Date of Award

January 2012

Document Type


Degree Name

Master of Science (MS)


Earth System Science & Policy

First Advisor

Rebecca J. Romsdahl


The Devils Lake flood is the longest, most expensive terminal lake flood in the history of the United States. In 1993, the Lake had a surface elevation of 433.9 m (1423.7 ft.) above mean sea level. Since that time it has risen 9.3 m (30.6 ft.), inundated 58,275 ha (144,000 acres) of land, and caused an estimated $1.6 billion (2011 USD) in total cumulative economic losses to the region. As a terminal lake in the Devils Lake sub-basin of the Red River Valley Basin, it has no natural outlet below 444.4 m (1458 ft.) and significantly poorer water quality than the rest of the basin waters. This impacts numerous downstream communities across state and national boundaries, making Devils Lake flooding a multi-definitional problem for policy makers on every political level. The economic, environmental, legal, and social ramifications of the State's response, combined with the Lake's unique hydrological features, systematic uncertainty, climactic fluctuation, and socio-political and technical complexity make Devils Lake an ideal case study of a Wicked Problem. Wicked Problems Theory is a subcategory of modern policy thought that is useful in assessing unique environmental and socio-political problems that lack a clear optimal solution or stopping point. This is indicative of the main Devils Lake solution, which consists of a three-pronged mitigation strategy known as Continuing Infrastructural Protection (CIP). Primarily an incremental infrastructural response to protect adjacent lake communities, CIP began in 1994, and has cost nearly $1.3 billion (2011 USD) since that time.

Although CIP was initially predicted to have a positive cost-benefit ratio, statistical data shows that the 18-year cost of CIP is greater than the value of all the property it was constructed to protect. In order to assess how policymakers and domain experts might have determined a more economically efficient solution, this thesis combines the economic concepts of expected and present value into an expected present value (EPV) model within a Wicked Problems framework. This model incorporates the lake level probability and the discount rate as variables, and produces the EPV of all future CIP costs from any point in time over the current course of flooding (1994-2011). Because the discount rate and lake level probability are unknown, the EPV of CIP was simulated under a range of potential discount rates and likely lake level probabilities and compared against the estimated cost of a one-time relocation and buyout of the adjacent Devils Lake communities. The model assumes that the threshold discount rate at which the relocation/buyout alternatives had an equivalent monetary value as CIP reflected the preference of decision-makers for CIP over other alternatives. The results suggested that policymakers preferred short-term solutions with smaller continuing costs over long-term solutions with large one-time costs, despite the fact that the long-term solution was ultimately cheaper in the long-run. Based on an examination of the relevant literature, governmental analysis, and anecdotal evidence, the analysis suggested that flood mitigation decisions were driven by a preference for the present over the future, the possible underestimation of long-term CIP costs, and the human tendency to place a lower priority on the elimination of extreme risks than their statistical probability implies is appropriate. When viewed as Wicked Problem, the results and theory support the conclusion that an `iterative risk-management framework', as described by the National Research Council, would have likely resulted in a more effective, resilient, and sustainable long-term flood mitigation response.