The Missouri River is the longest river in North America and flows from the semi-arid western states to the relatively moist Midwest. An integrated system of large reservoirs, constructed in the mid-20th century, provides important water storage, hydroelectricity and flood control benefits. This system has been managed by the US Army Corps of Engineers which has traditionally followed its original mandate to support navigation and flood control. As water uses and societal values have evolved, the management of the river has slowly evolved, and the Army Corps of Engineers has adopted adaptive management to incorporate biological uncertainties into its decision-making. Other evolution, including the incorporation of economic incentives into water management, has failed to occur.

## Introduction

At 3,767 kilometers, the Missouri River is the longest river in North America. The management of the Missouri River has been an issue of regional controversy since the 1930s. Efforts to utilize this key natural resource for navigation, irrigation, hydropower generation, and recreation while minimizing flood damage have led to a highly regulated river that has been greatly modified from its original state. It has also led to a management regime that prioritizes water flows to meet demands that were important in the 1930s–1960s and are relatively unimportant in the 21st century.

In the 1930s–1950s, the US Army Corps of Engineers (the Corps), which manages large multi-purpose dams, became the de facto river master of the Missouri River. Through congressional authorization under the US Constitution's interstate commerce clause, the federal government has maintained predominance in Missouri River management. Efforts to establish a river basin authority or an interstate river compact have failed. The Corps' role in river management has historical precedence in the Mississippi Basin, where navigation has been a key objective of national policy since the early 19th century (Niebling et al., 2014).

In the Missouri River Basin, the federal government must balance the generally consumptive water uses in the relatively dry upper basin states with predominantly non-consumptive water uses in the relatively wet lower basin states. Many western states, including the key upper basin states have adopted the prior appropriation doctrine of water allocation. These laws, which generally award water rights to claimants on a first in time, first in right basis, can easily conflict with navigation, which is a priority for river management in the eastern United States. Furthermore, the Endangered Species Act has led to increased conflicts between consumptive uses and instream uses of water. The resulting conflicts have resurrected decades-old disputes between upper and lower basin states regarding priorities for consumptive and non-consumptive uses of water, and between states' and federal power advocates.

This paper reviews how institutions that manage the Missouri River have evolved with changing societal and economic needs, and assesses whether this evolution is meeting changing needs. Because the focus of this paper is basin-wide institutions, federal and interstate institutions are featured. The second section of the paper gives a brief geographic overview of the river and its basin, and the third section provides historical background on institutional evolution in the basin. The fourth section reviews changing uses of Missouri River water and the fifth section assesses recent institutional evolution in the basin. The conclusion discusses further institutional innovation that would improve water management in the basin.

## Missouri River Basin geography

The Missouri River and its tributaries flow from the northern Rocky Mountains through the northern Great Plains merging with the Mississippi River near St Louis, Missouri (see Figure 1). The Mississippi then proceeds to the Gulf of Mexico. The Missouri River Basin covers 1/6th of the contiguous United States and includes areas in 10 states: all of Nebraska, much of Kansas, Montana, North Dakota, South Dakota, Wyoming, and parts of Colorado, Iowa, Minnesota, and Missouri. Ninety-five percent of the land use in the basin is agricultural. Some of the nation's most important grazing and crop lands are located in the basin. Principal metropolitan areas in the basin include Denver, Colorado, Omaha, Nebraska, and Kansas City, in Missouri and Kansas. Region 7 (2014) of the US Environmental Protection Agency reports the population of the US portion of the basin to be 10 million in 2014. This compares to the 1990 US census population estimate of 12 million, which shows a significant population decrease (US Army Corps of Engineers, Omaha District, 2006).
Fig. 1.

Map of Missouri River Basin. Source:Vineyard (1997).

Fig. 1.

Map of Missouri River Basin. Source:Vineyard (1997).

The 1,379,000 square kilometers in the basin include 25,123 square kilometers in the Canadian provinces of Alberta and Manitoba. The Canadian portion of the basin, which is sparcely populated, includes portions of the Milk River drainage. The apportionment of the Milk River between the USA and Canada began in 1909 and is part of the longest lasting apportionment agreement between these countries (Bankes & Bourget, 2013).

Thirty tributaries contribute to the mainstem Missouri. Major tributaries include: the Yellowstone River, currently the longest free-flowing river in the contiguous USA; and the Platte River, which has a number of irrigation storage reservoirs in its upper tributaries. In general, the semi-arid upper basin, west of the 98th meridian near the Fort Randall Dam, is dominated by irrigated agriculture and range, whereas the humid lower basin is dominated by rain-fed crop production (Otstot, 2011).

The basin is home to climate extremes, including long droughts, periodic floods, cold winters with frequent storms, and hot and dry summers. Future climate change is expected to increase the frequency of extreme weather events, such as droughts and floods, causing slight increases in mean temperatures and precipitation mostly in the spring and winter months (Shafer et al., 2014). This will increase the importance of well-managed reservoir storage.

## Evolution of water management in the Missouri River Basin

Institutions are the norms and rules that guide human behavior. As human behavior evolves, institutions should evolve in a manner that provides new guidance to meet new needs. Institutions that are durable provide stable expectations of future behavior. Such stability enables investment and growth. However, institutions that are stagnant might interfere with the evolving needs of societies, technologies, and economies. As societies become larger, more diverse institutions have gradually evolved away from informal ‘rules of behavior’ and toward formal codes, laws, agencies, and compacts. To meet societies' evolving needs, water management agencies have generally evolved and at times become more specialized. Some water management agencies focus on the needs of a single constituent sector, such as municipal water supply, whereas others attempt to meet broad societal goals with integrated multi-sectoral water management. Infrastructure development has necessitated the formation of additional institutions and sometimes new agencies. As infrastructure has become more complex, with multiple outputs and multiple beneficiaries, multi-purpose agencies and management authorities have developed (North, 1990; Saleth & Dinar, 2004; Hearne, 2007).

The basis of water law in the United States was established before the Louisiana Purchase made the Missouri River basin part of the United States in 1803. Water was managed by the separate states under state laws, which were adapted from the English Common Law including the doctrine of riparian rights. The US Constitution gave the federal government jurisdiction over interstate commerce and the interstate commerce powers gave the federal government authority over navigable rivers. In 1824, the US Supreme Court ruled that the federal government could regulate non-navigable tributaries of navigable waters in order to aid commerce (Niebling et al., 2014).

