan. Missouri River mainstem reservoirs behind Fort Peck Dam in Montana (Fort Peck Lake), Garrison Dam in North Dakota (Lake Sakakawea), and Oahe Dam in South Dakota (Lake Oahe) are three of the nation’s five largest human- made lakes (only Lake Mead and Lake Powell, both on the Colorado River, are larger). Although the river and its tributaries are extensively controlled by dams, channel modifications, and bank stabilization projects, the Mis- souri River is still subject to flooding, especially on the lower river. Like most major U.S. water projects, the Missouri River dams were authorized and built prior to the passage of modern environmental statutes such as the National Environmental Policy Act (1969) and the Endangered Species Act (1973), but not the Fish and Wildlife Coordination Act of 1934, which pre- dates most of the dams. The Corps of Engineers constructed and operates six of the Missouri’s seven mainstem dams (the Bureau of Reclamation constructed and operates Canyon Ferry dam, the comparatively small mainstem dam farthest up- stream). Operations of these six dams are guided by the Corps’ 1979 Missouri River Main Stem Reservoir System Reservoir Regulation Manual, usually referred to as the “Master Manual.” A severe drought across the basin in the late 1980s and early 1990s focused national attention on the tensions and conflicts among management objectives and competing ben- eficiaries. During this drought, upper basin reservoirs were drawn down (reducing benefits for recreation and tourism), and lower basin states expe- rienced disruptions to navigation and water supplies. The pronounced drought of 1988–1992 affected most parts of the Missouri River basin. Negative impacts on reservoir-based recreation (up- stream), on navigation (downstream), and on threatened and endangered species were so severe that in 1989, Congress directed the Corps to review the Master Manual. That review was conducted according to guidelines in the National Environmental Policy Act (NEPA), which requires the Corps to conduct an environmental impact statement (EIS) regarding changes in dam operations. As early as August 1994, the U.S. Fish and Wildlife Service (USFWS) issued jeopardy opinions (which state that a proposed action will jeopardize the existence of a threatened or endangered species) regarding operation of the Missouri River dams and the threat to federally 

14 THE MISSOURI RIVER ECOSYSTEM listed species (the Fish and Wildlife Service opinions were issued as part of the environmental impact study process). This followed the Corps’ issu- ance of the Master Manual Draft Environmental Impact Statement, which recommended changes in the management of the dams and reservoirs. The Corps conducted public hearings on this draft document. These hearings revealed controversies and passionately-held beliefs surrounding the river’s many users. A consensus emerged that recognized the need for improved ecological monitoring and scientific knowledge to improve river manage- ment. Nevertheless, the National Environmental Policy Act environmental impact statement process—initiated when the Corps began revisions to its Master Manual in 1989—and a final revision of the Corps’ Master Manual for operation of the Missouri River system had not been completed in early 2002, nearly 14 years after the Master Manual revision process began. Congress, the Missouri River basin states, and the basin’s water users and interest groups disagree on the appropriate water release schedule (includ- ing timing, locations, and quantities of water) for the Missouri River’s mainstem reservoirs. In 1999, with sponsorship of the U.S. Environmental Protection Agency (EPA) and the Corps of Engineers, the Water Science and Technology Board of the National Research Council (NRC) formed a committee of experts to help provide a better scientific basis for river management decisions in the Missouri River basin. This study complements similar NRC studies of the Columbia River basin, the Colorado River basin, the Florida Everglades, and the Upper Mississippi River. It also recognizes a growing public inter- est in redressing modifications made to large river ecosystems. This com- mittee was given the following charge: This committee will provide a general characterization of the historical and current status, and important ecological trends, of the Missouri River and floodplain ecosystem. The committee will provide a review of the available scientific information on the Missouri River and floodplain eco- system, and will identify and prioritize scientific information needs for improved Missouri River management. The committee will also recom- mend policies and institutional arrangements that could improve scientific knowledge of the Missouri River and floodplain ecosystem, and those that could promote adaptive management of the Missouri River and flood- plain ecosystem. The committee’s task was thus divided into three objectives: 1) Characterize the historical and current ecological status of the Missouri River and floodplain ecosystem. This overview will identify key ecologi- cal conditions, changes, and processes, endangered and threatened spe- cies, trends and relevant time scales, and gaps in and the limits of that knowledge. 

