Evaluating the long-term resilience of water resources systems: Application of a generalized grade-based combination approach

Abstract

Abstract This paper introduces a new methodology for evaluating water resource systems resilience under extreme events by taking advantage of a generalized grade-based technique to combine criteria. To determine extreme events, we evaluate the impacts of climate change scenarios using the Soil and Water Assessment Tool (SWAT). The proposed framework for evaluating the long-term resilience of water resources systems has three main steps. At first, three performance criteria (i.e., reliability, traditional resiliency, and vulnerability (RRV)), which show the resilience against crossing a performance threshold, are calculated considering the optimal allocated water to demand nodes. In the second step, a new resilience-based criterion is proposed considering five sub-criteria: robustness, redundancy, resourcefulness, rapidity, and durability (4R-D). In addition, a new social resilience-based criterion (i.e., SR) is proposed, and these two criteria are considered to evaluate the total resilience of the system. In the third step, some adaptive management scenarios are defined to improve the resilience of the system. The proposed methodology has been applied to the Zarrinehrud river basin, the main water supplier of the largest hypersaline lake in the Middle East, Lake Urmia. Regarding the qualitative and quantitative nature of the criteria, the generalized combination (GC) rule for the evidential reasoning approach (ER) is used to evaluate and integrate the criteria under uncertainty. The results show that the traditional RRV criteria do not accurately evaluate the temporal resilience of the system under extreme events. Therefore, 4R-D and SR criteria are taken into account to evaluate the long-term resilience of the water supply and demand system under extreme events with known and unknown probabilities of occurrence. According to the results, the RRV criteria are not sensitive to water resources management scenarios, but the scenarios improve 4R-D and SR criteria up to 0.6–0.8 and 0.76, respectively.