基于贝叶斯理论的中国近海网格化气象要素概率预报研究

Probabilistic forecast of gridding 10 meter wind over China sea based on Bayesian theory

基于贝叶斯理论建立了一种基本概率预报模式,将欧洲中期天气预报中心数值模式的确定性海面风场预报转换为概率预报,并对概率化后的释用产品进行了评估与检验。对2016年3月和4月释用产品的检验结果表明,在中国近海1425个渔区的全风速预报中,贝叶斯概率预报准确率均较高;对6级及以上大风的预报,贝叶斯概率预报的准确率也较高,且空报率低,无漏报。本文的研究成果为预报员进行业务预报具有指导性意义,同时将为进一步构建海面风场的网格化业务预报体系奠定了技术基础。

Based on Bayesian theory, a basic model of the probabilistic forecast was established out of the deterministic forecast of European Centre of Medium-Range Weather Forecast （ECMWF）. The probability forecast from March to April 2016 was evaluated by using ERA-Interim. The results show that the probability forecasts of both all wind and strong wind have better forecast values over China sea. This study not only provides a guidance forecast for forecaster and user, but also lays the foundation for building gridding operational forecast system of 10 meter wind over China sea.