Authors: Masood Ali Khan – University of Vermont
Title: Emergency Managers’ Practices and Social Vulnerability Trends in U.S. Flood-Affected Counties: A Decade-Long Mixed-Methods Study (2014–2024)
Presentation Type: Poster
Abstract: This study assesses the longitudinal impact of flooding on social vulnerability across 2,357 U.S. counties between 2014 and 2024, using fixed-effects panel regression. The Social Vulnerability Index (SVI), ranging from 0 (least vulnerable) to 1 (most vulnerable), serves as the primary outcome measure.
Results indicate that floods significantly increase overall social vulnerability in both the first year (β = 0.536, p < 0.001) and the third year (β = 0.546, p < 0.001) following an event. However, effects vary by baseline vulnerability level. In counties with low initial vulnerability (SVI 0–0.25), vulnerability significantly rises in the second (β = 0.014, p = 0.041) and third years (β = 0.029, p = 0.021), suggesting delayed but growing risk. Conversely, counties with moderately high (SVI 0.5–0.75) and high vulnerability (SVI 0.75–1) experience significant declines in SVI in the second year after flooding (β = –0.036 and –0.039, respectively; p < 0.001), possibly due to aid influx or out-migration.
Model fit varied across subgroups, with the highest explanatory power in high-SVI counties (R² = 0.1066) and the overall model showing significance (F-statistic = 1,196,000; p < 0.001). These findings highlight the complex, time-sensitive nature of flood impacts on vulnerability, emphasizing the importance of differentiated resilience strategies based on local vulnerability baselines.