中国冬季区域性极端低温事件分类及其与气候指数极端性的联系

基于区域性极端低温事件客观识别技术, 对1951–2010年中国冬季的区域性极端低温事件进行客观识别. 根据事件的空间分布特征, 将综合指数前60位的事件划分为全国型、东部型、东北-华北型、华北-华南型、北方型和西北-华南型六类; 通过分析不同区域类型低温事件形成的环流背景场验证了分类的有效性. 在此基础上, 以1971年1月21日开始的典型事件为例, 分析了事件对应的海温场、高度场和风场的异常, 确定与区域性极端低温事件联系较密切的可能气候因子, 进而分析不同类型事件与各气候指数异常的对应关系. 总体而言, 赤道中东太平洋海温指数异常偏小、北太平洋涛动指数异常偏小、北极涛动指数异常偏小和冬季风异常偏强时, 发生区域性极端低温事件的概率较高; 且这四种指数的历年冬季平均值达到15%(或85%)极端阈值的年份中, 发生区域性极端低温事件的百分率分别达到80.0%, 77.8%, 60.0%和62.5%, 从而为区域性极端低温事件的诊断和预测研究等提供了一定的参考.关键词：区域性极端低温事件空间分类气候指数极端

The classification of winter regional extreme low temperature events in China and their corresponding relationship to climatic indices extreme anomaly

We identify China regional low temperature extreme events (RELTEs) in winter during the periods from 1951 to 2010 using objective identification technique for regional low temperature extreme events (OITRELTEs). The 559 RELTEs are identified and classified into 6 types, i.e., nationwide style, east style, northeast-north China style, north-south China style, south style, and northwest-south China style, according to the spatial distribution of these events. The circulation backgrounds of different styles of low temperature events are also analyzed. In addition, taking the classical event that began from January 21st in 1971 for example, anomaly characteristics of sea surface temperature, geopotential height and winds vectors are investigated specifically. Based on these analyses, the corresponding relationships between different types of events and anomalies of climatic indices are further studied, and the relations between mainly influencing index and event are obtained for different types of events. On the whole, when the NINO3.4, the Pacific decadal oscillation, and the Arctic oscillation are small and the winter wind index is strong, the probability with which the RELTE happens is high; in the years in which the winter average values of the four indices reach 15% of extreme threshold, the percentages of occurrence of RELTE reach up to 80.0%, 77.8%, 60.0% and 62.5%, respectively. Therefore, certain signals can be offered for diagnosing and predicting the RELTE from the index anomalies.
Keywords：regional low temperature extreme eventsspatial distributionclimatic indexextreme