INTRODUCTION 15 2) Identify and describe the general state of existing scientific information on the Missouri River and floodplain ecosystem. Identify and prioritize the key scientific questions to be addressed and the key scientific informa- tion needed for improved Missouri River management. 3) Recommend policies and institutional arrangements for improving Mis- souri River and floodplain ecosystem monitoring and research, and those that could promote an adaptive management approach to Missouri River and floodplain ecosystem management. This committee began its two-year study late in 1999. Five meetings were held along the river: Bismarck, North Dakota; Columbia, Missouri; Great Falls, Montana; Omaha, Nebraska; and Pierre, South Dakota (a sixth meeting was held at the National Academies’ Beckman Center in Irvine, California, in February 2001). The committee spoke with federal and state scientists and engineers, representatives from Indian tribes, ex- perts on Missouri River institutions and policies, groups interested in Mis- souri River ecology and river management, the commercial navigation in- dustry, and many citizens. This report focuses on the Missouri River ecosystem. However, an understanding of the larger context of water resources development is help- ful in explaining some of the patterns reflected across the Missouri basin. Namely, changing values and water management policies in the United States are part of a larger global shift in which assumptions about the benefits of dams and the ability to appropriately distribute those benefits are being rethought. ECOLOGICAL CONDITIONS AND TRENDS IN U.S. RIVERS The rivers of the United States underwent considerable hydrologic and ecological changes during the twentieth century. The most obvious of these changes was the inundation of extensive stretches of these rivers behind dams. The twentieth century saw the Corps of Engineers and the Bureau of Reclamation, along with local, state, and private entities, construct hun- dreds of dams and greatly increase water storage. For example, in a given year, 60 percent of the United States’ entire river flow can be stored behind dams (Hirsch et al., 1990). Dams in the Missouri River basin have the capacity to hold roughly 106 million acre-feet of water, with the six Corps of Engineers Missouri mainstem reservoirs having a combined capacity of roughly 73.4 million acre-feet, making it North America’s largest reservoir system (USACE, 2001). The waters stored by these reservoirs are intended to serve multiple purposes, including irrigation, recreation, and controlled releases for navigation enhancement. The reservoirs are also operated so 

16 THE MISSOURI RIVER ECOSYSTEM that flood-control storage is available, an amount that fluctuates through the year in response to snowpack and precipitation conditions in the basin. Major hydrologic changes in some sections of the Missouri River at- tended the closures of the Corps’ mainstem Missouri dams: Fort Peck in 1937, Fort Randall in 1952, Garrison in 1953, Gavins Point in 1955, Oahe Dam in 1958, and Big Bend Dam in 1963. When they were constructed, the Missouri River mainstem dams were intended to help control river flows and to reduce streamflow variability. Today, however, there is a better understanding of and appreciation for the ecological values and services supported by streamflow variability (Box 1.1 describes the values of ecosystem goods and services). Decreases in riverine wetlands and other riparian (streamside) habi- tats in U.S. river systems have resulted from population growth and eco- nomic development, as well as from structural alterations (e.g., straighten- ing of channels, bank stabilization, and construction of wing dams) designed to constrict flows to a main channel. A variety of indicators might be used to measure changes in these ecosystems. To use one example, a National Research Council committee estimated that total wetland acreage in the contiguous United States decreased by approximately 117 million acres— half the original total—by the mid-1980s (NRC, 1995). Another study found that two-thirds of the pre-European settlement areas of riparian vegetation in the United States have been replaced by other land uses (Moberly and Sheets, 1993). Regardless of the measure chosen, U.S. ripar- ian ecosystems have been greatly altered during the past century. Human impacts on U.S. rivers have reduced populations of many aquatic species, including some extinctions. In the Columbia River basin, Pacific salmon have disappeared from about 40 percent of their historical breeding ranges over the past century (NRC, 1996). In the Upper Missis- sippi–Illinois River system, the number of mussel species has declined by 23 percent to 44 percent since European settlement in the nineteenth century (USGS, 1999). Only four of eight endemic fishes remain in the Grand Canyon reach of the Colorado River, and some of these are threatened or endangered (Minckley, 1991). Engineered changes in the nation’s rivers have enhanced competition, predation, and other detrimental interactions between native and nonnative species (Minckley and Deacon, 1991), which has contributed to the demise of native species. Missouri River reservoirs and river segments presently contain populations of exotic fishes, including cisco, several salmon and trout species, and several Asian carp species (Hesse et al., 1989). Some of these species have contributed to the development of economically impor- tant recreational fisheries. On the Upper Mississippi River, scientists re- ported increased abundance of species such as bluegill and largemouth 

