This article is a case study of environmental decision making within a complex system, the Red River Basin (Basin). It explores the competing interests and policies that drive flood control efforts in the Basin of the upper Plains, taking as a point of departure the review process under the National Environmental Policy Act (NEPA) for a major flood control project involving the paired Red River cities of Fargo, North Dakota, and Moorhead, Minnesota (Fargo–Moorhead). Fargo–Moorhead, together with the U.S. Army Corps of Engineers (Corps), proposed a flood protection project that would include a 36-mile long, $1.7 billion diversion channel (the Fargo–Moorhead Diversion) to intercept flood flow upstream (south) of the two cities, conduct it in a semicircle to the west, and return it to the river channel at a point downstream (north) of the cities. The environmental impact statement process on the Fargo–Moorhead Diversion lays bare the multilayered and conflicting interests involved in decision making on Red River flood measures generally, and the Fargo–Moorhead Diversion in particular. It also reveals the inherent limitations of NEPA as an environmental decision tool for projects within complex systems of this character.
The 550-mile Red River of the North (the Red) drains an area of more than 50,000 square miles, from western Minnesota and eastern North Dakota northward to Lake Winnipeg in Canada. Its watershed includes the bed of an ancient glacial lake, an extensive shallow depression that fills seasonally with snow meltwater and, because of its topography, drains very slowly. A peculiarity of this river is its “reverse” south to north flow. This “reverse” flow amplifies flood stages (height) as water from its southerly end, where spring thawing occurs earlier, arrives downstream just as runoff from later melting further north reaches the river. See Charles Anderson & Al Kean, Red River Basin Flood Damage Reduction Framework (2004), at 2.
Frequent and catastrophic floods of the Red and its tributaries, with damage totals in the tens or hundreds of millions of dollars, are a fact of life for communities, large and small, in the Basin, from Breckenridge in southwestern Minnesota to Winnipeg, Canada. Climate change in the region has increased wintertime precipitation, with the result that while 48 of the last 109 years have seen the Red exceed its National Weather Service flood stage, such flood stage has been exceeded in every year from 1993 through 2011. Dan Gunderson, Fargo Moorhead Flood Diversion Moves Forward MPR News, Oct. 26, 2011.
There is general scientific consensus on the natural and human causes of flooding in the Basin. In addition to its unusual topography and hydrology, wetland drainage over the years has reduced the Basin’s natural flood storage capacity. Structural flood control projects, such as levees or channelization, on the river itself may concentrate flood flows, compounding flood effects for downstream neighbors. Luther Aadland et al., Stream Restoration for Flood Damage Reduction in the Red River (Working Paper No. 5, 1998). There is much less consensus on solutions.
Responsibility for flood management on the U.S. portion of the Red is divided between the federal government, notably the Corps, the states of North Dakota and Minnesota, and a number of agencies in both states, counties, and numerous individual cities and towns. In addition, special districts, called watershed districts in Minnesota and water resource districts in North Dakota, manage water bodies and watercourses, and permit and implement local flood control and other water-related projects. These districts (water districts) have limited taxing powers for water projects and other activities within their jurisdiction. Eight Minnesota water districts in the Basin work to implement local flood damage reduction projects under a joint powers agreement authorized under Minnesota law. Minn. Stat., § 471.59. This agreement creates the Red River Watershed Management Board (RRWMB), known locally as the “Red Board,” which serves as an umbrella planning, development, and funding organization. The equivalent entity in North Dakota is the Red River Joint Water Resource District (RRJWRD), formed under North Dakota’s joint powers law and composed of ten counties and four water districts in that state.
Consultative bodies and commissions have proliferated in response to Basin flooding. The Red River Basin Board is an international body with representatives from North Dakota, Minnesota, and Manitoba. It maintains close liaison with the Red River Water Resources Council, an intergovernmental entity formed by North Dakota, Minnesota, and Manitoba. The Red River Basin Commission, a grassroots watershed-based nonprofit organization of flood management professionals from the United States and Canada, is a research entity funded by both North Dakota and Minnesota. Its draft report, Finding Long Term Flood Solutions Together— Report to State and Federal Officials on a Comprehensive Plan of Flood Strategies for the Basin of the Red River of the North 2009-2011, was released November 11, 2011.
