ABSTRACT Accurate climate modelling and prediction are essential for understanding climate change impacts, optimising water resource management, and mitigating drought and flood risks. However, simulations of extreme precipitation and high‐temperature events remain highly uncertain due to the strong sensitivity of regional climate models to physical parameterisation schemes and the lack of systematic identification of optimal scheme combinations, particularly in regions characterised by complex terrain and reservoir regulation. To address this gap, this study systematically evaluates the performance of the Weather Research and Forecasting (WRF) model under multiple physical parameterisation schemes for extreme precipitation and high‐temperature events in the Wanzhou District of the Three Gorges Reservoir region. The optimal schemes are identified based on a comprehensive assessment of temporal and spatial consistency indices, error statistics, and agreement between simulations and observations. In addition, extreme gradient boosting (XGBoost), random forest (RF), and long short‐term memory (LSTM) models are applied to optimise WRF outputs through post‐processing. The results show that Scheme A5 (WSM6‐GD‐Noah‐RRTM/Dudhia) provides the best overall performance for extreme precipitation simulations, while Scheme B5 (WSM6‐GD‐Noah‐RRTMG/CAM) performs best for high‐temperature events. Furthermore, coupling the optimal WRF configurations with the RF model yields the greatest improvement, increasing the temporal and spatial consistency indices by 38.03% and 57.30% for extreme precipitation, and by 25.25% and 55.79% for high‐temperature events, respectively. The integration of machine learning models into WRF substantially improved the accuracy and reliability of extreme event predictions, providing vital scientific evidence and technical support for disaster risk reduction and meteorological forecasting in the Wanzhou District of the Three Gorges Reservoir region, thereby contributing to the sustainable development of regional environments and socioeconomics.
Zhou et al. (Wed,) studied this question.