In the early 19th century, the Corps was the only federal government unit with engineering expertise. Congress was eager to employ this specialized unit with peacetime efforts to improve navigation on the nation's waterways. In addition to constructing harbor facilities and fortifications, the Corps dredged and cleared waterways. The Corps' expertise expanded to building locks, dams, and canals to facilitate navigation. The Corps' survey work on Mississippi River flooding was initiated before the US Civil War and led to a long-lived preference toward construction of levees for flood control. However, with the passage of the 1936 Flood Control Act, the federal government was granted responsibility for flood control and the Corps began building and operating multi-purpose reservoirs (Flood Control Act of 1936, 1936; Arnold, 1988).

The legal doctrine of prior appropriation evolved from the mining custom and was particularly well suited for developing incentives to invest in the frontier. The doctrine was tested in the Colorado territorial case of Yunker v. Nichols in 1872. An appeals court ruled that a non-riparian landowner had the right to convey water across the land of a riparian landowner in the Platte River basin. Other cases followed and Colorado, Kansas, Montana, Nebraska, North Dakota, South Dakota, and Wyoming formalized seniority-based user rights (Cech, 2003; Hobbs, 2004).

By recognizing the principle of first-in-time, first-in-right for beneficial uses of water, prior appropriation provides incentives to develop water capture and transport infrastructure. Prior appropriation does not place a priority on the type of use, although in most states the beneficial use of water includes most consumptive uses and originally excluded instream and aesthetic uses of water. Once a farmer, rancher, municipality or industry claimed water and demonstrated beneficial use, the claim could be registered with the state, and thereafter, considered to be a permanent right to water that was generally transferable under the stipulations of state water law. Western water law has evolved substantially since the 1870s and each state has its own water laws with provisions for conflict resolution. Most of these states allow for well-regulated market transactions of water and water-use rights (Josephson, 1987; Hobbs, 2004).

The imposition of the prior appropriation doctrine on the western US frontier eventually led to conflict with land and water rights claimed by Native American tribes. The principle of reserved water rights for sufficient water to fulfill the purpose of the tribal reservations was established by the US Supreme Court in Winters v. United States. The rights reserved for tribal reservations date back to the establishment of the reservations, which in most cases are senior to the claims of the settlers. Over the decades, there has been substantial adjudication of the particular allotments to be reserved for the tribes. These adjudication proceedings have been complicated by the predominance of states' courts in water allocation cases and the fact that tribal nations are represented in state courts by the federal government, which is often biased toward the predominant, non-tribal, economic and societal interests (Veeder, 1972; Membrino, 1992; Getches, 2009).

The authority of western states to allocate water rights was already established before the federal government developed large scale irrigation projects. At the time, much of the area in the western states was federal land and the federal government had become heavily involved with infrastructure development. The Reclamation Act of 1902, initiating the federal government's efforts in developing irrigation projects, maintained that states allocated water. This law established the US Reclamation Service, which was renamed the Bureau of Reclamation (BuRec) in 1923. The BuRec was fully engaged in large dam construction and operation before the Pick-Sloan plan was implemented (Bureau of Reclamation History Program, 2011).

Because large dam projects produce complementary and competing services, there are advantages to having dams managed by multi-purpose intersectoral agencies. Flood control, hydroelectric generation, and inter-seasonal water storage are generally complementary, whereas consumptive uses of water, and at times navigation, are competing services.

The infrastructure projects that enable the Corps to manage the Missouri River were constructed from 1933 to 1963, and initiated in the 1930s with the construction of the Fort Peck Dam in Montana. This depression-era public works project was greatly expanded under the Flood Control Act of 1944, which promoted post-war resettlement and economic development in the basin. The Act proved to be a relatively expensive compromise between upper basin consumptive use interests and lower basin navigation and flood control interests. Two established, and often competing, federal government agencies participated in this compromise. The Pick-Sloan plan for the Missouri River and tributary dams represented a merger between a Corps plan to build mainstem dams and lower river levees to provide navigation from the Mississippi River to Sioux City, Iowa and a more ambitious BuRec plan that included 90 water storage, conduction, irrigation, and hydroelectric generation projects. The resulting Pick-Sloan plan featured six mainstem dams and reservoirs managed by the Corps and smaller tributary dams and irrigation projects managed by BuRec. This merger resulted from a consensus among powerful interest groups that large multi-purpose dams should be constructed at the end of the war to stimulate post-war employment and regional economic growth. This consensus did not extend to the use of the water behind these large dams (Flood Control Act of 1944, 1944; Guhin, 1985; Ferrell, 1993).

The debate over the allocation of water between upper and lower basin states manifested itself in Congress before the passage of the 1944 Flood Control Act. The O'Mahoney-Milliken amendment to the act, which authorized the Pick-Sloan plan, stipulated that uses of water east of the 98th meridian should not conflict with present or future beneficial uses of water west of the 98th meridian for irrigation, livestock, mining, and domestic and industrial uses. This amendment protected traditional consumptive uses that were important to the western states, and allowed for inter-basin transfer of water (Guhin, 1985; Ferrell, 1993).

## The evolution of water use in the Missouri River Basin

The evolution of water management institutions should follow the changing use and value of water. The demographics and the economy of the Missouri River Basin has changed significantly since the 1940s and water-use has also changed. The value of the services provided by the water and the infrastructure has also evolved.

Flood control has always been a high priority service of the Pick-Sloan reservoir system. Both the upstream reservoirs and the downstream levees protect low-lying riparian areas from flooding and allow these areas to be developed for infrastructure and crops. Flood protection benefits all riparian areas, but a majority of these benefits occur in downstream areas, including the lower Mississippi floodplains (Guhin, 1985; Committee on Missouri River Ecosystem Science, 2002). Cumulative flood protection benefits, including lower Mississippi River flooding, were estimated to reach $18,000,000,000 in 1999 (US Army Corps of Engineers, Omaha District, 2000). Reducing the extent of flooding in areas that were formally floodplains by the construction of levees and other structural measures allows for the establishment of permanent infrastructure and crops on these lands. Flood control benefits are expected to increase over time as infrastructure in riparian areas protected from floods becomes more developed and more valuable. And global climate change is expected to increase flood events. In the Great Flood of 1993 in the upper Mississippi and Missouri Rivers, damage was estimated to be$15,000,000,000 (in 1993 dollars) (Larson, 1996). The 2011 flood produced record runoff and significant damage throughout the river's riparian areas. This prompted the Corps to review the amount of reservoir storage that is dedicated to flood protection. That review concluded that dedicating additional storage space exclusively for flood protection would decrease valuable multi-purpose storage and the total value of reservoir storage capacity (US Army Corps of Engineers, Northwestern District, 2012).

