Abstract
The combined trends of urbanization and climate change elevate the criticality of planning resilience against natural hazards in urban areas, but quantifying key metrics and responses for planning purposes has proven elusive. A key need for cutting-edge quantitative research in urban resilience is datasets that are large and detailed but also broadly applicable to different settings and questions. The goal of this paper is to introduce an infrequently used dataset in the risk science field-the U.S. Census Bureau's American Housing Survey (AHS). We evaluate the pros and cons of the AHS and highlight specific examples of how researchers can use the dataset to advance goals toward more resilient urban futures. Using predictive models and multiple dimension reduction, we evaluate spatial and temporal trends within and between cities, and we investigate relationships between often disparate variables (e.g., socio-demographic and built environment) for anticipating negative consequences of hazards and perceived risk. In so doing, we demonstrate the usefulness of AHS's longitudinal nature, cross-city data, varying climatic regions, and broad range of variables for risk scientists.