NSF Award
2427928
Understanding how people move and respond during natural disasters is crucial for improving disaster response strategies and building resilient communities. This project seeks to address gaps in current research by developing a comprehensive framework that uses advanced algorithms and large-scale data from anonymized mobile phones to model human behavior and movement during disasters. This innovative approach will identify potential vulnerabilities and forecast areas at greater risk, thereby enhancing disaster preparedness and response. The project aims to introduce three novel algorithms to better understand the interplay between human mobility and socioeconomic factors. Doing so will provide a detailed representation of how individuals and communities react to catastrophic events. This work supports the national interest by promoting scientific progress, advancing public health and welfare, and enhancing national security. The broader impacts of this project include guiding the creation of more effective and equitable disaster preparedness and response strategies, which are accessible to all community members, particularly vulnerable groups.<br/><br/>The project aims to advance the understanding of human mobility patterns under natural hazard impacts by developing and integrating three novel algorithms: a multidimensional Dynamic-Time-Warping Self-Organizing Map (mDTW-SOM) for clustering human mobility trajectories, a Weighted Greedy Gaussian Multivariate Segmentation (WGGS) for characterizing aggregated human mobility patterns at the political geographic unit level, and a Multinomial Geographically Weighted Elastic Net (M-GWEN) for selecting socioeconomic impactors under conditions of spatial non-stationarity. These algorithms will analyze large-scale spatiotemporal datasets to capture the intricate relationship between human mobility and socioeconomic factors, providing an in-depth analysis of human behavior during disasters. The project’s methodological innovations significantly advance disaster response analysis, offering crucial insights for enhancing disaster response strategies and developing predictive models of human behavior in crises. The findings will significantly contribute to the understanding of disaster resilience and response mechanisms, facilitating more refined analyses of population behaviors and movements during disasters.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