INTRODUCTION 17 Box 1.1 Ecosystem Goods and Services Although knowledge of the importance of ecosystems to societies and econo- mies dates back centuries, discussion of nature’s importance in terms of goods and services is a relatively recent phenomenon. Because many functions provided by ecosystems are not monetized and are not traded in markets, values are often under-appreciated by the public and by decision makers (Daily, 1997). For example, clean air and clean water provided by ecosystems are fundamental to healthy societies and economies, but price tags are generally not affixed to air and fresh- water systems. They thus may be treated as having no monetary value in market- based decisions. But if rational natural resources decisions are to be made, it is important to understand how ecosystems provide value to societies and the mag- nitude of those values. Ecosystem goods and services include fish protein, fish-based recreation, bio- mass fuels, wild game, timber, clean air and water, medicines, species richness, maintenance of soil fertility, and natural recharge of groundwater. Consistent with the fact that ecosystem goods and services are generally not priced, are not traded in markets, and are not owned, they tend to become needlessly scarce. To rem- edy this, conservation programs that protect fish and wildlife are enacted, parks and natural reserves are created, the use of the biosphere as a sink for wastes is regulated, and programs aimed at restoring natural habitat are mandated. A variety of approaches are in use to correct for the unowned and unpriced nature of many ecosystem goods and services. Tradable rights or quotas have been introduced for certain pollutants and in some fisheries. Charges are levied by governments to prevent the overuse of certain goods and services. Several methods are used to place simulated market values on goods and services that would otherwise be unpriced in policy making or in court decisions. bass, which colonized habitats with slow-moving (lentic) water, after the river’s navigation dams were constructed (Fremling and Claflin, 1984). SHIFTING VALUES AND PUBLIC PREFERENCES Large, regional water projects no longer enjoy the widespread political support they once did. The economic rationale for these projects has eroded and there is today more concern over these projects’ environmental and social costs. As a result, the arid and semiarid western United States is shifting from the reclamation era—characterized by the construction of large, federally subsidized regional water projects—to an era of realloca- tion, conservation, and ecosystem restoration. The Bureau of Reclamation today focuses on management and maintenance of existing projects and on ecological improvements in degraded stream systems. Similarly, the Corps of Engineers is faced with the challenge of carrying out engineering and construction activities while balancing competing social, environmental, 

18 THE MISSOURI RIVER ECOSYSTEM and economic demands in highly developed and highly controlled river systems. The Corps’ traditional roles have been expanded by Congress to include environmental restoration and programs that address environmen- tal problems associated with existing projects. For example, the Corps plays a central role in the multi-billion dollar Florida Everglades restoration project. The value of dams today is questioned by segments of society that value environmental preservation and enjoyment. Some smaller U.S. dams have been breached or removed (e.g., Edwards Dam on the Kennebec River in Maine was breached in 1999), and others are scheduled to be removed (e.g., Elwha Dam in the state of Washington). In addition, the possible removal or decommissioning of some large federal dams (e.g., four dams on the lower Snake River and Hetch Hetchy Dam in Yosemite National Park) has been discussed (ASCE, 1997; Gleick, 2000). During the 1980s and 1990s, Congress passed specific legislation aimed at protecting and/or restoring aquatic ecosystems in the Columbia River basin, the Florida Ever- glades, the Grand Canyon, and the San Francisco Bay Delta. The need to consider the conditions under which dams and hydroelectric power facili- ties should be retired has also drawn attention from professional engineer- ing groups (ASCE, 1997). Changing views toward large dams are reflected in the recent report of the World Commission on Dams (WCD, 2000). In 1997, the World Bank and the International Union for the Conservation of Nature and Natural Resources (The World Conservation Union) assembled a group to discuss the highly controversial issues associated with dams. These parties agreed to a proposal to work together to establish a World Commission on Dams (WCD), which in 1998 began a comprehensive global and independent review of the performance and impacts of large dams. The commission’s final report was issued in 2000 (WCD, 2000). Although care must be taken in applying findings from the commission’s global review of dams to the Missouri River, many of the commission’s findings regarding environment, indigenous people, equity, and sustainability are applicable to the Missouri River basin and to the United States. This committee reviewed the com- mission’s report with interest and, where its findings are relevant, refers to this report. ADAPTIVE MANAGEMENT To properly balance social, economic, and environmental consider- ations in large river ecosystems, organizations and management policies must be able to respond to and take advantage of changing environmental, social, and economic conditions, as well as address extreme events. The concept of adaptive management promotes the notion that management 

