Search for: All records

Abstract. Regional hurricane risk is often assessed assuming a static housing inventory, yet a region's housing inventory changes continually. Failing to include changes in the built environment in hurricane risk modeling can substantially underestimate expected losses. This study uses publicly available data and a long short-term memory (LSTM) neural network model to forecast the annual number of housing units for each of 1000 individual counties in the southeastern United States over the next 20 years. When evaluated using testing data, the estimated number of housing units was almost always (97.3 % of the time), no more than 1 percentage point different than the observed number, predictive errors that are acceptable for most practical purposes. Comparisons suggest the LSTM outperforms the autoregressive integrated moving average (ARIMA) and simpler linear trend models. The housing unit projections can help facilitate a quantification of changes in future expected losses and other impacts caused by hurricanes. For example, this study finds that if a hurricane with characteristics similar to Hurricane Harvey were to impact southeastern Texas in 20 years, the residential property and flood losses would be nearly USD 4 billion (38 %) greater due to the expected increase of 1.3 million new housing units (41 %) in the region. 

Abstract We hypothesize that for disaster risk mitigation, many households, despite being aware of their risk and possible mitigation actions, never seriously consider doing anything about them. In mitigation-focused decisions, since there is no equivalent to warning messages, the decision process is likely to evolve over an extended time. We explore what activates hurricane mitigation protective action decisions through three research questions: (1) to what extent are homeowners unengaged in protective action decision making? (2) What homeowner characteristics are associated with lack of engagement? And (3) to what extent do different life events trigger engagement in the decision-making process? We use the Precaution Adoption Process Model to conceptualize engagement as distinct from decision making; the broader protective action decision-making literature to explore drivers of engagement; and Life Course Theory to examine potential transitions from unengaged to engaged. We use survey data of homeowners in North Carolina to examine these questions empirically. Findings suggest that one-third of respondents had never engaged in protective action decisions, that life experiences differ in their occurrence frequency and effect on households’ mitigation decisions, and that some events, such as renovating, reroofing, or purchasing a home may offer critical moments that could be leveraged to encourage greater engagement in mitigation decision making.