地下水位时间序列噪声分析

英文篇名：THE NOISE ANALYSIS OF UNDERGROUND WATER LEVEL OBSERVATION TIME SERIES
中文刊名：防灾减灾学报
英文刊名：Journal of Disaster Prevention and Reduction
作者：刘俊清 ; 张宇 ; 郑传芳 ; 张晨侠 ; 丁广
英文作者：LIU Jun-qing1 ; 2 ; ZHANG Yu1 ; ZHENG Chuan-fang1 ; ZHANG Chen-xia1 ; DING Guang1(1.Earthquake Administration of Jilin Province ; Jilin Changchun 130117 ; China ; 2.College of Geoexploration Science and Technology ; Jilin University ; Jilin Changchun 130026 ; China)
中文关键词：时间序列 ; 噪声模型 ; 最大似然估计 ; 幂律噪声
英文关键词：time series ; noise model ; MLE ; power law noise
出版日期：2013-03-15
机构：吉林省地震局;吉林大学地球探测科学与技术学院;
年：2013
期：01
出版单位：防灾减灾学报

摘要

地震的孕育时间长达几十年甚至几百年,前兆观测往往获得相当长时间的观测数据,时间序列中包含大量的随机信号,也就是噪声。本文应用噪声分析软件CATS,引入经典的噪声统计模型,运用最大似然估计法分析吉林省地震前兆地下水观测时间序列的噪声特性,结果表明地下水观测噪声可用闪烁噪声加白噪声联合模型来描述。
The earthquake preparation lasts for decades or even centuries.So earthquake precursory observation often get long time observation data.The observed time series contains a large number of random signal,namely noise.In this paper,based on the application of noise analysis software CATS,the classical statistical noise model and the maximum likelihood estimation,the noise characteristics of the groundwater observation time series of the seismic precursor in Jilin province are analyzed.The results show that the groundwater observation noise can be described by the combined noise model with flicker noise and white noise.

引文

[1]Langbein,J.,and H.Johnson.Correlated error ingeodetic time series:Implications for time-dependent deformation[J].J.Geophys Res.,1997,102:591-604.
    [2]Zhang J,Bock Y.Southern California permanent GPS geodetic Array:eroranalysis of daily position estimates and site velocities[J].J Geophys Res,1997,102:18035-18056.
    [3]黃立人.GPS基准站时间序列的噪声特性分析[J].大地测量与地球动力学,2006,26(2):31-33.
    [4]黃立人,符养.GPS连续观测站的噪声分析[J].地震学报,2007,29(2)197-202.
    [5]Agnew,D.C.,The time-domain behavior of power-law noises[J].Geophys.Res.lett.,1992,19,333-336.
    [6]史忠科.最优估计的计算方法[M].北京:科学出版社,2001.
    [7]Williams,S.D.P.CATS:GPS coordinatet ime series analysis software[J].GPS Solut.DOI2007,10.1007/s10291-007-0086-4.