A recent addition is the Red River Retention Authority (RRRA). This is a new joint powers entity formed between individual water districts in both North Dakota and Minnesota. The RRRA, established in late 2010, has as its principal purpose expediting local retention projects, through the development of a memorandum of understanding with the Corps regarding project permitting under Section 404 of the Clean Water Act (CWA), and the securing of federal funds (e.g., through the Natural Resources Conservation Service of the U.S. Department of Agriculture).
At the municipal level, flood control has centered on local flood protection projects as opposed to flood flow reduction or prevention. These projects are typically dikes, levees, and diversion channels directed to the flood threat to a specific city or town. Unlike many flood retention or nonstructural flood reduction projects, such as wetland restoration, flood protection projects are immediately visible in (and thus politically appealing to) the affected communities. Municipal flood management projects reflect an ethos of self-reliance in the region: each town takes responsibility for its own flooding, with neighborly cooperation in sandbagging flood fights an established tradition. Notwithstanding such individualism, municipalities along the Red do not hesitate to call on the Corps for assistance with larger projects, such as the diversion and levees at Breckenridge, Minnesota (federal share $26 million), Wahpeton, North Dakota (federal share $12 million), and Grand Forks, North Dakota, and East Grand Forks, Minnesota (total cost $412 million). See U.S. Army Corps of Engineers St. Paul District, Flood Damage Reduction (Sept. 1, 2011). The most recent high-profile example is the Fargo–Moorhead Diversion itself.
In contrast, the flood control projects of water districts are largely prevention measures, such as water storage or retention and natural resource enhancements such as riverine and wetland habitat restoration. Originally formed primarily for drainage and watercourse control, water districts have evolved into a broad natural resource management role. Because they have permitting authority over a range of projects, including agricultural drainage, they are subject to local pressures (e.g., from farmers and others opposed to actions seen as inimical to agriculture), such as wetland restoration or temporary flood storage. Nonetheless, water districts possess significant technical and leadership resources and have produced detailed engineering proposals for flood flow reduction Basin-wide, such as the “Twenty Percent Flow Reduction Strategy” of the Red River Watershed Management Board, developed in coordination with the Red River Basin Commission.
Water districts have a cooperative but wary relationship with the Corps, whose function may be perceived as competing with local control of projects. Historically, water district projects received little Corps oversight beyond routine CWA Section 404 permitting. But in the early 1990s, concern about cumulative impacts from such projects in the Red River Basin prompted the Corps and the Minnesota Department of Natural Resources (DNR) to initiate a joint Environmental Impact Statement (EIS) under Minnesota’s version of NEPA, the Minnesota Environmental Policy Act (MEPA). Minn. Stat. Ch.116D. In January 1993, the Corps put CWA permit actions on Minnesota water district projects on hold pending completion of the EIS. In response, the RRWMB brought a state court action against the Corps and DNR challenging their EIS. Project permits were gridlocked.
The Minnesota legislature intervened, authorizing funding for a mediated settlement reached in 1998. An ad hoc stakeholders workgroup, the Minnesota Flood Damage Reduction Work Group (FDRWG), was formed with representation from the water districts, the Corps, the Minnesota DNR, environmental and conservation nonprofits, and others, but, significantly, no Basin cities. The mediated settlement document incorporates a dozen agreed upon flood damage reduction principles, among which are the storage or management of water as close as feasible to where it falls and the use of a systems approach to manage the timing of flow contributions from multiple minor watersheds. FDRWG discussions on implementation of the mediated settlement continue today.
In addition to local, state, and federal agency stakeholders, flood control on the Red has other constituencies. One is academia, exemplified by an extensive study, the so-called “waffle” plan modeled by the Energy & Environmental Research Center of the University of North Dakota. This would be a system of temporary flood storage using primarily existing ditches and farm fields bounded by roads, with a network of connecting channels and control structures. Dean A. Bangsund et al., Benefit-cost Analysis of the Waffle®: Initial Assessment (July 2008). Its authors conclude that “[i]n the case of Fargo/Moorhead, the anticipated crest height reductions [in a major flood] appear to be substantial,” and that overall, the “waffle” appears capable of generating $200 million to $600 million in net benefits over a 50-year period. While the “waffle” proposal is regarded by many as utopian and infeasible to implement on the scale proposed, it has influenced policy thinking about distributed flood retention.
