LAUSR

Abstract Hydropower has a damaging effect on aquatic biodiversity globally. Extensive research has demonstrated the significance of plant diversity in sustaining wetland services and functions. However, wetland plant diversity conservation has rarely been considered in hydropower operation optimization, and the relationship between hydropower generation and plant diversity indicators has rarely been quantified. In this study, a new optimization approach is proposed for reservoir operation to balance hydropower generation and plant diversity conservation in downstream wetlands. We identify key hydrological indicators affecting wetland plant diversity, accounting for the richness and abundance of plants as well as the evenness of multiple plant species. An artificial neural network (ANN) is subsequently adopted to quantify the complex relationship between hydrological and plant diversity indicators. Using the ANN model for plant diversity prediction, we develop a multi-objective optimization model for reservoir operation to maximize hydropower generation and wetland plant diversity indicators. This approach is applied to a large plant-dominated wetland (Baiyangdian) in China as a case study. Using a genetic algorithm to solve the optimization model, we suggest a reservoir operation scheme which can increase multi-year averaged plant diversity indicator in this wetland by 20% while decreasing reservoir hydropower generation by 10%. We also generate a set of Pareto-optimal solutions from the optimization model, which quantifies the tradeoff relationship between hydropower generation and plant diversity conservation. This relationship provides a reference for managers in determining reservoir operation rules depending on the demands on these two conflicting objectives.