Improved navigation has always been a principal benefit of the Corps' river management activities. Since the completion of the Pick-Sloan infrastructure, navigation on the Missouri River has decreased considerably. At the time of the 1944 Flood Control Act, the 1181 kilometer long, 91 meter wide, and 2.7 meter deep channel from Sioux City, Iowa to the confluence with the Mississippi River was considered to be a boom for the transport of agricultural commodities to export markets. Having an uninterrupted channel with sufficient depth during the months of operation, generally late March until December, was expected to bring increased river commerce (Guhin, 1985).

In the last few decades, rail transport has dominated the movement of agricultural commodities throughout the basin. The increase in agricultural commodities shipped through Pacific ports has decreased the movement of these goods south and east on the Missouri River. Agricultural tonnage shipped on the Missouri River has declined steadily since the 1970s (Committee on Missouri River Ecosystem Science, 2002). Public expenditure on maintaining the navigation channel caused the total cost of Missouri River transport to be very high, with cost per ton mile approximately 100 times greater than on the Mississippi River. A 2009 US Government Accountability Office study reported that 84% of the tonnage shipped on the Missouri River was sand and gravel; this percentage increased during the period 1994–2006. Most of these sand and gravel trips were for distances less than one mile. Although the navigation channel abuts four states, Missouri accounted for 84% of shipments (US Government Accountability Office, 2009).

A major objective of the Pick-Sloan dams, especially those in the more arid upper basin states, was the use of water for irrigation. The Dust Bowl era of the 1930s was recent history when the Pick-Sloan plan was authorized. Thus, improved crop yields from irrigation were featured in the rhetoric leading to its authorization. The plan touted the benefits of having the BuRec develop 2.1 million hectares of irrigated land in the basin. To date, only 223,000 hectares of irrigated land has been developed. Many of the proposed irrigation projects were de-authorized, although over 1 million hectares of land may still be irrigated (Otstot, 2011).

The failure of the BuRec to fully develop federally sponsored surface irrigation projects casts doubt on the claim that upper basin states needed to retain Missouri River water stored in mainstem dams for irrigation and the priority granted to this service under the O'Mahoney-Milliken amendment. Meanwhile, groundwater irrigation has become much more important in the basin. The development of improved pumps and center pivot sprinkler systems in the late 1940s and 1950s greatly expanded groundwater irrigation especially in the Ogallala Aquifer, which substantially overlaps with the Missouri River Basin (Hornbeck & Keskin, 2011). In Nebraska alone, 2.9 million hectares are being irrigated by groundwater sources representing 94% of the total irrigated hectares in the state (Gollehon & Winston, 2013). Unfortunately, the Ogallala reservoir is being depleted rapidly due to groundwater mining. A 2013 US Geological Survey report states that nearly one-third of the total decline in aquifer storage has occurred during the eight-year period 2001–2008 (Konikow, 2013).

Municipal and industrial (M&I) water supply, which was included in the O'Mahoney-Milliken amendment, is a high priority for consumptive water uses in the basin. Compared to the ambitious plans for irrigation, M&I use of water was not considered to be substantial. In 2006, the Corps estimated that total water withdrawal for M&I uses in the basin was 3.45 cubic kilometers. Much of this water was used for industrial cooling, 87% of which was returned to the basin as return flows. Although the total M&I consumption remains small and the basin population is decreasing, increasing population in key urban areas implies municipal water supply is a growing use of Missouri River water (Guhin, 1985; US Army Corps of Engineers, Omaha District, 2006).

The use of the Missouri River for industrial purposes has also been controversial, although this is a very minor use of water. Numerous thermal generation plants with a total generating capacity of 15,000 Megawatt (MW) use Missouri River water for cooling. Industrial water use has become, in certain instances, a very valuable use of water.

An effort to divert Missouri River water from Lake Oahe in South Dakota to a coal slurry operation in Wyoming resulted in an important 1988 US Supreme Court ruling in ETSI Pipeline Company v. Missouri. The ETSI company wanted to purchase South Dakota water rights through a special conservancy district and purchase the water from the BuRec. The Supreme Court ruled against this proposed transfer on the basis that only the Corps has a mandate to sell surplus water (Thorson, 1994).

The sale of surplus water for energy development opened new controversy in 2010 when the Corps released a Draft Environmental Impact Statement for a proposal to sell surplus Lake Sakakawea water to oil producers in North Dakota. The tremendous expansion of western North Dakota's Bakken shale region due to hydrological fracturing has increased the industrial and residential demand for water. The abundance of water in Lake Sakakawea and the failure to develop irrigation uses for the lake's water presented an opportunity to acquire water from Lake Sakakawea for this new industrial use. The proposal to sell water to oil producers was in accordance with the 1944 Flood Control Act and the ETSI decision, but without precedent. The state of North Dakota protested vigorously against the sale of water in Lake Sakakawea because the state considered that water to be theirs (US Army Corps of Engineers Omaha District, 2010; Sando, 2011). Eventually, the Corps temporarily suspended its plan to charge industrial users for this water. The 2014 Water Resources Reform and Development Act placed a ten-year moratorium on water charges for surplus water. In 2012, the Corps initiated the Missouri River Municipal and Industrial Water Storage Reallocation Study with the intent of addressing this issue (US Army Corps of Engineers, Omaha District, 2013; Congressional Research Service, 2014).

The production of hydroelectric power was always a valuable benefit of the Pick-Sloan dams. As the demand for electricity expanded, the production of hydroelectricity from the Pick-Sloan dams greatly surpassed initial plans of the 1940s. The 2009 hydropower potential for the Pick-Sloan system was 2,991 MW, which is four times greater than the original Pick-Sloan plan (Otstot, 2011). Because hydropower generation is subordinate to other dam storage uses, such as flood control, the potential for hydropower generation in the basin has not been achieved (Committee on Missouri River Ecosystem Science, 2002).

