摘要:利用1961～2012年全国553站逐日最低气温资料，根据百分位值法，定义了不同台站的极端低温阈值，并统计分析了逐年逐站发生极端低温事件的频数，并且使用经验正交函数分解（EOF）的方法对其时空变化特征进行了研究。研究结果表明：近51年来，我国冬半年极端低温事件总体呈明显的减少趋势，但仍存在明显的时空差异。EOF分析结果显示我国极端低温事件发生频率主要存在3种空间模态，即空间一致型和南北振荡型以及东西振荡型。并且对PC1（第一主分量）时间序列作突变检验的结果表明，我国冬半年极端低温事件频数在1986年前后存在明显突变现象，这与1986年前后东亚冬季风发生年代际突变时间点一致。进一步相关回归分析结果表明，我国冬半年极端低温事件频数与东亚冬季风指数存在很好的相关性，这可能与西伯利亚高压的变化有显著的联系，即冬半年西伯利亚高压强时，极端低温事件发生较多，反之亦然。分析的结果表明，1986年前后，中国冬半年极端低温事件发生频次的突变可能与西伯利亚高压的南移并且减弱有关。
关键词:极端低温/
东亚冬季风/
时空特征/
冬半年
Abstract:Based on the daily minimum temperature dataset from 553 stations in China from 1961 to 2012, extreme low temperature (ELT) thresholds are determined for different stations, and occurrence frequency of ELT events during boreal winter for each station is estimated and analyzed using empirical orthogonal function (EOF). The relationship between ELT events and East Asian winter monsoon (EAWM) circulation is studied. The results suggest that the ELT events in China during the boreal winter have remarkably decreased in the past 51 years, with some spatial and temporal differences. The results of EOF analysis show that there are three main spatial modes in the ELT events in China, namely, the spatial uniformity mode, the North–South Oscillation mode, and the East–West Oscillation mode. In addition, the results of mutation test of the PC1 (the first pricipal component) series show that the abrupt change phenomenon of the occurrence frequency of ELT events during boreal winter is ubiquitous and occurred around 1986. Furthermore, the results of correlation analysis show that a good relationship was found between the frequency of ELT events and EAWM index, which may be due to the changes of the Siberian High. Besides, the results of composite analysis suggest that the abrupt change around 1986 may have something to do with the obviously southwardly movement and a weakening trend of the Siberian High.
Key words:Extreme low temperature/
East Asian winter monsoon/
Spatial and temporal characteristics/
Boreal winter
PDF全文下载地址:
http://www.iapjournals.ac.cn/qhhj/article/exportPdf?id=4e5c223d-9b36-497e-90e2-581d418605a1