Another constituency is Canada. Manitoba’s downstream position makes it the beneficiary of the United States’ obligations under the 1909 Boundary Waters Treaty, implemented by an International Joint Commission (IJC) established under Article VII of the Treaty. Canadian concerns, heightened after record flood levels in 1997 and 2009, are documented in an IJC publication, How Are We Living with the Red? Corps of Engineers’ modeling of the effects of its initial design of the Fargo–Moorhead Diversion, discussed below, showing flood stage increases as far north as the Canadian–U.S. border, may have played a significant (if unacknowledged) role in the project’s redesign.
The Fargo–Moorhead Diversion Project
In 2008, the Corps began a feasibility study of a flood protection project for Fargo and Moorhead. In May 2010, the Corps produced a draft Integrated Feasibility Report and Environmental Impact Statement, constituting a NEPA draft environmental impact statement (DEIS). While acknowledging the regional scope and cause of flooding at Fargo–Moorhead, the DEIS framed the NEPA project purpose and need as “reduc[tion] of flood risk, flood damages and flood protection costs related to the flooding in the Fargo–Moorhead Metropolitan Area.” In keeping with the Corps’ institutional focus on flood protection, the Corps of Engineers’ narrow formulation of the project purpose responded to the needs of its local development partners, the two sponsoring communities.
CEQ NEPA regulations, 40 CFR §1502.23, authorize but do not mandate agency cost-benefit analysis as part of its NEPA compliance. The Corps applies such analysis through its 1983 Economic and Environmental Principles and Guidelines for Water and Related Land Resources Implementation Studies (Study Principles). This document directs the development of a project plan that best “maximizes net national economic development benefits, consistent with the Federal objective.” Study Principles, at iv. In addition to such a plan, known as the NED Plan, the Corps is to prepare, if necessary, other plans that “reduce net NED benefits in order to further address other Federal, State, local and international concerns.” Id.
Accordingly, the Corps initially screened five alternatives as separate stand-alone plans (i.e., plans that would be completely effective in and of themselves). Among these alternatives were two forms of upstream flood flow reduction: distributed storage in retention ponds and the “waffle” (described above). Both were eliminated from the final alternatives analysis in the DEIS because the Corps believed they would be less physically effective and less cost-effective, as measured by the NED cost-benefit standard, than three final diversion channel alternatives.
The Corps’ alternatives analysis in the DEIS itself included no retention or other flood reduction measures. It rejected the possibility that such measures could be included as project components, as opposed to stand-alone alternatives. It made no mention of the RRWMB “Twenty Percent Flow Reduction Strategy” study. Overall, the DEIS proceeded on the implicit premise that the alternative selected would be implemented entirely within the project area, metropolitan Fargo–Moorhead.
The 2010 Corps proposal consisted only of a 36-mile flood diversion channel, sized to handle the major portion of the flow of a 100-year flood and to reduce the flood stage at Fargo–Moorhead in a 500-year flood to 36 feet, a reduction in flood risk deemed tolerable by the two sponsoring cities. Subsequent to the publication of its DEIS in April 2010, the Corps remodeled the flood stage impacts produced by the project downstream (north) of Fargo–Moorhead. Contrary to earlier results, this later modeling showed substantial additional flood stage increases downstream as far north as the Canadian border (and potentially beyond) attributable to the project, as water from the diversion channel rejoined the main stem of the Red.
In response, the Corps redesigned the project, incorporating a new and substantial flood staging/retention component immediately upstream of Fargo–Moorhead, consisting of a 50,000 acre-foot storage basin (4,360 acres) behind a nearly 12-mile embankment and an adjacent 150,000 acre-foot floodplain staging area. In April 2011, the Corps published a Supplemental Environmental Impact Statement (SEIS) for this revised project. The proposal, now consisting of the storage basin, the staging area, and the diversion channel, is more complex than the 2010 channel-only project, not only in design but in operation, since Fargo–Moorhead, as the operator, would now have, in each flood event, a range of options: flood storage or diversion of incoming flood water, or both, either simultaneously or sequentially.