INTRODUCTION 19 policies should be flexible and should incorporate new information as it becomes available. New management actions should build upon the results of previous experiments in an iterative process. It stresses the continuous use of scientific information and monitoring to help organizations and policies change appropriately to achieve specific environmental and social objectives. Adaptive management promotes collaborative and consensus-based decision-making. Adaptive management promotes “thinking outside the box” and stakeholder discussions about the desired state of the ecosystem. Responsive organizations and policies that can adjust decisions on dam operations are needed to meet changing scientific and social goals: “Dams and the context in which they operate are not seen as static over time. . . . Management and operation practices must adapt continuously to changing circumstances over the project’s life and must address outstanding social issues” (WCD, 2000). Adaptive management requires an organizational and political frame- work for its full and proper implementation. To be successful, it should be implemented by organizations with sufficient legal authority and political legitimacy to appropriately adjust management policies. Scientific investi- gations will never eliminate all economic, engineering, environmental, and social uncertainties in large ecosystems like the Missouri River basin. Policy decisions must account for these uncertainties. Organizations responsible for promoting adaptive management must have the legal authority and the stakeholder support necessary to make and enforce recommended changes in current management regimes. Adaptive management also promotes the advancement of scientific knowledge through carefully designed experiments and monitoring sys- tems. Water resources managers and scientists across the United States are conducting numerous experiments, at a range of spatial and temporal scales, with water releases and diversions to benefit select species and ecosystems. Perhaps the most prominent experiments in river and dam management are controlled releases of high flows from reservoirs. The most famous con- trolled release in United States water management was a controlled flood from Glen Canyon Dam in March 1996. The controlled flood in the Grand Canyon aimed to restore beaches that had been damaged by decades of low hydrologic variability. The notion of operating a dam to purposely create a large flood represented a milestone in U.S. water management. Former Secretary of the Interior Bruce Babbitt described the event and the process leading up to it: “There was simply no precedent on the Colorado River—or as far as we know anywhere in the history of civilization—for what Interior was proposing to do” (Babbitt, 1999). Beyond reservoir releases, possible adaptive management actions for the Missouri River in- 

20 THE MISSOURI RIVER ECOSYSTEM clude changing the length of navigation seasons, changing patterns of irri- gation water withdrawals, changing elevations of navigation pools, and constructing notches in flood-control levees. This committee studied carefully the history of efforts to create coordi- nated management schemes for the Missouri River basin through federal river authorities modeled on the Tennessee Valley Authority (TVA), through interstate compacts, and through an entity composed of federal, state, and tribal representatives. In general, the proposed organizations lacked the necessary political support to achieve agreement on implementation and have thereby been unable to resolve most intra-basin conflicts. The lack of such a management authority in the Missouri River basin has created a management vacuum that has been filled by the Corps and increasingly by the courts (Thorson, 1994). If adaptive management is chosen as a para- digm by which to coordinate Missouri River management organizations and policies, it must be considered and implemented in the context of these current and historical organizational efforts. It would require Congress, federal and state agencies, Indian tribes, and other public and private stake- holders to forge an agreement placing adaptive management at the center of a process for reaching compromises on the full array of river management issues. This committee addressed its charge against a backdrop of over a cen- tury of actions devoted to developing and managing the Missouri River for economic and social ends. Before evaluating contemporary Missouri River management issues, a review of the historical development of the Missouri River and its floodplain is appropriate.