Sichuan Rainfall Prediction Using an Analog Ensemble

This study aimed to address the significant bias in 0–44-day precipitation forecasts under numerical weather conditions. To achieve this, we utilized observational data obtained from 156 surface stations Sichuan region and reanalysis grid National Centers for Environmental Prediction Climate Forecast System Model version 2. Statistical analysis of spatiotemporal characteristics was conducted, followed by a correction experiment based on Analog Ensemble algorithm different seasons region. The results show that, terms spatial distribution, amounts days Province gradually decreased east west. Temporally, highest number occurred autumn, while maximum amount observed summer. effectively reduced error model forecast However, effectiveness varied seasonally, primarily because differing performance AnEn method relation events various magnitudes. Notably, effect poorest heavy-rain forecasts. In addition, degree improvement initial times lead times. As time increased, weakened.