Tellingly, the Corps’ SEIS did not justify the addition of the upstream retention/staging component in terms of flood flow reduction at Fargo–Moorhead itself, which would of course experience less flood flow on account of it, but as a means to reduce flood stages downstream of Fargo–Moorhead. Just as it had in the 2010 DEIS, the Corps did not view upstream stage reduction as a component of the project itself. Rather, and in accordance with CEQ NEPA regulations, it characterized the retention/staging area as a mitigation measure. (Despite the evident impacts at the Canadian border from the 2010 proposal, no mention was made in the 2011 SEIS of potential Boundary Waters Treaty problems.) The diversion channel itself remains, in the Corps’ eyes, the “project,” albeit now reduced in size on account of the lower flood flow it would carry. The Corps’ NEPA alternatives analysis in the 2011 SEIS is virtually identical to that of the 2010 DEIS: little more than a ranking of different locations for the diversion channel, each essentially an iteration of the same design idea. If the project is built, however, the operation of the upstream retention/staging areas, the largest on the Red, may serve an educational function as a dramatic and visible demonstration of flood flow reduction through upstream retention.
Reaction to the Fargo–Moorhead Diversion
Local reactions to the Fargo–Moorhead Diversion were mixed. The cities of Fargo and Moorhead were, of course, strongly supportive. So too was the adjoining suburb of West Fargo, until it became evident that the channel location would block that city’s western expansion, at which point it became an opponent. Communities and water districts outside the Fargo–Moorhead metro, largely excluded from a “seat at the table” in the planning process, remained either neutral or opposed.
When the original DEIS was published in 2010, communities upstream and downstream of Fargo–Moorhead were cautiously united in opposition. There was an understandable “us versus them” feeling that the project devoted massive federal resources to a permanent flood solution for Fargo–Moorhead while leaving other communities to shift for themselves. But this tenuous unity was fractured by the publication of the SEIS for the revised 2011 Fargo–Moorhead Diversion proposal. Upstream interests, opposed to the new retention/staging component on their doorstep, were now reluctant to make common cause with downstream communities, whom they saw as beneficiaries of the project’s redesign, however justified the Corps’ reason for it. As of November 2011, the comment period on the final Corps EIS has passed, although no record of decision has yet been issued; whether the required congressional appropriation for the project will be forthcoming remains an open question.
One impact from the Fargo–Moorhead Diversion is already being felt. The diversion controversy has sparked increased efforts for local retention projects. The newly formed RRRA is supporting an initiative by Minnesota Democratic Representative Collin Peterson, the former chair and now ranking minority member of the House Agriculture Committee, to include a 10-year, $500 million funding program in the 2012 Farm Bill for regional projects. The goal would be to create one million acre-feet of distributed water retention in the Basin through storage basins, wetland restoration, and temporary flood detention. The RRRA would oversee the local expenditure of the federal funds.
Observations from the Diversion Controversy
Reaction to the Fargo–Moorhead Diversion illustrates the political tension between the constituencies for larger, urban, Corps-backed projects, on the one hand, and for smaller, more rural, distributed retention, or other prevention projects, on the other. The two constituencies are supported by two very different federal agencies, the Corps and the U. S. Department of Agriculture (USDA). The Corps’ flood control mission is carried out through large construction projects. Its command structure is hierarchical, centered in Washington. The USDA, whose subsidy programs fund many water district activities, is rural and agricultural in focus. Its decision making is largely local. USDA is among the most important federal agencies for elected federal officials from farm states. Representative Peterson’s recent distributed storage proposal is easily understood in this context.
But extensive development of distributed storage or other retention over the entire Basin would be a major, long-term undertaking. To achieve flood stage reductions at Fargo–Moorhead comparable to those achievable by the Fargo–Moorhead Diversion would require distributed upstream measures implemented over a wide area. The problem is compounded by the fact that no two flood events are the same. Depending on precipitation and weather patterns, flood flows in any particular year may originate disproportionately from some tributary sources than from others. This in turn requires that to protect a given location, upstream storage/retention capacity, Basin-wide, must be great enough to handle flood flows from any upstream direction.
