气象学报2026，Vol.84Issue(3)：532-549,18.DOI:10.11676/qxxb2026.20250145
基于标准化异常梯度提升算法的集合预报后处理
Post-processing of ensemble forecasts based on a gradient boosting algorithm with standardized anomalies
摘要
Abstract
In recent years,ensemble forecasting has become a vital tool for major global weather forecasting centres.However,ensemble forecasts often exhibit underdispersion and systematic biases,making the application of statistical post-processing methods essential.The standardized anomalies model output statistics(SAMOS)is a commonly used post-processing technique that provides a complete description of the forecast distribution.Nevertheless,SAMOS typically relies only on predictors either directly related to the forecast variable or selected based on subjective judgment,potentially overlooking other valuable predictors.Moreover,directly incorporating too many predictors into SAMOS may lead to overfitting.Therefore,effectively selecting key predictors from numerous variables provided by forecast models remains a significant challenge.Boosting-based variable selection and optimization algorithms have proven to be effective in mitigating overfitting and identifying the most important predictors.This study proposes the standardized anomalies gradient boosting(SABST)method by integrating the strength of SAMOS with a boosting-based variable selection algorithm.SABST is applied to calibrate ensemble forecasts of 2 m air temperature,2 m relative humidity,and 10 m wind speed.The SABST model is developed using the European Centre for Medium-Range Weather Forecasts(ECMWF)high-resolution ensemble forecast(ensemble prediction system,ENS)products during 2019-2020 and is systematically compared with SAMOS,focusing on its calibration performance and bias correction ability.Results show that compared to ENS and SAMOS,SABST performs better in addressing underdispersion in probabilistic forecasts and improving the accuracy of deterministic forecasts.Based on the continuous ranked probability skill score(CRPSS),SABST improves the average CRPSS by 9.5%,15.3%,and 4.6%across all forecast lead time compared to SAMOS.These findings demonstrate the advantage of introducing additional potential predictors in the SABST framework and highlight the method's potential for application in ensemble forecast post-processing.关键词
统计技术/预报技术/数值天气预报/后处理/气候态Key words
Statistical techniques/Forecasting techniques/Numerical weather prediction/Postprocessing/Climatology分类
天文与地球科学引用本文复制引用
刘普,张南,王玉虹,MARKUS Dabernig,AITOR Atencia,王勇..基于标准化异常梯度提升算法的集合预报后处理[J].气象学报,2026,84(3):532-549,18.基金项目
河北省气象台复杂地形精细化气象要素客观预报系统(2020h490)、河北省气象局2023年省级财政投资业务建设项目《气象业务核心能力提升工程-INCA温度预报技术改进》、四川省科技计划(2025YFNH0006)、北极阁青年基金项目(BJG202510)、江苏省气象局面上项目(KM202515). （2020h490）
