Improved Monthly and Seasonal Multi-Model Ensemble Precipitation Forecasts in Southwest Asia Using Machine Learning Algorithms

Southwest Asia has different climate types including arid, semiarid, Mediterranean, and temperate regions. Due to the complex interactions among components of Earth system, forecasting precipitation is a difficult task in such large The aim this paper propose learning approach, based on artificial neural network (ANN) random forest (RF) algorithms for post-processing output models, order provide multi-model ensemble monthly southwest Asia. For purpose, four GEM-NEMO, NASA-GEOSS2S, CanCM4i, COLA-RSMAS-CCSM4, included North American (NMME) project, are considered algorithms. Since each model nine lead times, total 108 ANN RF models trained month year. To train proposed an ERA5 reanalysis dataset employed. compare performance algorithms, evaluation criteria calculated model. results indicate that better than individual NMME models. Moreover, outperformed all times months