PCA-based approach for mapping social vulnerability to hazards in the Chennai metropolitan area, east coast of India

Abstract

ABSTRACTSocial vulnerability shows the lack of capacities of a person or groups across space and time to prepare for, respond to, and recover from the impact of natural hazards. It involves a combination of socioeconomic and demographic factors that determine the degree to which a (human) system is susceptible to, or unable to cope with, the adverse effects of a disastrous event. Social Vulnerability Index (SoVI) is an effective tool to measure the social vulnerability of an area. Though SoVI has successfully applied in many different contexts and places for socioeconomic development and disaster risk reduction, most societies still lack awareness of how social differences within their population play a role during disastrous events. To address this gap, the present study aims to map the social vulnerability and identify the locations of a socially vulnerable community in the Chennai Metropolitan Area (CMA) through an inductive approach (e.g. factor analysis) using demographic and built-environment data in ArcGIS and SPSS environment. We analysed twenty-three individual variables from five different vulnerability components, such as population, housing, economics, healthcare service, and exposed elements using Principal Component Analysis, that reduced to a smaller set of multidimensional components that explained 71.2% of the total variance and calculated the final SoVI score by adding all five-factor scores. The resultant SoVI map identifies the most vulnerable areas in the highly populated and tightly packed residential areas of Chennai city and the least vulnerable areas on the outskirts of Chennai city. The constructed SoVI could assist planners and policymakers at the national, state, and local government level in making appropriate decisions at all phases of the disaster management cycle and help prioritize the implementation of Government welfare schemes.

Keywords

• Social Vulnerability Index (SoVI)
• Descriptive statistics
• Principal component analysis (PCA)
• Hazards
• and geographic information system (GIS)