Copula-based Multivariate Vulnerability Assessment of Coastal Systems in South Carolina

Abstract:

Comprehensive assessment of coastal systems vulnerability to natural hazard requires inclusion and integration of various information. Combining the biophysical factors with socio-economic vulnerability to examine the adaptive capacity of a society has gained serious attention for coastal hazards vulnerability assessment. The objective of this study is to improve the conventional vulnerability cost function by construction of bivariate copula based probabilistic function. The biophysical factors include two main groups of information, physical, such as elevation, distance from coast, and landuse, and hydroclimate, such as number of coastal hazard events, hurricane track density, surge height, rainfall intensity, and sea level rise. The socio-economic vulnerability consists of the number of historical and archeological structures, cost of fatalities, and social vulnerability index. The biophysical factors and socio-economic factors, as well as their join probability, are estimated for the six coastal counties of South Carolina based on 26 different Copula functions. The MvCAT tool is used to form the Copula functions, fine tune the Copula parameters using local optimization approach, and select the best fitted function using minimum Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). The bi-variate coastal vulnerability index (CVI) is then estimated based on the joint probability distribution between the biophysical vulnerability and socio-economic vulnerability of coastal hazards. Results suggest that most biophysically vulnerable places did not intersect or overlay with the socio-economically vulnerable locations. While a place might be jointly vulnerable for both biophysical and socio-economic classes, residents might maintain greater adaptive capacity associated with health insurances, financial capacity, etc. to absorb the adverse effects of a disaster. The results indicate that multi-variate based vulnerability analysis provide additional information needed for coastal managers to simultaneously view the biophysical and socio-economic impacts of hazards on coastal systems.