The Fargo–Moorhead Diversion also points up the limitations of NEPA as a decision tool on large flood projects. NEPA is directed exclusively to project impacts. However, flood control project planning is as concerned with the causes of the conditions creating the project as with project impacts themselves. Control of causes may be largely beyond the scope of an individual project. In addition, to the extent that cost-benefit considerations are required in the NEPA alternatives analysis, nonstructural or distributed preventive measures, by their nature, may be shown to be uncompetitively costly or to have less immediate flood stage reduction at the sponsoring municipality than local structural measures. This is particularly true if such measures are required—as they are in the Corps’ approach—to stand or fall on their merits as entirely sufficient solutions (i.e., as stand-alone measures, rather than as components of other actions).
Viewed from the wider, policy end of the telescope, distributed storage/retention measures should be part of the comprehensive Basin-wide approach so often urged by consultative and advisory bodies. Indeed, the Corps’ own Planning Guidance Notebook, Appendix H, directs that flood project planning be “conducted in a systems/watershed context.” But from the perspective of the project needs of the sponsoring community, cost-benefit considerations may lead to consideration of only local structural measures. As a practical matter, as flood protection is currently planned and executed in the Basin, the range of implementable flood solutions for a project on the scale of the Fargo–Moorhead Diversion may, in fact, be nearly as constricted as Corps’ alternatives analysis for that project.
On the other hand, a Basin-wide planning approach, with Basin-focused implementation at the requisite scale, might produce a different outcome. If flood protection projects were planned and implemented on the basis of regional, not purely local, cost-benefit ratios, individual flood control projects like the Fargo–Moorhead Diversion might look very different. While there would likely be a need for structural protective measures locally, the scale and cost of these measures would be lessened by the existence of upstream storage or retention capacity already in place. Such an approach would not eliminate individual project cost-benefit analysis, but would broaden the range of specific project options. With less flood stage to address, a broader range of local flood control designs would “pencil out” in a given case.
A New Federal Flood-control Agency for the Red River Basin?
Would it be possible implement flood protection on the Red with each project measured by benefit and impact to the population and natural resources of the Basin as a whole? Such a Basin-wide program would require extensive flood protection, reduction, and prevention measures, cross multiple jurisdictional boundaries, and pose coordination challenges in both construction and operation. A new, most likely federal, umbrella agency would be required. Rather than a development agency, like the Depression-era Tennessee Valley Authority, it could be a planning and regulatory agency with the power to set standards and goals (e.g., for flood stage reductions in the 10-year, 100-year, and 500-year flood), as well as to provide project permitting and funding assistance to existing implementing agencies whose project responsibilities would otherwise remain unchanged.
Numerous technical and economic issues would have to be addressed. How would Basin-wide flood stage standards be set? How would projects be selected? Something like the Basin-wide approach required of state water quality standard-setting under Section 319 of the Clean Water Act could be employed; each Basin state would be responsible for developing flood stage reduction goals and implementing projects for the Basin area within its borders, but ultimate approval would rest with the federal agency. Projects would be selected for their benefit Basin-wide. This would affect project types and sizes: A larger project, while more costly and only marginally more advantageous locally, might be more efficient and economically beneficial when viewed at a wider (e.g., Basin) scale.
A final challenge would be operational. How would all the various component facilities be managed during a flood? A decision system would be required to determine, for example, what retention or storage basins would be filled or released, and in what sequence. A mechanism would be needed to equalize flood effects, to the extent feasible. Compensation might have to be paid to private parties for temporary flood easements, or the equivalent. The authors of the “waffle” plan have explored the flood easement/compensation issue in depth.
Ultimately, some communities along the Red may not survive without an effective Basin-wide flood solution. Smaller towns, some going back to the nineteenth century, have largely depopulated and are at risk of disappearing entirely as their aging and economically disadvantaged populations are increasingly unable to bear the burden of periodic flood fights and unending repair to flood damaged roads and infrastructure. Recently, newspapers reported the planned closing of the U.S. post offices in two small towns (Hendrum, Minnesota, population 350, and Perley, Minnesota, population 105).
Climate change in the Upper Plains has already altered weather patterns, accelerating social and economic transformation in the region. Water management in the Red River Basin, including the governmental, legal, and political framework of flood control, must adapt to keep pace.