README – U.S. Climate Risk Dataset (IPCC AR6 Framework)

This dataset provides a consistent and easy-to-use foundation for climate change and natural hazard risk assessments across the contiguous United States (CONUS). It is structured around the IPCC Sixth Assessment Report (AR6) understanding of risk, which considers four key components: Hazard, Exposure, Vulnerability, and Resilience.

What’s Included:
130 variables, processed and ready for analysis,
Available at both county (3,108) and census tract (83,481) levels,
In CSV format. 
The dataset can support research and decision-making for 18 natural hazard types, including hurricanes, flooding, wildfire, drought, extreme heat, coastal hazards, severe storms, and others.

2. How the Data Is Organized:
All variables are grouped into four categories that align directly with the IPCC AR6 risk components:
Hazard: Measures related to the presence or likelihood of natural hazards
Exposure: Populations, buildings, infrastructure, and land features that may be affected
Vulnerability: Social and economic factors that influence how severely communities are impacted
Resilience: Community capacity to prepare for, respond to, recover from, and adapt to hazards

You can choose to work at either the county or tract scale depending on your project needs.
A separate reference file (keyword.xlsx) is included, listing for each variable:
Category, Variable Name, Description, Keyword, and Data Source

3. Data Sources:
The dataset brings together information from several sources. The primary contributors include but not limited to:
Baseline Resilience Indicators for Communities (BRIC) – University of South Carolina
National Risk Index (NRI) – FEMA
Social Vulnerability Index (SVI) – CDC
Additional publicly available datasets were added to provide stronger coverage of exposure-related variables.

4. Data Processing:
To make the variables comparable and ready for index construction, two key steps were applied:
Winsorization at the 5th and 95th percentiles, to limit the influence of extreme outliers
Fuzzy membership normalization, which converts all variables to a consistent 0–1 scale
These steps preserve the variability in the data while making it easier to combine different variables in index development, spatial modeling, or statistical analyses.

5. Intended Use:
The dataset is designed to be “plug-and-play” for climate and hazard risk assessments. Common applications include:
Building composite climate risk indices
Mapping multi-hazard risk in GIS
Socio-environmental vulnerability and resilience research
Supporting adaptation, planning, and policy studies

Contact:
Questions or feedback are welcome.
Email egecankurter@gmail.com