NINO3海面温度异常时间序列的小波分析

本文对连续小波变换和离散小波变换进行了简要介绍,对连续小波变换中小波函数与尺度的选取、小波尺度与Fourier频率的关系等进行了阐述。结合太平洋中部地区1871年至2005年6月期间NINO3海面温度异常时间序列进行了小波功率谱分析,提取了其时频特征,分析结果与主要厄尔尼诺和拉尼娜事件的发生时间吻合较好。最后,利用离散小波分解与重构对时间序列进行消噪处理,并对不同层数强制消噪效果进行了比较。

Wavelet analysis in temperature abnormity time series of NINO3 sea surface

In this paper,the continuous and discrete wavelet transform are introduced briefly.The choice of mother wavelets and scales for continuous wavelet transform is explained together with the relationship between scales and Fourier period.Wavelet power spectrum is used to analyze the time series of sea surface temperature abnormity of NINO3 from 1871 to 2005,6,and to obtain the characteristic of time and frequency.The analysis result is in accord with the happening time of primary El Nino and La Nina.Finally,the noise of the time series is reduced in discrete wavelet decompose and reconstruction,and the effects of different filtering are compared.