论文部分内容阅读
地震反射数据中包含有小尺度阻抗变化和平滑参考速度模型信息。给定一个参考速度模型,通过线性偏移反演可以确定反射层。如果参考速度不精确,则通过不同数据子集反演获得的反射层将是相干的。本文提出用图象的相干性反演,求取背景速度分布。我们已研究一种两步迭代反演法,用此法我们由长周期参考速度模型分离出小尺度变化的地震速度的修正值。对一个特定的初始背景速度模型,我们用一个波形失配函数完成小尺度速度变化的反演。对于线性处理步骤,采用了线性化偏移-反演法,该法以我们和Lambare和
Seismic reflection data contains small-scale impedance changes and smooth reference speed model information. Given a reference velocity model, the reflection layer can be determined by linear offset inversion. If the reference speed is not accurate, the reflection layers obtained by inversion of different data subsets will be coherent. This paper proposes to use the coherence inversion of the image to find the background velocity distribution. We have studied a two-step iterative inversion method, by which we use the long-period reference velocity model to isolate the small-scale change in seismic velocities. For a particular initial background velocity model, we use a waveform mismatch function to perform small scale velocity inversion. For linear processing steps, a linearized offset-inversion method is used, which takes us and Lambare and