Understanding Social Disaster Resilience Using GeographicallyWeighted Regression: A Case Study of Seoul Metropolitan Area

Abstract

Quantifying urban resilience under disaster events have been interested scholars through considerable period of time. Although many researches have tried to assess various dimensions of community resilience, it is still challenging to select significant attributes that can be used for the assessment model, especially for social resilience. This study proposes a practical assessment model of social resilience through the following steps: (1) examining appropriate variables that are considered to be related to disaster damage, (2) analyzing the impact of spatial heterogeneity of the social attributes by using Geographically Weighted Regression (GWR) method. A Geographic Information System (GIS) software, was used for the visualization of the assessment result. Through an experimental case study on Seoul Metropolitan Area, the author proposes meaningful variables and distinguishes the relationship between disaster damage and social resilience. 5 variables including population density, vulnerable age, disability, administrative service, and multi-cultural population are determined as related proxy variables to social resilience. Significant coefficients representing the influence of each variable to social resilience is derived in positive and negative values. The study looks forward to support the decision-making process of disaster management and to improve current disaster mitigation project planning.