By

Global flood damages may increase 20-fold by the end of the twenty-first century due to climate change and continued development in flood-prone areas (Winsemius et al. 2016). The UN estimates that 200 million people may be displaced due to the effects of climate change by 2050. Island nations stand to lose the most from these now seemingly inevitable changes, and countries such as Fiji, Kiribati, and the Maldives are already being heavily disrupted by rising sea levels, changing ecosystems, and more intense storms and flood events. For small coastal and riverine communities and settlements alike, wholesale relocation is being considered as a feasible alternative to seawalls, levees, and other structural forms of protection, and example of which is shown in Fig. 1. However, the effectiveness and feasibility of managed retreat has sometimes been debated. Proponents of managed retreat claim that it can offer a more environmentally friendly, long-term, and sometimes a more cost-effective approach to hazard mitigation than structural flood protection such as levees, dikes, weirs, and dams (Wenger 2015; David and Mayer 1984; Rose et al. 2007; NIBS 2018; Pinter 2021). Critics of managed retreat cite the social implications of displacing communities (Schnell and Haddock 2004; Dannenberg et al. 2019; Oliver-Smith 1991), along with the unequal distribution of federal property acquisition funds (Muñoz and Tate 2016; Howell and Elliott 2019). By quantifying metrics of success for completed on ongoing case studies of managed retreat, we may assess the practice as a whole, understand under what conditions it is most effective, and draft a roadmap to guide future implementation. For nine cases of riverine managed retreat completed in the US Midwest, three pragmatic and preliminary metrics were developed: (1) flood losses avoided, (2) population trends, and (3) trends in home sales prices. Flood-loss modeling shows that community relocation can be cost-effective, reducing economic flood exposure by over 95% (Fig. 2). Analysis of population data and residential real-estate sales shows that Midwestern floodplain towns have lagged non- floodplain towns, and homes on mapped floodplains have sold at a discount to non-floodplain homes since at least the early 1990s. Similar success can be found in cases of coastal managed retreat in both Japan and Fiji, but similar quantitative analysis for cases of coastal retreat have yet to be completed (Pinter et al., 2019).

James.Rees@colorado.edu

Geography Graduate Student, CU Boulder