An index of social fabric for assessing community vulnerability to natural hazards: Model development and analysis of uncertainty and sensitivity

Indices of socio-environmental vulnerability, resilience, and sustainability are well-established in assessing, mitigating, and planning for natural hazards in mitigation. Although there are many such indices, we still lack an index to measure connections and cohesion within a community-the social fabric (as we refer to it) that connects its members with each other and with the place. We propose an indicator-based index social fabric index (SoFI) to both measure and map the spatial dimension of the social fabric within and across communities. We investigate approaches to constructing social fabric indices and conduct uncertainty and sensitivity analyses for each stage of constructing the underlying model of the social fabric, which includes indicator transformation, normalization, principal component selection and rotation, and weighting. From this effort, we find that the precision of the index increases with social fabric levels, corresponding to lower social vulnerability. Global sensitivity analysis shows that the coordinate transformation and PCA selection play especially important roles in determining total uncertainty. Furthermore, the preceding step tends to absorb the uncertainty contributions of the following steps through construction interactions. Finally, we consider the importance of the emotional and psychological effects of the social fabric concept and provide suggestions for future work so that we will be able to develop a more transparent, comprehensive, and robust social fabric index model.