Fostering Collective Disaster Resilience via Cross-city Learning of Post-disaster Mobility Dynamics and Collaborative Data Governance

As climate changes rapidly, many cities and communities are facing disaster events at unprecedented scales. The lack of prior data poses significant challenges in learning from and building urban simulation models that forecast both short- and long-term socio-economic impacts. More specifically, human behavior data critical for urban planning and disaster preparedness is often not accessible, interoperable, or re-usable.

To fill the data gap, this project supports research on foundational computational methods to improve forecasting impacts by leveraging advances in language models, with results intending to enhance disaster preparedness and response strategies. Universality and heterogeneity of post-disaster mobility behavior are examined from over 150 disaster events in the US.

This approach contributes to the need for innovation beyond traditional data sources to enhance disaster resilience and climate change adaptation.

Research completed in association with this project looks to develop a data governance mechanism - a data commons platform - that offers inclusive access to and transparent re-usage of synthetic human behavior data under various disaster scenarios. This platform facilitates collaboration between policymakers, researchers, and citizen groups as well as convergent activities across scientific disciplines, including urban science, infrastructure engineering, economics, public health, and social sciences.

The policy relevance lies in its potential to transform disaster preparedness and response strategies through informed, data-driven decision-making leveraging non-traditional data. Furthermore, the public-facing platform enables stakeholders to test various mitigation approaches, fostering a culture of proactive planning and actions.

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.