Since the formation of the Department of Energy in 1977, electric power has been transmitted and sold by the Western Area Power Administration (WAPA). The WAPA is an energy wholesaler, required to sell its energy at cost. WAPA does not charge the full cost of generating electricity because it does not contribute to the cost of dam construction. Although power production is a potentially lucrative source of revenue, the mandating legislation stipulated a preference in the sale of power to public institutions, municipalities, Indian tribes and cooperatives. This is very much in accordance with the rural electrification programs of the 1930s' New Deal.

Although some of the power generated by the system is sold at spot market prices for electricity, WAPA sells most of the power it generates through wholesale contracts with preferred customers, including public institutions, municipalities, Indian tribes, and cooperatives. Guhin (1985) emphasizes that re-selling this low cost power has been lucrative for the preferred customers. Guhin (1985) stresses that the practice of using such surpluses to subsidize municipal expenditures is not uncommon. The lost revenue to the federal government of preferential power sold at cost by the WAPA has not escaped the attention of Congress. However, a 1984 proposal to auction Hoover Dam energy to the highest bidder was rejected by the House of Representatives in a tight vote (Guhin, 1985).

Recreational uses of the river were not considered important prior to the passage of the 1944 Flood Control Act. However, despite their distance to major population centers, the mainstem reservoirs have become very valuable for recreation. Visitorship increased from less than 5 million hours per year in the early 1950s to over 60 million hours per year by the end of the 1990s (US Army Corps of Engineers, Omaha District, 2010). Recreational benefits exceeded \$87 million in 1994, with over 60% of these benefits generated in North and South Dakota (Committee on Missouri River Ecosystem Science, 2002). Maintaining sufficient water volume in mainstem dams for boating and fishing is one of the principal concerns of upper basin states, and the reason for their efforts to reduce releases from the mainstem dams (Boldt, 2003).

## Continued or stalled evolution of water management in the Missouri Basin

The Corps manages the river by regulating releases from the mainstem dams. It periodically updates its Master Manual for reservoir system management, which stipulates the plan for managing reservoir storage. In general, the mainstem reservoir space is allocated to: (1) permanent pools of water in each dam; (2) a large space for inter-annual storage for multi-purpose uses; (3) additional space for annual multi-purpose use and flood control; and (4) an exclusive storage space reserved for flood control (see Figure 2). With fealty to its original mandate, the Corps attempts to provide an eight-month navigation channel from April through November. The actual details of quantity and timing of releases are guided by the Master Manual, which was originally released in 1960 and revised in 1973, 1975, 1979, 2004, and 2006. That manual is updated each spring with an Annual Operating Plan that considers actual precipitation and river conditions. Although the Corps has argued that the Master Manual is an internal document that allows flexibility in implementation, courts have ruled that the Corps is bound by this plan (Seeronen, 2009).
Fig. 2.

Total system storage for Missouri River mainstem dam operations, end of July 2013. Source:US Army Corps of Engineers, Northwestern Division, 2014. Units are million acre feet (MAF) or 1.2335 cubic kilometer.

Fig. 2.

Total system storage for Missouri River mainstem dam operations, end of July 2013. Source:US Army Corps of Engineers, Northwestern Division, 2014. Units are million acre feet (MAF) or 1.2335 cubic kilometer.

In the 19th and 20th centuries, the Corps was not a multi-purpose agency. It developed an expertise and adhered to its mandate to support navigation and flood control. As manager of the Pick-Sloan dams, the Corps has become the de facto Missouri River master because two institutional alternatives, the basin authority and the river compact, were never adopted. State acquiescence to federal authority occurred despite the great fear of western state politicians that the Pick-Sloan plan would forfeit control of water to the federal government (Billington et al., 2005). A Missouri Valley Authority (MVA) was proposed in the light of the success of the Tennessee Valley Authority (TVA). The latter is a much publicized regional development agency and poster child of the Roosevelt Administration's New Deal. The TVA faced a number of fortuitous conditions that made its creation politically feasible. It also had the freedom to expand beyond water management. By providing low-cost electricity and fertilizer to the impoverished Tennessee River basin, the TVA became a successful regional development agency (Ferrell, 1993). This new institutional format for regional development and basin management was favored by President Roosevelt and his administration.

The political conditions that enabled creation of the TVA were not present during the 1940s when the Pick-Sloan plan was developed. In the initial legislative process that occurred in late 1945, the proposed MVA was stripped of any authority for navigation and flood control in the Senate Commerce Committee and the committee did not remove these responsibilities from the Corps. During the late 1940s, a number of MVA proposals were brought to legislative committee without passage. The two lead federal government agencies for the Missouri River System, BuRec and the Corps, collaborated with each other, which reduced political pressure to create a third agency for river management. In addition, President Truman did not share his predecessor's desire for an MVA.

The river compact is the other institution that was never adopted for Missouri River management. The US Constitution allows states to form congressionally approved compacts. Interstate water management compacts are common. The US Fish and Wildlife Service lists 39 interstate agreements for river management, 25 of which are interstate compacts for water apportionment, although some of these are different agreements that cover the same water body. These compacts include compacts for the Republican, South Platte, Upper Niobrara, and Yellowstone Rivers, all in the Missouri River basin. All ten states in the Missouri River basin, except for Iowa and Missouri, have been signatories to interstate water compacts. Most of the compacts on rivers in the eastern United States focus on water quality and flood control. There are no river compacts for the Mississippi, Missouri, and Columbia Rivers. These compacts are formed when agreements negotiated by states' representatives are approved by the required number of states' legislatures. Most of these compacts establish interstate commissions with representatives from the signatory states, and, sometimes, the federal government. For compacts that require an interstate commission to interact with the federal government on behalf of the signatory states, Congress recognizes the compact and the federal government often becomes a signatory (US Fish & Wildlife Service, undated). Some of these commissions have regulatory authority. Most do not have dispute resolution authority (US Government Accountability Office, 2007; US Fish & Wildlife Service, undated).

