近似熵检测气候突变的研究  

The research of detecting abrupt climate change with approximate entropy

作　　者：王启光[1,2] 张增平[1,3] 

机构地区：[1]扬州大学物理科学与技术学院,扬州225009 [2]中国科学院大气物理研究所东亚区域气候-环境重点实验室,北京100029 [3]国家气候中心,中国气象局气候研究开放实验室,北京100081

出　　处：《物理学报》2008年第3期1976-1983,共8页Acta Physica Sinica

基　　金：国家重点发展基础研究项目(批准号:2006CB400503);国家自然科学基金(批准号:40675044)资助的课题~~

摘　　要：引入近似熵方法应用于理想时间序列,检验了该方法具有区别不同动力学结构和检测其突变的特性.然后应用于中国气象局国家气候中心740站点1960—2000年的逐日气温和降水观测资料.分析结果表明,温度和降水分布存在明显的区域分布特征,反映出气候系统动力学结构区域间的差异;在20世纪70年代末到80年代初各站点具有不同程度的突变现象,说明了突变时空分布的不均一性以及气候系统演化的复杂性.The approximate entropy method is applied to an ideal time series. It is indicated that the method is capable of distinguishing between different structural dynamics and detecting structural dynamics mutation. And then, this method is applied to the observation data of the daily temperature and precipitation of the 740 sites between 1960 and 2000. The results showed that, the approximate entropy distribution of the temperature and precipitation has a characteristic of obvious district distribution, and displays the territorial difference of dynamic structures of climate system. Abrupt climate changes have been detected by the approximate entropy of the annual precipitation and temperature between 1970s and 1980s. The characteristics of the in homogeneity of temporal and spatial distribution of abrupt climate changes and the complexity of climate systems have been demonstrated.

关 键 词：气候突变 温度 降水 近似熵 理想时间序列 检测 

分 类 号：P467[天文地球—大气科学及气象学]