|Abstract:||Since 1992, Devils Lake in northeastern North Dakota has risen nearly 30 feet, destroying hundreds of homes, inundating thousands of acres of productive farmland, and costing more than $1 billion for road raises, levee construction, and other flood mitigation measures. In 2011, the lake level is expected to rise at least another 2 feet above the historical record set in 2010 (1,452.0 feet above the National Geodetic Vertical Datum of 1929), cresting less than 4 feet from the lake‘s natural spill elevation to the Sheyenne River (1,458.0 feet). In an effort to slow the rising lake and reduce the chance of an uncontrolled spill, the State of North Dakota is considering options to expand a previously constructed outlet from the west end of Devils Lake or construct a second outlet from East Devils Lake. Future outlet discharges from Devils Lake, when combined with downstream receiving waters, need to be in compliance with applicable Clean Water Act requirements. This study was completed by the U.S. Geological Survey, in cooperation with the North Dakota Department of Health Division of Water Quality, to evaluate the various outlet alternatives with respect to their effect on downstream water quality and their ability to control future lake levels.
A Devils Lake stochastic simulation model developed in previous studies was modified and combined with a downstream stochastic routing model developed for this study to simulate future (2011-30) Devils Lake levels and water quality, and outlet discharges, flows, and water quality (specifically, dissolved sulfate and total dissolved solids concentrations) for key downstream locations. Outlet alternatives include: (1) a 250 cubic feet per second west-end outlet (the current outlet) combined with a 250 cubic feet per second east-end outlet (W250E250); (2) a 350 cubic feet per second west-end outlet combined with a 250 cubic feet per second east-end outlet (W350E250); and (3) a 250 cubic feet per second west-end outlet combined with a 350 cubic feet per second east-end outlet (W250E350). In addition to satisfying current (2011) flow and water-quality requirements for the upper Sheyenne River, each of the outlet options was simulated with a less restrictive downstream sulfate constraint (750 milligrams per liter) and a more restrictive downstream sulfate constraint (650 milligrams per liter) for the outflows from Baldhill Dam. Thus, there were a total of six outlet scenarios (three outlet alternatives, each with the less restrictive and more restrictive downstream sulfate constraint). In addition, a baseline simulation in which there were no outlet discharges was used for comparison with the outlet simulations.
Simulation results indicate all six outlet scenarios substantially reduce, but do not eliminate, the chance of a spill. For the baseline simulation, the chance of a spill would be 0.6 percent this year (2011), about 14 percent by next year (2012), about 28 percent by 2015, and about 45 percent by 2030. The outlet scenarios reduce the chance of a spill to 0.2 percent this year, about 9 percent next year, 14 to 15 percent by 2015, and 17 to 19 percent by 2030. The chances of a spill are slightly less for the larger outlets (W350E250 and W250E350) compared with the smaller outlet (W250E250) and slightly greater for the more restrictive downstream sulfate constraint (650 milligrams per liter) compared with the less restrictive constraint (750 milligrams per liter). All of the outlet scenarios prevent most spills that would have occurred after 2015, but many of the spills that occur before 2015 are not prevented by any of the outlet scenarios.
All of the outlet scenarios are effective for drawing the lake down in future years, but the more restrictive downstream constraint results in slower drawdown compared with the less restrictive constraint. For the baseline condition, the chance the lake would be above 1,450.0 feet is 99 percent in 2015 and 38 percent in 2030. For the outlet scenarios w