There have been a number of attempts to develop a Missouri River Compact. In 1950, the Truman administration's Water Resources Policy Committee recommended broad changes to basin management and supported regional commissions to integrate federal, state, and local initiatives. In the early 1950s, several states attempted to form a compact. State efforts to continue the Missouri River Basin Commission (MRBC) and form a compact were derailed in the 1980s by the distrust that coincided with the proposed ETSI transfer of water and the subsequent lawsuits. As recent as 2011, upper basin governors met to discuss a compact (Guhin, 1985; Ferrell, 1993; Thorson, 1994; Sioux City Journal.com, 2011). None of these measures were able to overcome interstate rivalry and distrust, which worked against a formal compact agreement. One reason why states are reluctant to agree to compacts is that once these compacts are approved by Congress they achieve the status of federal law and the states cannot retire from these compacts (Thorson, 1994).

The lack of a central authority over the Missouri River Basin and the need to establish protocols for interagency cooperation led to the development of the Missouri Basin Inter-Agency Committee, which had representatives of key federal agencies, including the Corps and BuRec, and states' representatives. This committee did not have a permanent staff and policy-making authority and was dominated by federal agencies. It eventually became ineffectual (Ferrell, 1993). In an effort to improve planning, the MRBC was established in 1972 by executive order. The MRBC included representatives from more than a dozen federal agencies, ten basin states, and a few tributary commissions. It was allocated a small budget for hiring staff and initiating a planning process. The MRBC was terminated by executive order in 1981 (Ferrell, 1993). Further attempts at inter-agency cooperation came with specific purposes. The Missouri River Basin Interagency Roundtable was established by a Memorandum of Understanding among 16 federal agencies in 1996, and recommitted in 2010, in order to promote improved ecosystem management (US Institute for Environmental Conflict Resolution, 2015). The Missouri River Flood Task Force was initiated in response to the flood of 2011 and included federal agencies as well as state and tribal representatives. The task force facilitated communication and cooperation with the purpose of recovery and flood risk reduction (Missouri River Flood Task Force, 2012).

The lack of state cooperation in the management of the Missouri River increased the dominant role of federal agencies. Two additional agencies became influential in the management of the river. After implementation of the Clean Water Act (CWA) of 1972, the US Environmental Protection Agency (EPA) began to regulate point source discharges and to subsidize sewage treatment plants. Under the CWA, the EPA requires states to implement surface water quality standards. If state monitors designate that water quality is impaired, then new restrictions on discharges can be imposed. The EPA has fostered state agencies to implement water quality programs that are in general different from those that oversee water allocation. Indeed basin states have gradually enhanced their capacities to deal with water allocation, CWA implementation, and conflict resolution (Davidson, 2000; Schempp, 2013).

Efforts to improve water quality through point source regulation have not been followed by sustained improvements in nonpoint source nutrient reduction. Nutrients in the lower Missouri River have increased over the period 1993–2003 (Sprague et al., 2009). One of the leading causes of impairment in the Missouri Basin is sediment, although natural levels of sediment in the river are greater than they are today. Conflicts between implementing state water quality standards designed to reduce turbidity and sedimentation, and the Endangered Species Act which would attempt to restore turbid river bottoms for the endangered pallid sturgeon, have resulted. The growth of the hypoxia in the Gulf of Mexico has demonstrated that high levels of nutrients from agriculture in the Mississippi Basin require further water quality efforts throughout the basin (Zellmer, 2011; Rabalais, 2014).

The CWA is jointly implemented by the EPA and state agencies that are, in general, different to the state agencies that manage water quantities and water rights under prior appropriation. Although there is a relationship between water quantity and water quality, the separation of the management of these services is fundamental in the US federal system where the states maintain primacy in water quantity management, especially in prior appropriation states, and the Federal government regulates water quality under the CWA. Conflicts have emerged when efforts to improve water quality have threatened prior appropriation rights and where state law does not consider the improvement of water quality to be a beneficial use of water. The capacity of the prior appropriation doctrine to evolve has been demonstrated in Montana and Nebraska where the appropriation of water rights to improve water quality has been accepted as a beneficial use (Schempp, 2013).

The US Fish and Wildlife Service (FWS) is another important federal agency in Missouri River management. The FWS became very powerful with the implementation of the Endangered Species Act of 1973. This law ‘requires Federal agencies to insure that any action authorized, funded or carried out by them is not likely to jeopardize the continued existence of listed species or modify their critical habitat’ (US Fish & Wildlife Service, undated). Three endangered species have critical habitat that are modified by Missouri River dam management: the least tern, listed as endangered in 1985; the piping plover, listed as threatened in the Great Plains in 1985; and the pallid sturgeon, listed as endangered in 1990. The Endangered Species Act requires federal agencies to consult with the FWS when they modify critical habitat. Thus, any change in the Corps' plans to manage reservoirs through its Master Manual must be accompanied by a Biological Opinion (BO) from the FWS.

Although the Pick-Sloan dams have stabilized water flows, periodic droughts and floods lead to conflict. There was frequent litigation from the 1980s through the 2000s in which both upper and lower basin states sought court rulings to change the Corps' reservoir management in their favor. Much of this litigation was stimulated by the FWS' recommendations to modify reservoir operation to provide for more natural river flows. Such flows can be detrimental to downstream navigation, especially in drought years. In general, the federal courts: (1) upheld the Corps' management decisions; (2) reduced the Corps' discretion in implementing the Master Manual; (3) upheld the primacy of navigation and flood control over recreation; (4) denied the use of state surface water standards under the Clean Water Act to override Corps' reservoir management decisions; (5) denied the claim that total economic benefits should guide reservoir management under the Flood Control Act of 1944; and (6) upheld the Corps and FWS' implementation of the Endangered Species Act. Another result of this litigation is the increased experience and expertise of Federal courts in managing water disputes (Seeronen, 2009).

After significant delays from litigation and political interference, the Corps and the FWS collaborated on a 1997 Master Manual that incorporated: (1) releasing water from the three large upper basin dams on a rotation, so that each reservoir would have low water levels only one year in three in order to support the growth of riparian vegetation and improve plover and tern habitat; (2) spring pulses of water from the Gavins Point Dam to aid pallid sturgeon habitat; and (3) adaptive management of the river system in order to modify management as biological conditions change. According to the Corps (Barnes, 2009), ‘[a]daptive management is an organized and documented undertaking of goal-directed actions, while evaluating their results to determine future actions’. The National Research Council (2004) states that ‘adaptive management promotes flexible decision making that can be adjusted in the face of uncertainties as outcomes from management actions and other events become better understood’.

