亚洲季风区过去700年来夏季极端干/湿事件多尺度变化特征分析

Multi-scale Analysis of the Extreme Dry / Wet Events in Asian Monsoon Region in Summer During Last 7 Centuries

长期器测数据的缺乏妨碍了对亚洲季风较长历史时期的时空变化特征及对这一变化背后的驱动因子的深入研究,Monsoon Asia Drought Atlas(MADA)资料是根据亚洲季风区内300多个不同地点的树木年轮而测定的一个历时700年(1300-2005年)的在整个亚洲范围内的季风变化记录。利用MADA资料中504个格点逐年夏季的帕尔马干旱指数(Palmer Drought Severity Index,PDSI),以5年为一个间隔得到1300-2005年夏季整个区域内处于极端湿润、严重湿润、极端干旱和严重干旱等级格点数的变化序列,分析了各序列的整体变化特征;针对气候系统的非线性/非平稳性和气候层次理论,采用集合经验模态分解(Ensemble empirical mode decomposition,EEMD)方法对亚洲季风区过去700年来的极端干/湿事件进行多尺度分析,求出各个序列的本征模态分量(IMF),分析了各IMF分量的方差贡献率,并对各个IMF分量进行了显著性检验;基于不同时间尺度的IMF分量,分析了亚洲季风区过去700年来极端和严重等级干湿状态在不同时间尺度的振荡变化特征,以及各个IMF分量的振荡对整个亚洲季风区干湿变化的影响作用。

Monsoon failures,mega droughts,and extreme flooding events have repeatedly affected the agrarian peoples of Asia over the past millennium. A scarcity of long-term instrumental climate data for many remote regions of Monsoon Asia impedes progress toward resolving these issues. To better elucidate the spatial complexity of the Asian monsoon,a large-scale,spatially explicit,long-term data set is needed. This context is provided here by our Monsoon Asia Drought Atlas（ MADA）. The MADA provides a seasonal-to centennial-scale window into the Asian monsoon＇s repeated tendency for extended dry and wet extremes with distinct spatial flavors of response. Recently,ensemble empirical mode decomposition（ EEMD） method is developed for non-linear and non-stationary signal analysis. The EEMD method is instituted and utilized in several fields such as de-noising,ocean surface measurement,metrology,image processing and so on. The method can work on nature signals（ non-linear and nonstationary signals） as well as reducing the speckle noise. The EEMD method is like as a filter bank that the signal is decomposed into several intrinsic mode functions（ IMFs） and the frequencies of IMFs are arranged in decrease order（ high to low） after the EEMD processing. The scaling mode of the EEMD method is similar to wavelet transform,but the signal resolution at different frequency domain is not decrease by down-sampling. In this paper,we propose the EEMD method to extract the multi-scale characters of the variability of extreme dry / wet events in Asian monsoon area. Using the Palmer Drought Severity Index data of MADA from 504 stations in Asian monsoon area in summer from 1300-2005 a and EEMD method,we get the series of the number of grids which is especially and serious wet or dry in this region by an interval of 5 years,and analysis the variations of these series. Based the feature of nonlinear / nonstationarity and multi-scale in climatic system,applying EEMD to the series of the number of grids which is especially and serious wet or dry for the purpose of multi-scale analysis,different scale signals（ IMFs）have been separated,and give the variance contributes and do the test of significant for each IMF. By analyzing the variations of IMFs,hereby,the changes in different scales for the extreme dry / wet events in Asian monsoon region in summer during last 7 centuries have been uncovered,and get the effect of different IMFs on the changes of drywet in Asian monsoon region.