摘要
瑞利波具有强能量和频散特性,面波勘探采集瑞利面波数据分析反演得到横波速度结构,在浅地表勘探领域得到了广泛的应用,根据面波频散反演速度结构是面波勘探的重要环节之一.面波频散与地下介质的弹性参数是非线性关系,全局优化方法是解决非线性反问题的有效办法,贝叶斯方法是一种基于统计的全局优化方法.贝叶斯公式反映了先验信息和条件概率乘积与后验概率之间的正比关系,利用马尔科夫链蒙特卡洛采样方法可以获得后验分布的采样.与其他方法相比,该方法不是给出一个模型最优解,而是统计出模型参数的平均值和方差,平均值即是反演的模型解,方差能对该反演结果的不确定性做出评价.本文采用贝叶斯方法对模拟和实测瑞利面波数据频散曲线进行反演,结果显示,该方法能获得比较精确的横波速度和厚度参数,验证了本文方法可行.在高丽营地区采集了多个单炮面波数据,对数据进行频散谱处理拾取频散曲线,通过贝叶斯反演方法获得了该测线二维横波速度剖面图,有助于划分和解释该测区地层及断层地质构造.
Rayleigh wave is characterized by strong energy and dispersion.Surface wave prospecting is used to image shear wave velocity structure from Rayleigh wave dispersion and widely applied to shallow geological investigation.Velocity structure inversion is one of the key processing steps based on Rayleigh surface wave dispersion relationship between elastic parameters and phase velocity.The dispersion relationship is non-linear.So it is difficult to solve inversion problem.Global optimization method is an effective way to solve the inverse problem.Bayesian method,based on statistics,is one of global optimization methods.And Bayesian formula expresses the model posterior probability from prior information and conditional probability.For dispersion inversion,prior information shows elastic parameter distribution and is generally uniform.And the conditional probability is Gauss distribution.The sampling of posterior distribution can be obtained by Markov chain Monte Carlo sampling method.Compared with other methods,the proposed method provides the mean and variance,instead of an optimal solution,for the model parameters.The mean is a solution of inversion problem and uncertainty of the solution can be evaluated from variance.In this paper,Bayesian method is used to invert for different models and real data.The fundamental dispersion curves are extracted from the dispersion spectrum.And dispersion inversions are conducted by the proposed Bayesian method.The results show that accurate shear wave velocity and thickness parameters are obtained and their uncertainties are provided,which verifies the feasibility of this method.We conduct a surface wave prospecting test in Gaoliying area,Beijing.The dispersion curves are picked and inverted.A pseudo-two-dimensional shear wave velocity profile is obtained,which help to deduce the active fault structure of surveying area.
作者
刘昊楠
张致付
陈凯
LIU Hao-nan;ZHANG Zhi-fu;CHEN Kai(School of Geophysics and Information Technology,China University of Geosciences(Beijing),Beijing 100083,China)
出处
《地球物理学进展》
CSCD
北大核心
2020年第4期1584-1589,共6页
Progress in Geophysics
基金
国家自然科学基金(41530321)资助。
关键词
面波
频散曲线
贝叶斯反演
横波速度
Surface wave
Dispersion curve
Bayesian inversion
Shear wave velocity
