摘要
This paper examines climate change impacts on the water resources system of the Manicouagan River (Québec, Canada). The objective is to evaluate the performance of existing infrastructures under future climate projections and the associated uncertainties. The main purpose of the water resources system is hydropower production. A reservoir optimization algorithm, Sampling Stochastic Dynamic Programming (SSDP), was used to derive weekly operating decisions for the existing system subject to reservoir inflows reflecting future climate, for optimum hydropower production. These projections are simulations from the SWAT hydrologic model for climate change scenarios for the period from 2010 to 2099. Results show that the climate change will alter the hydrological regime of the study area: earlier timing of the spring flood, reduced spring peak flow, and increased annual inflows volume in the future compared to the historical climate. The SSDP optimization algorithm adapted the operating policy to the future hydrological regime by adjusting water reservoir levels in the winter and spring, and increasing the release through turbines, which in the end increased power generation. However, there could be more unproductive spills for some power plants, which would decrease the overall efficiency of the existing water resources system.