Dynamics of Resilience over Time: A global empirical study on the community level for flood risks

A significant challenge of resilience measurement lies in taking a complex, multi-dimensional concept and operationalizing it in a concrete and measurable way. The next generation FRMC (Flood Resilience Measurement for Communities) framework and tool is providing such a measurement within a standardized approach (e.g. not dependent on the location it is applied to), and which therefore can be used across the globe. It is based on the Sustainable Livelihood Framework and includes 44 indicators called ‘sources of resilience’ that are distributed across and represent critical aspects of five complementary ‘capitals’ (5C). The sources are selected for the roles they play in helping people on their development path and/or providing capacity to withstand and respond to shocks. We present the dynamics of these resilience indicators over time based on a large-scale empirical assessment of communities across the globe that are exposed to flood risks. In more detail, resilience indicators are measured for a baseline as well as endline period (over 2-4 years) and in case of flood hazard events, a post-event analysis was performed to identify corresponding damages. The baseline survey involved 325 communities across 22 developing countries, with data collected from over 19,000 households as well as focus groups, key informants, and secondary sources. This survey represented a total community population of more than 1 million people and generated over 2.5 million data points from 14,300 graded sources. The endline survey engaged 280 communities of the 325 in 19 developing countries. Post-event surveys were conducted in 66 communities across 7 developing countries that have experienced a flood event. Lastly, an interventions survey analysed in which communities’ interventions were implemented. Based on a Confirmatory Factor analysis, Structural Equation modelling as well as a Boosted Regression Tree approach we found important differences in the dynamics of resilience over time which are not only dependent if hazard events have realized but also in regard to the resilience levels communities are starting from during the baseline period. Our empirical findings should therefore provide a better understanding about actual resilience trajectories that can take place and the important dimensions that may influence them over time.