对比频率偏振分析方法和概率谱密度函数方法  

Frequency polarization analysis method and PSDF method

作　　者：李晓梦 徐彦[1] 王嗣尹 解朝娣[1] LI XiaoMeng;XU Yan;WANG SiYin;XIE ZhaoDi(College of Earth Sciences,Yunnan University,Yunnan 650500,China;Wenshan Zhuang and Miao Autonomous Prefecture Water Conservancy and Electric Power Survey and Design Institute,Wenshan 663000,China)

机构地区：[1]云南大学地球科学学院,昆明650500 [2]文山壮族苗族自治州水利电力勘察设计院,文山663000

出　　处：《地球物理学进展》2022年第1期41-50,共10页Progress in Geophysics

基　　金：国家自然科学基金(41564003,41964001)资助。

摘　　要：本文从IRIS上下载了2012年10月21日—10月31日位于俄罗斯Obninsk的OBN台站(Ⅱ台网)和美国东部宾夕法尼亚州石油城的州立公园内的M54A台站(TA台网)各自共264 h的三分量数据.我们运用频率偏振分析和概率谱密度函数方法对数据进行处理,来探讨两种研究背景噪声随频率变化的处理方法的差异.研究结果显示这两种方法得到的三分量信号能量随频率变化特征大致相同,并且两种方法得到的第二地脉动(3-10 s)和第一地脉动(10-16 s)的峰值周期几乎重合.概率谱密度函数方法得到的功率谱密度曲线比频率偏振分析方法的结果更加平滑,这可能是由于概率谱密度函数方法在计算过程中进行了两次平滑.由于概率谱密度函数方法的运算时间较短,因此更适合用于数据观测质量的日常监控工作.而频率偏振分析方法在高频段(周期小于2 s)上得到的信号能量比概率谱密度函数方法的结果高,这可能是由于方法本身造成的.频率偏振分析方法能提供比概率谱密度函数方法更多的质点运动信息,例如偏振程度(β^(2))、信号的后方位角(Θ_(H))、入射角(Θ_(V))、垂直分量和水平分量的相位差(φ_(VH))、两个水平分量的相位差(φ_(HH))随频率的变化情况.这些角度信息可以帮助我们分析信号的类型(体波或面波)及来源方向,从而可对海上气旋活动进行跟踪和研究.We downloaded 264 hours of three-component seismic data from IRIS since October 21, 2012 to October 31, 2012 of two broadband seismic stations OBN(II network) of OBNINSK in Russia and M54 A(TA network) in the state park of Oil City, Pennsylvania in the eastern United States. We use two different data processing methods that frequency polarization analysis method and the probability spectral density function method to explore the difference between the two methods of processing the ambient seismic noise with frequency. We found that the variation characteristics of noise power with frequency fluctuation obtained by frequency polarization analysis method and probability spectral density function method are roughly the same, and the peak periods of double-frequency microseismic(3-10 s) and single-frequency microseismic(10-16 s) obtained by the two methods almost coincide. However, the probability spectral density function method is more suitable for daily monitoring of data observation quality because of its smoothing variation with frequency and short operation time. It may be caused by the probability spectral density function method is smoothed twice in the calculation process. The signal power obtained by the frequency polarization analysis method is higher than the probability spectral density function method in the high frequency band(period less than 2 seconds), which may be caused by the method itself. And The frequency polarization analysis method can obtain more information about particle motion, For example, the degree of polarization(β^(2)), the back azimuth of the signal(Θ_(H)), the incident Angle(Θ_(V)), the phase difference between the vertical component and the horizontal component(φ_(VH)), the phase difference between the two horizontal components(φ_(HH)) changes with frequency. This Angle information can help us to analyze the type of signal(body wave or surface wave) and the direction of source, so as to track and study the activity of cyclone over the sea.

关 键 词：环境背景噪声 频率偏振分析 概率谱密度函数 

分 类 号：P315[天文地球—地震学]