Limitations and considerations of using composite indicators to measure vulnerability to natural hazards.

Vulnerability assessment plays a critical role in disaster management and requires the consideration of multiple dimensions that involve both the built and social environments. A common approach to address this problem is the use of composite indicators, which offer a simplified method to combine information across different dimensions and facilitate decision making. However, composite indicators present limitations in the context of hazard vulnerability. This study investigates the source of these limitations and provides ways to overcome shortcomings in the interpretation of composite vulnerability indicators. To conduct this investigation, a composite indicator is developed to assess the vulnerability of power and transportation infrastructure, while considering social vulnerability, to capture community hazard vulnerability. Using a case study of Harris County in Texas, we investigate the disparities in outcomes resulting from different calculation methods, such as sub-indicator weighting. The case study shows that the value of the indicator is not consistent across different calculation methods. Additionally, weighting the sub-indicators plays an important role in the value of the indicator. Combining infrastructure and social factors is found to be misleading in the interpretation of hazard vulnerability, and the use of bivariate maps is proposed to better distinguish between infrastructure and social vulnerabilities.