Lesson learned from adaptation response to Devils Lake flooding in North Dakota, USA

The water level of Devils Lake, a terminal lake in the northeast of North Dakota, U.S., has risen nearly 10 m since 1993, inundating farm land and causing significant damages to the infrastructure and surrounding communities. The currently adopted flood adaptation policy is continuing infrastructural protection (CIP), under which the infrastructure, such as levees or roads, is raised gradually in response to the rising water. Since the Devils Lake city and the adjacent communities are built within the historical confines of the lake, we introduce and estimate potential costs of two diametrical, but long-term targeted adaptation strategies involving either buyout or relocation (B/R) to contrast against the CIP. The comparison shows that B/R, while more expensive initially, would be cheaper than CIP accumulated over longer term. We further explore the reason that policymakers prefer CIP: the near-term uncertainty regarding the future lake levels has led to significant discount of long-term potential risk. With future risk significantly discounted, not only B/R, but any other alternatives that feature long-term perspective, could become less attractive than CIP. Were we able to predict the flooding more accurately or willing to discount future less, we might have chosen a different plan of action. The ultimate choice of CIP among the possible adaptation options reveals that policymakers have privileged the present and favored near-term remedies instead of long-term solutions. Although certainly not on the same scale, the lesson we have learned in dealing with Devils Lake has far-reaching implications on climate change adaptation strategies. The passive decision to wait and ignore the full potential for disaster ultimately leaves us with no other option, but to continue to wait and see and then spend more.