基于P/S波解耦和上下行波分离的二维VTI介质弹性波逆时偏移

Elastic reverse time migration in VTI medium based on P/S wave-mode decoupling and upgoing/downgoing wavefield decomposition

VTI介质弹性逆时偏移可以有效地揭示地下地层结构,是当前流行的成像方法之一。首先构建VTI介质下的Christoffel方程,对该方程进行特征分析得到VTI介质下P波和S波的极化方向向量。根据Helmholtz分解原理,将各向异性弹性波场投影到P/S波极化方向矢量上,得到VTI介质下向量P/S波波场分离公式。然后通过一阶泰勒展开式对P/S波分离公式中的归一化项进行近似,得到一种高效且适用于任意复杂介质的各向异性P/S波解耦方法,并应用于弹性波逆时偏移。此外,低频噪声和成像假象严重干扰了逆时偏移的成像质量,前者由同路径波场成像造成,后者由于炮点上行波与检波点下行波相关形成。为此,将上下行波分离成像条件引入VTI弹性波逆时偏移成像,提出一套结合各向异性P/S波模式分离技术和上下行波分离的成像流程,有效地压制了成像中的低频噪声和串扰假象,提高了成像质量。最后,简单和复杂模型的成像结果均验证了方法的有效性。

Elastic reverse time migration in VTI media can effectively reveal the underground stratum structure,and thus has become one of the popular imaging methods.In this paper,firstly,the Christoffel equation in VTI medium is constructed,and the polarization direction vectors of P-wave and S-wave are obtained by solving the equation.According to the Helmholtz decoupling principle,the anisotropic elastic wavefield is projected onto the polarization direction,and the wavefield decoupling formula of P/S wave in VTI medium is obtained.Then,the normalization term in the P/S wave decoupling formula is approximated by the first-order Taylor expansion,and an efficient P/S wave decoupling method suitable for any complex medium is obtained and applied to the reverse time migration of elastic waves.In addition,low-frequency noise and imaging artifacts seriously affect the imaging quality of reverse time migration.The former is caused by the imaging of the same path wavefield,and the latter is caused by the correlation between the upgoing wavefield of source and the downgoing wavefild of the receivers.Therefore,the upgoing/downgoing wavefield separation imaging conditions are introduced to the VTI elastic wave inverse time migration,a set of imaging processes including P/S wave-mode decoupling and upgoing/downgoing wavefield separation,which effectively suppresses the low-frequency noise and crosstalk artifacts,and improves the imaging quality.Finally,the imaging results of simple and complex models validate the effectiveness of the method.