A Comprehensive Geospatial Dataset for Risk Assessments of Natural Hazards in the United States

ABSTRACT

We present a homogeneous and spatially explicit dataset at county and tract levels that supports climate change risk assessment across the contiguous United States, designed to enable user‐defined construction of hazard, exposure, vulnerability, resilience, and composite risk indices through application‐specific weighting of variables and selection of hazards or future hazard projections. The dataset follows the Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report (AR6) risk framework, which defines risk through four key components: hazard, exposure, vulnerability, and resilience. It includes 130 variables compiled from multiple national sources for 3198 counties and 83,481 tracts. Major data sources include the Baseline Resilience Indicators for Communities (BRIC) from the University of South Carolina, the National Risk Index (NRI) from the Federal Emergency Management Agency (FEMA), and the Social Vulnerability Index (SVI) from the Centers for Disease Control and Prevention (CDC). Additional datasets were integrated to provide a more comprehensive set of exposure variables. To facilitate integration of variables with different units and scales, all data were winsorised at the 5th and 95th percentiles and normalized using fuzzy membership functions. The resulting normalized values represent relative positions within the distribution of each indicator and provide a consistent numerical framework for visualization, spatial analysis, and user‐defined risk modelling. This process reduces the influence of outliers while preserving data variability. The resulting dataset provides a consistent foundation for constructing composite risk indices, replicating or extending IPCC AR6‐aligned analyses, and supporting climate risk and adaptation research. It is particularly suited for applications in systemic risk assessment, spatial modelling, and policy evaluation.