An Adaptive Surrogate Assisted CE-QUAL-W2 Model Embedded in Hybrid NSGA-Ⅱ_ AMOSA Algorithm for Reservoir Water Quality and Quantity Management

摘要

The Meimeh dam construction is a project, planned to provide sustainable livelihoods, social, and economic developments in the Meimeh River Basin, Ilam, Iran. However, due to the high concentration of TDS in the Meimeh River and its tributaries, river impoundment and water storage can be harmful to the Meimeh Reservoir. The upstream inflow control and the reservoir operation management in a selective withdrawal scheme (SWS) were used to mitigate the potential environmental degradation of Meimeh Reservoir's low water quality. CE-QUAL-W2 and WEAP (Water Evaluation and Assessment Programming) models were employed to evaluate the effects of various upstream saline inflow control scenarios. The analysis indicated that the diversion of the Siyoul tributary flow rate in the Meimeh River could result in lower violations of Total Dissolved Solids (TDS) concentrations and better water supply satisfaction. Then, the optimal reservoir operation management strategies in a SWS were derived in the best upstream inflow control scenario. The adaptive surrogate-assisted WQSM (water quality simulation model), coupled to hybrid NSGA-II_AMOSA (Non-dominated Sorting Genetic Algorithm-II_Archived Multi-Objective Simulated Annealing) algorithm, has been applied to derive the suitable reservoir operation strategies in SWS, improve the water supply satisfaction and alleviate the adverse effects of reservoir outflow TDS violations. The performances of the best water quality and supply scenarios have been compared with the scenario based on the standard operation policy (SOP) in the Meimeh Reservoir. The results show that most violations of TDS criteria occur during the peak agricultural seasons and significant water deficit in some dry years happens in the SOP.