Applied to the Missouri River system, adaptive management is an institutional innovation that accounts for uncertainty about future socioeconomic and biological conditions influencing the system and adjusts system management over time, as needed, to improve ecosystem health without inflicting measurable harm on other stakeholders (Grinwald, 2003; Prato, 2003; Seeronen, 2009). With its belated embrace of adaptive management, the Corps has evolved toward managing infrastructure to protect ecosystem services. This change required the Corps to gain expertise in dynamic biological systems, which is a far cry from its traditional bailiwick of rule-based management (Tarlock, 2004; Kronholm, 2013).

Ideally, adaptive management is collaborative. Development of the 2007 Master Manual included consultation with a Plenary Group of state, tribal, and stakeholder group representatives. The large number of environmental groups and tribes represented tended to dilute the states' influence. Similarly, the Missouri River Recovery Implementation Committee was established by the Corps to provide feedback on its adaptive management program. Invited as participants are representatives of eight states (excluding Colorado and Minnesota) and 28 tribes, appointed representatives from 16 stakeholder categories, and a federal working group consisting of members from 13 federal agencies (US Army Corps of Engineers, Omaha District, 2008; Seeronen, 2009; Bonneau et al., 2011). Involvement of this committee in the Corps' adaptive management processes is expected to occur through annual strategic reviews and could be minimal given the more direct collaboration of Corps and FWS personnel in the process. Clearly the Corps has not fully engaged the states in its decision-making. This is despite the growing expertise of states' water management agencies.

Another institutional failure in the Missouri River Basin is the inability to adjudicate the water rights claims of Native American tribal reservations under the doctrine established in Winters v. United States. The doctrine awards tribal reservations senior water rights sufficient to meet the authorized purposes of the reservation. The amount of water reserved under tribal rights generally equals the amount sufficient to irrigate all of the reservations' irrigable land that existed at the time the reservation was established. Because much of the land flooded by the Pick-Sloan reservoirs was tribal reservation land, the amount of irrigable land was reduced substantially when the reservoirs were created. This loss of riparian land has been called the worst injustice by the United States against any tribe (Thorson, 1994). In an effort to provide restitution to two affected tribes, the Fort Peck-Montana Compact was developed, which specifies reserved water rights for the Fort Peck Assiniboine and Sioux Tribes. Congressional approval of this compact has never been obtained due to objections of a Missouri senator in an effort to maintain the priority traditionally given to downstream navigation (Thorson, 1994).

The moratorium on water charges written into the 2014 Water Resources Reform and Development Act demonstrates another failure of institutional evolution. Development of the Pick-Sloan reservoirs was made at a time of Federal government largesse. The norm was that the Federal government built development projects that were given to constituencies with minimal payments for services provided. Under the ‘single project ultimate development concept’, neither flood control nor navigation was expected to contribute to covering the costs of the Pick-Sloan dams. Sales of irrigation water were expected to contribute to recover some of the project's costs, but without interest payments. Sales of hydroelectricity to preferred customers was expected to contribute toward capital costs with interest payments. The exception to the government largesse was the potential sale of surplus water to municipal and industrial users (Guhin, 1985). Although unprecedented, the sale of surplus water to the North Dakota oil industry was consistent with the 1944 Flood Control Act, the ETSI decision, and economic theory, which suggests that efficient allocation of water would be motivated by price incentives. The period of Federal government fiscal austerity from 2011 through 2014 provided an opportune time to challenge the norm of Federal largesse and adopt price incentives as a way to more efficiently allocate a small portion of Missouri River water in the upcoming century.

## Conclusions and observations

Societal needs for water and water resources management have evolved since the establishment of the Pick-Sloan reservoir system. The economic contribution of large surface irrigation projects and river barge navigation in the Missouri River has decreased substantially during the last 70 years. Meanwhile societal values for ecosystem services such as habitat and recreation have increased. Flood control remains a valuable service of the dam system and this value may increase with the increased severe climate events expected to accompany global climate change. Municipal and industrial water is expected to increase in importance with increased urbanization and continued expansion of the energy industry in the upper basin. However given the relatively small volumes of water involved, the M&I sector is not expected to become a major consumer of Missouri River water.

The need to modify water management and to reallocate water as societies' priorities change is not unique to the Missouri Basin. However the difficulties in managing an interstate river without a basin authority or an interstate compact are highlighted in this basin. The states and political leaders have failed to cooperate in establishing a river compact or a river management authority. Given the disparate goals of upper basin and lower basin states, interstate cooperation is not expected to advance in the near future. Thus the Corps has become and is expected to remain the de facto river master. The Corps' traditional expertise in construction and management of infrastructure is well suited to the goal of managing an integrated reservoir system to meet complementary and competing objectives. Whereas there will always be controversy with dam management the Corps does protect stakeholders from much of the whims of management to meet short-term jurisdictional political ends.

Because the Corps should be expected to maintain its predominant role in managing the mainstem Missouri River it will need to evolve, accept new roles, and meet new needs. It has recently embraced adaptive management and accepted biological determinants in its reservoir management decisions. It has thus expanded its expertise to include ecosystem restoration, which encompasses a wider skill set than the Corps employed in the 20th century. The strategy of spring pulses has demonstrated the Corps' evolution toward environmental interests and threatened the water quantities required for downstream navigation. The three-year rotation of water releases from the three largest upper basin reservoirs should be complementary to the non-consumptive requirements for upstream recreation. Continued evolution of the Corps away from its 19th century role of navigation enhancement toward a more complex multi-objective management would be expected to reduce the priority given to maintaining a Missouri River navigation channel. Such evolution will be needed in order to better mitigate the increased extreme climate events that are expected with global climate change.

As the Corps continues to manage the mainstem Missouri through its reservoirs it needs to establish protocols for collaboration with other Federal agencies − principally the FWS, EPA, BuRec and WAPA – and the states. The current collaboration within the adaptive management process dilutes the input from the states, and consequently disrespects the political representation of the vast majority of the basin's residents. Although collaboration is a means to improved management and not an end in itself, the Corps has so far missed the opportunity to provide leadership for more substantial cooperation that could facilitate communication, improve decision-making, and reduce conflict.

