A Stacking Ensemble Learning Model for Monthly Rainfall Prediction in the Taihu Basin, China

The prediction of monthly rainfall is greatly beneficial for water resources management and flood control projects. Machine learning (ML) techniques, as an increasingly popular approach, have been applied in diverse climatic regions, showing their respective superiority. On top that, the ensemble model that synthesizes advantages different ML models deserves more attention. In this study, based on stacking approach was proposed. Four prevalent models, namely k-nearest neighbors (KNN), extreme gradient boosting (XGB), support vector regression (SVR), artificial neural networks (ANN) are taken base models. To combine outputs from weighting algorithm used second-layer learner to generate predictions. Large-scale climate indices, large-scale atmospheric variables, local meteorological variables were predictors. R2, RMSE MAE, evaluation metrics. results show performance varied among nine stations Taihu Basin, while generally performed better than four showed spring winter summer autumn. During wet months, accuracy significantly. whole, measures, it concluded proposed multi-ML can provide a flexible reasonable framework applicable other regions.