A Multiscale Method to Improve Social Vulnerability Estimation for Water Infrastructure System Planning and Management

In recent years, concerns have grown over the risks posed by climate change, droughts, floods, and other hazards to the most vulnerable populations. As a result, the quantification of socioecologic vulnerability has been increasingly considered in risk planning and management. However, current vulnerability quantification methods often suffer from a spatial attribution problem, in which the spatial scale of the information does not align with the spatial scale at which risk-reduction interventions are designed and implemented. This study seeks to address these limitations by proposing a new quantitative multiscale method for socioecologic vulnerability. Using the planning of green stormwater infrastructure (GSI; e.g., rain gardens, bioswales, green roofs) as a case study, this study explores the spatial variability of socioecologic vulnerability under the use of different dasymetric mapping techniques and methods for the quantification of the social susceptibility index. The proposed approach is applied to four cities in the United States. The results show that computing vulnerability at the Census Block scale induces less sensitivity to methodological choices and produces lower spatial variability in identifying the most vulnerable communities when compared to using Census Block Groups. Furthermore, using an absolute count with ordinal normalization produces the least variability in identifying spatial clusters of high vulnerability.