Examining Vulnerability to Tornadoes Using Census Tract-Level Demographic Data and Tornado Damage Survey Paths

Increasing resilience to tornadoes requires advancements in forecast techniques and technology, effective threat communication, additional knowledge of societal vulnerability, and much more. For emergency managers, knowledge about vulnerable populations within their community can help with planning for and responding to tornadoes. Advancements in geographic information systems can increase this knowledge via detailed mapping of demographic variables, land-use land-cover, and tornado impacts (e.g., National Weather Service damage surveys; fatality locations) at high spatial resolutions.

The purpose of this study is to examine demographic vulnerability to tornadoes by relating census tract-level demographic data to National Weather Service tornado damage assessments (i.e., tornado paths) and fatality locations with the ultimate goal of providing tornado vulnerability maps to emergency managers. To determine which demographic variables might be most related to fatalities, a linear regression model was used to analyze demographic data for 156 census tracts that intersected with the surveyed paths of 13 killer tornadoes responsible for 127 fatalities. Using the top variables determined by the model (e.g., population density, poverty density, mobile home density, etc.), census tract-level tornado vulnerability maps were produced based on spatial area impacted by the tornado and the number of fatalities predicted by a linear model. Land-use land-cover was also considered to quantify the amount of developed space impacted by each tornado and relate that to vulnerability and fatalities. The lead author will discuss key findings of this ongoing study.