State and Federal agencies have cooperated in addressing water quality, but nutrient levels in the Missouri River have not decreased. Given the distance between the Missouri Basin and the Gulf of Mexico, and the high demand for agricultural commodities in recent decades, this will not be an easy challenge to overcome. However this is a national issue that will be need to be addressed eventually in the Missouri Basin and will require further policy innovations.

Surface irrigation never developed as expected under the Pick-Sloan plan, but groundwater irrigation has developed rapidly and aquifers are being depleted. Innovative ways to transfer water from the Missouri River's surface reservoirs to the basin's aquifers, including the Ogallala Aquifer, should be explored. The predicted increase in extreme climate events, such as floods and droughts, with global climate change will increase the value of an integrated surface and groundwater storage system. In order to achieve the benefits of integrated storage, engineering and institutional innovations will be required.

Water allocation is in general handled by state agencies, and challenges in the allocation of water remain. One is the settlement of reserved Native American rights under the Winters decision. The other is an agreed-upon definition of the surplus water that the Corps is allowed to sell as stipulated by the 1944 Flood Control Act. The presence of unallocated water in the upper basin states while senior rights to tribal reservation land remain unfulfilled implies that one problem can be the solution to the other.

The era of federal government largesse and New Deal regional development projects ended in the last decades of the 20th century. Resources such as the Missouri River and the Pick-Sloan integrated reservoir system need to be used wisely, and economic incentives should be respected. Congress should allow the Corps to market surplus water to industry. The potential sales are not expected to amount to a large percentage of water nor a large sum of money. But the incorporation of pricing incentives would raise revenue and move toward the allocation of water to more valuable uses.

## Acknowledgements

This material is based upon work supported, in part, by the National Institute of Food and Agriculture, US Department of Agriculture, under Project No. ND01315.

