Social Vulnerability and Hurricane Impact Modeling

Abstract

Numerical hurricane loss and impact prediction models have become widely used as tools to determine property damages. To yield loss estimates, models often combine building stock, economic data, and vulnerability functions in an effort to relate wind and surges to expected damages. Within this framework, vulnerability refers only to damage or economic loss to physical structures. Societal factors influencing building exposure and damage are disregarded although research shows that growing populations, increased wealth, and demographic shifts in high hazard areas are partially responsible for increased loss during the past 25 years. This paper addresses this missing societal component in hurricane loss modeling by examining two questions: (1) to what extent can a quantified measure of social vulnerability be incorporated into numerical hurricane impact modeling to improve loss prediction; and (2) which indicators of social vulnerability explain spatial differences in impacts from the wind and storm surge hazards associated with hurricanes? The social vulnerability index was coupled with a surface wind analysis, a storm surge inundation model, and a damage database along Mississippi's Gulf Coast in the aftermath of Hurricane Katrina. All components were input into a series of multiple regression analyses to explore the relationships between social vulnerability and hurricane impact. A relationship between social parameters, hurricane winds, storm surge inundation, and impact exists at both extensive and catastrophic levels of damage, and results reinforce the conclusions of earlier studies that illustrate the effects of social vulnerability on hurricane impacts.