Adaptive flood control operation of the Xin'an Reservoir in future precipitation extremes under climate change

In this study, a generalized extreme value (GEV) distribution-based statistical model has been developed and proposed to simulate historical and future precipitation extremes in the Xin'an River basin, and the vertical mixed runoff model was driven by future precipitation extremes to simulate the hydrological response to extreme flood events. An adaptive flood control operation model has been established and solved using genetic algorithm in order to ensure the safety of dam and downstream areas under precipitation extremes. In view of the precipitation events for the period 1951-2017, the monthly extreme precipitation events are expected to rise in the period 2020-2100 by 10.4%, 11.0%, and 11.4% at a 10-, 20-, and 50-year return period, respectively. After optimal regulation, the maximum release is reduced by 60.8%, 43.6%, and 42.7%, while the average reservoir water level is reduced by 0.13 m, 0.14 m, and 0.11 m in extreme flood events with a 10-, 20-, and 50-year return period, respectively. In conclusion, the adaptive flood control operation can ensure the safety of dam and downstream areas and mitigate possible impacts of extreme flood events under climate change.