## References

References
Arnold
J.
(
1988
).
The Evolution of the 1936 Flood Control Act. Office of History United States Army Corps of Engineers. Fort Belvoir, Virginia. http://www.publications.usace.army.mil/Portals/76/Publications/EngineerPamphlets/EP_870-1-29.pdf
(accessed July 2015)
.
Bankes
N.
Bourget
E.
(
2013
).
Apportionment of the St. Mary and Milk Rivers
. In:
Water Without Borders?: Canada, The United States, and Shared Waters
.
Norman
E.
Cohen
A.
Bakker
K.
(eds).
University of Toronto Press
,
Toronto
.
Barnes
T.
(
2009
).
(accessed September 2014
)
.
Billington
D.
Jackson
D.
Melosi
M.
(
2005
).
The History of Large Federal Dams: Planning, Design, and Construction in the Era of Big Dams. Denver. Bureau of Reclamation. 623 p
. http://www.usbr.gov/history/HistoryofLargeDams/LargeFederalDams.pdf
(accessed July 2014)
.
Boldt
M.
(
2003
).
Judge removes order affecting Lake Sakakawea releases. The Bismarck Tribune. 29 May 2003
.
Bonneau
J.
Fleeger
T.
Fleming
C.
Hale
C.
(
2011
).
Missouri River Recovery Program Adaptive Management Process Framework. US Army Corps of Engineers, Omaha District. Paper 82. http://digitalcommons.unl.edu/usarmyceomaha/82
.
Consulted August 2014
.
Bureau of Reclamation History Program
(
2011
).
Brief History Bureau of Reclamation
. http://www.usbr.gov/history/
(accessed November 2014)
.
Cech
T.
(
2003
).
Principles of Water Resources: History, Development, Management, and Policy
.
Wiley
,
New York
.
Committee on Missouri River Ecosystem Science
(
2002
).
The Missouri River Ecosystem: Exploring the Prospects for Recovery
.
,
Washington
.
Congressional Research Service
(
2014
).
Library of Congress Summary: H.R. 3080: Water Resources Reform and Development Act of 2014. https://beta.congress.gov/bill/113th-congress/house-bill/3080
(accessed August 2014)
.
Ferrell
J.
(
1993
).
Big Dam Era: A Legislative and Institutional History of the Pick-Sloan Missouri Basin Program. US Army Corps of Engineers. Omaha Division. http://cdm16021.contentdm.oclc.org/cdm/ref/collection/p16021coll4/id/160
(accessed September 2014)
.
Flood Control Act of 1936
.
CHS. 649–651. 20, 22, June 1936. http://legisworks.org/congress/74/pubres-117.pdf
(accessed July 2015)
.
Flood Control Act of 1944
.
Act of 22 December, 1944, ch. 665, 58 stat 887. http://www.usbr.gov/power/legislation/fldcntra.pdf
(accessed July 2015)
.
Getches
D.
(
2009
).
Water Law in a Nutshell
,
4th edn
.
Thomson West
,
St. Paul
.
Gollehon
N.
Winston
B.
(
2013
).
Groundwater Irrigation and Water Withdrawals: The Ogallala Aquifer Initiative. NRCS. Economic Series no. 1. https://prod.nrcs.usda.gov/Internet/FSE_DOCUMENTS/stelprdb1186440.pdf
(accessed November 2014)
.
Grinwald
M.
(
2003
).
Bush v. The Missouri River: washed away
.
The New Republic
,
27 October 2003
.
16
18
.
Guhin
J.
(
1985
).
The law of the Missouri River
.
South Dakota Law Review
30
,
350
488
.
Hearne
R.
(
2007
).
Evolving water management institutions in the Red River Basin
.
Environmental Management
40
,
842
852
.
Hobbs
G.
(
2004
).
Citizen's Guide to Colorado Water Law
,
2nd edn
.
,
Denver
.
Hornbeck
R.
Keskin
P.
(
2011
).
The Evolving Impact of the Ogallala Aquifer: Agricultural Adaptation to Groundwater and Climate. NBER Working Paper 17625. National Bureau of Economic Research. Cambridge, MA. http://www.nber.org/papers/w17625
(accessed April 2014)
.
Josephson
R. M.
(
1987
).
An analysis of the potential conflict between the prior appropriation and public trust doctrines in Montana water law
.
Public Land Law Review
8
,
81
114
.
Konikow
L.
(
2013
).
Groundwater depletion in the United States (1900–2008): U.S. Geological Survey Scientific Investigations Report 2013 − 5079, 63 p. http://pubs.usgs.gov/sir/2013/5079
(accessed May 2014)
.
Kronholm
S.
(
2013
).
Adaptive management strategies in the United States Army Corps of Engineers – an analysis. University of Minnesota Water Resources Center, Minnegram. http://wrc.umn.edu/pubs/minnegramonline
(accessed August 2014)
.
Larson
L.
(
1996
).
The Great USA Flood of 1993. Paper presented at the IAHS Conference Anaheim, CA. June 23–28 1996. http://www.nwrfc.noaa.gov/floods/papers/oh_2/great.htm
(accessed May 2014)
.
Membrino
J.
(
1992
).
Indian reserved water rights, federalism, and the trust responsibility
.
Land and Water Law Review
27
,
36
58
.
(
2012
).
(accessed July 2015)
.
Niebling
W.
Baker
J.
Katz
S.
Smet
K.
(
2014
).
Challenge and response in the Mississippi River Basin
.
Water Policy
16
,
87
116
.
North
D.
(
1990
).
Institutions, Institutional Change, and Economic Performance
.
Cambridge University Press
,
Cambridge
.
Otstot
R.
(
2011
).
An Overview of the Pick-Sloan Missouri River Basin Program
.
Bureau of Reclamation
,
Great Plains Region
. p.
26
.
Prato
T.
(
2003
).
Adaptive management of large rivers with special reference to the Missouri River
.
Journal of the American Water Resources Association
39
,
935
946
.
Rabalais
N.
(
2014
).
Louisiana Universities Marine Consortium, Press Release. August 4, 2014. http://www.gulfhypoxia.net/news/default.asp?XMLFilename=201408210800.xml
(accessed October 2014)
.
Saleth
R. M.
Dinar
A.
(
2004
).
The Institutional Economics of Water: A Cross-Country Analysis of Institutions and Performance
.
Edward Elgar
,
Cheltenham
.
Sando
T.
(
2011
).
Letter to Colonel Robert Ruch. United States Army Corps of Engineers. Omaha District. http://www.swc.nd.gov/
(accessed August 2014)
.
Schempp
A.
(
2013
).
At the Confluence of the Clean Water Act and Prior Appropriation: The Challenge and Ways Forward
.
Environmental Law Institute
,
Washington
.
Seeronen
J.
(
2009
).
Judicial challenges to Missouri River mainstem regulation
.
Missouri Environmental Law & Policy Review
16
,
59
107
.
Shafer
M.
Ojima
D.
Antle
J. M.
Kluck
D.
McPherson
R. A.
Petersen
S.
Scanlon
B.
Sherman
K.
(
2014
).
Ch. 19: Great Plains
. In:
Climate Change Impacts in the United States: The Third National Climate Assessment Melillo
,
J. M.
Richmond
T. C.
Yohe
G. W.
(eds).
U.S. Global Change Research Program
, pp.
441
461
. Doi:10.7930/J0D798BC: .
Sioux City Journal.com
(
2011
).
North Dakota governor suggests Missouri River compact. August 4, 2011
. .
Sprague
L. A.
Mueller
D. K.
Schwarz
G. E.
Lorenz
D. L.
(
2009
).
Nutrient trends in streams and rivers of the United States, 1993–2003: U.S. Geological Survey Scientific Investigations Report 2008–5202, US Geological Survey, Reston, Virginia. 196 p
.
Tarlock
D.
(
2004
).
First look at a modern legal regime for a post-modern United States army corps of engineers
.
Kansas Law Review
52
,
1285
1325
.
Thorson
J.
(
1994
).
River of Promise, River of Peril: The Politics of Managing the Missouri River
.
University Press of Kansas
,
Lawrence
.
US Army Corps of Engineers, Northwestern District
(
2012
).
Missouri River Mainstem Reservoir System Post 2011 Flood Event Analysis of Missouri River Mainstem Flood Control Storage
. .
US Army Corps of Engineers, Northwestern District
(
2014
).
Missouri River Mainstem Reservoir System Missouri River Mainstem Reservoir System Summary of Actual 2013 Regulation
.
Available at. http://www.nwd-mr.usace.army.mil/rcc/aop.html (accessed October 2015)
.
US Army Corps of Engineers, Omaha District
(
2000
).
Missouri River Main Stem Reservoirs Summary of Actual 1998–1999 Operations
. .
US Army Corps of Engineers, Omaha District
(
2006
).
Missouri River Mainstem Reservoir System Master Water Control Manual Missouri River Basin
. .
US Army Corps of Engineers, Omaha District
(
2008
).
Missouri River Recovery Program: Missouri River Recovery Implementation Committee. Paper 80
. .
US Army Corps of Engineers, Omaha District
(
2010
).
Garrison Dam/Lake Sakakawea Draft Surplus Water Report
.
http://www.swc.nd.gov/ (accessed August 2014)
.
US Army Corps of Engineers, Omaha District
(
2013
).
Review Plan: Missouri River Municipal and Industrial Water Supply Reallocation Report and Environmental Impact Statement
. .
US Environmental Protection Agency Region 7
(
2014
).
The Missouri River Mainstem
. .
US Fish & Wildlife Service
(
undated
).
Digest of Federal Resource Laws of Interest to the U.S. Fish and Wildlife Service: Interstate Compacts
. .
US Government Accountability Office
(
2007
).
Interstate Compacts: An Overview of the Structure and Governance of Environment and Natural Resource Compacts. Report GAO-07-519
.
(accessed July 2014)
.
US Government Accountability Office
(
2009
).
Missouri River Navigation: Data on Commodity Shipments for Four States Served by the Missouri River and Two States Served by Both the Missouri and Mississippi Rivers. Report GAO-09-224R
. .
US Institute for Environmental Conflict Resolution
(
2015
).
About Missouri River Basin Interagency Roundtable
. .
Veeder
W.
(
1972
).
Indian water rights in the Upper Missouri Basin
.
North Dakota Law Review
48
,
617
637
.
Vineyard
J. D.
(
1997
).
Missouri State Water Plan Series Volume VI, Water Resource Sharing – The Realities of Interstate Rivers, Missouri Department of Natural Resources’ Division of Geology and Land Survey
,
Water Resources Report No. 50
, p.
76
.
Zellmer
S.
(
2011
).
Mudslinging on the Missouri: Can endangered species survive the Clean Water Act?
Drake Journal of Agricultural Law 16
,
89
115
.
Available from http://digitalcommons.unl.edu/lawfacpub/112 (accessed October 2015)
.