Abstract:
Human-flood interaction models have the potential to help decision makers in flood-prone areas predict the impact of proposed development strategies and policies. In this talk, I will discuss a case study of operationalizing a human-flood interaction model for Biloxi, MS. Biloxi is a coastal community that has been impacted by and recovered from flooding throughout its history. About 25% percent of damaged areas didn't recover after Hurricane Katrina. Stakeholders from city government are weighing options for increasing resilience to flooding, such as changing base flood elevation maps and increasing green infrastructure. It is unclear if these strategies will lead to a sustainable population level. We adapted a minimalistic model of human-flood interactions to provide decision support to our stakeholders. Biloxi lacks physical flood protection structures, so we represented protection level as the base flood elevation plus required freeboard, and tested scenarios with more or less conservative approaches to updating flood elevation maps. We developed a spatially explicit version of the model to account for heterogeneity in flood hazard and community resilience. Flood hazard statistics were based on LISFLOOD 8.0 model outputs. To parameterize the socioeconomic model equations at the census block group level, we drew on a long-term monitoring dataset of post-Katrina recovery to develop a logistic regression model for redevelopment that considered social, economic, institutional, infrastructural, community-based, and environmental dimensions of resilience. Results from the stochastic model were highly dependent on the trajectory of flood events, with either substantial population decline or sustained coastal development possible in any scenario. The scenario that included frequent updates to flood elevation maps after flood events resulted in lower flood losses but lower expected long-term population in certain neighborhoods, demonstrating a tradeoff between the two strategies. I will also discuss lessons learned about operationalizing sociohydrologic models in the process of stakeholder engagement and model development. Future efforts will consider the role of nuisance flooding in decision making by property owners.
Bio:
Frances O’Donnell earned her Ph.D. in Civil and Environmental Engineering from Princeton University in 2013 and completed a postdoctoral fellowship in Earth Science and Environmental Sustainability at Northern Arizona University. She joined the faculty of Civil and Environmental Engineering at Auburn in 2016 as part of the Climate, Human and Earth System Science faculty cluster.
