摘要
该文依据１９８３～１９９２年６～８月间２７６次强对流天气过程，研究北京地区强对流天气的形成条件及其展望预报方法．通过５００ｈＰａ逐日分型，将影响北京地区的大尺度环流型划分为５种类型．在各型中随机抽取出现和不出现强对流天气样本共２９８个，分别作合成分析，研究各环流型出现和不出现强对流天气的合成形势及差异．结果表明，出现和不出现强天气的环流系统特征、结构及物理量分布都有明显不同，从而概括出概念模式，导出各环流型预报着眼点．在此基础上，从各型计算的数十个物理量参数中，筛选出最佳预报因子，采用判断树预报流程，逐型建立强天气的展望预报方法．经１９９３、１９９４年两年业务试验，该预报方法具有计算速度快，微机自动预报，预报结论客观的特点，在实际工作中取得十分满意的预报效果．
The forming condition and the prospect forecasting method of severe convective weather in Beijing area are studied according to 276 severe weather cases in June, July and August from 1983 to 1992. By means of classification of 500 hpa patterns day by day, the largescale circulation patterns in Beijing area are divided into five types. From the cases of convective occurrence and non-occurrence in each type 298 weather samples are randomly adopted and analysed for studying the circulation pattern and difference of convective occurrence and no-occurrence. It is shown that circulation features, structure and the distribution of the scalars for convective occurrence and no-occurrence are different obviously. Therefore a concept model is summerized and the starting point about circulation pattern forecast is obtained. On this basis, after calculating dozens of physical parameters, the best forecasting factor is selected, and the prospect forecasting technique is developed by use of dicision tree method. The operational experiments in 1993 and 1994 show that the forecasting is objective, fast and automatic.
出处
《应用气象学报》
CSCD
北大核心
1996年第3期336-343,共8页
Journal of Applied Meteorological Science
基金
空军(九五)重点科研资助
