A NEW DERIVATION OF THE WAVEFRONT CURVATURE TRANSFORMATION AT AN INTERFACE BETWEEN TWO INHOMOGENEOUS MEDIA*

The principles for ray-tracing and wavefront curvature calculations in a three-dimensional medium are reviewed. A new derivation of the transformation of the wavefront curvature matrix at an interface between two inhomogeneous media is given. The derivation is based on a Taylor series expansion of the ray refraction equation at the interface between two inhomogeneous media, and only elementary geometric arguments are used. The wave-front curvature transformation at the interface is obtained by neglecting all terms in the direction of the surface normal. With proper definition of the variables, the derivation is also valid for a reflected wave-front. A simplified transformation rule is derived for a reflected wave of the same type as the incident wave.     

计算最小走时和射线路径的界面网全局方法

用慢度分块均匀正方形模型将介质参数化，仅在正方形单元的边界上设置计算结点，这些结点构成界面网．根据Ｈｕｖｓｅｎｓ和Ｆｅｒｍａｔ原理，由不断扩张、收缩的波前点扫描代替波前面搜索，在波前点附近点的局部最小走时计算中对波前点之间的走时使用双曲线近似，通过比较确定最小走时和相应的次级源位置，记录在以界面网点位置为指针的３个一维数组中．借助这些数组通过向源搜索可计算任意点（包括界面网以外的点）上的全局最小走时和射线路径．这一方法不受介质慢度差异大小限制，占内存少，计算速度较快，适于走时反演和以Ｍａｓｌｏｖ射线理论为基础的波场计算．     

HTI介质中的反射纵波方位属性

利用横波响应进行裂缝性各向异性介质的检测在实际应用中取得了很好的效果,但技术复杂、成本较高使该方法的广泛使用受到限制,而纵波资料采集和处理技术的精细有效保持了纵波的各种属性,这为直接利用纵波资料进行裂缝检测创造了条件.地下垂直定向裂缝通常用HTI介质模型来描述,为此,本文利用射线追踪和反射率法计算了层状各向同性介质背景下的HTI介质顶、底界面的反射纵波旅行时和反射系数,并分析了这些属性随观测方位的变化规律.研究表明,HTI介质底界面反射纵波旅行时和HTI介质顶界面反射系数表现出了明显的方位各向异性;旅行时、振幅和AVO梯度属性均在0°观测方位和90°观测方位上存在最大差异,可以用多种属性联合来精确判定裂缝的发育方向.     

适于Kirchhoff叠前深度偏移的地震走时李代数积分算法

本文基于拟微分算子理论和李代数积分法,根据程函方程和波场坐标变换,提出一种新的适于横向变速介质Kirchhoff叠前深度偏移的地震波走时算法.该算法与Kirchhoff叠前时间偏移所用李代数时间积分表达相比,差异在于增加了波数一次项,且二次项的系数在求积时亦需进行修正.针对单平方根算子象征、李代数积分、指数映射和走时多项式的求解而言,皆需对以往Kirchhoff叠前时间偏移中所用算法进行深化调整.文中数值算例对比了本文李代数积分表达与时间积分的区别,本算法计算结果与线性横向变速介质中的理论值相当吻合.通过走时多项式中各项对结果的影响分析,可知非对称项使计算精度得到了进一步提高.数值试验表明,本算法对横向变速介质中走时求取是可行的,且不需要存储海量走时表,有利于提高Kirchhof叠前深度偏移的精度和效率.     

3D P‐wave traveltime computation in transversely isotropic media using layer‐by‐layer wavefront marching

Subsurface rocks (e.g. shale) may induce seismic anisotropy, such as transverse isotropy. Traveltime computation is an essential component of depth imaging and tomography in transversely isotropic media. It is natural to compute the traveltime using the wavefront marching method. However, tracking the 3D wavefront is expensive, especially in anisotropic media. Besides, the wavefront marching method usually computes the traveltime using the eikonal equation. However, the anisotropic eikonal equation is highly non‐linear and it is challenging to solve. To address these issues, we present a layer‐by‐layer wavefront marching method to compute the P‐wave traveltime in 3D transversely isotropic media. To simplify the wavefront tracking, it uses the traveltime of the previous depth as the boundary condition to compute that of the next depth based on the wavefront marching. A strategy of traveltime computation is designed to guarantee the causality of wave propagation. To avoid solving the non‐linear eikonal equation, it updates traveltime along the expanding wavefront by Fermat's principle. To compute the traveltime using Fermat's principle, an approximate group velocity with high accuracy in transversely isotropic media is adopted to describe the ray propagation. Numerical examples on 3D vertical transverse isotropy and tilted transverse isotropy models show that the proposed method computes the traveltime with high accuracy. It can find applications in modelling and depth migration.     

A Fast Sweeping Scheme for Calculating P Wave First-Arrival Travel Times in Transversely Isotropic Media with an Irregular Surface

Seismic wave propagation shows anisotropic characteristics in many sedimentary rocks. Modern seismic exploration in mountainous areas makes it important to calculate P wave travel times in anisotropic media with irregular surfaces. The challenges in this context are mainly from two aspects. First is how to tackle the irregular surface in a Cartesian coordinate system, and the other lies in solving the anisotropic eikonal equation. Since for anisotropic media the ray (group) velocity direction is not the same as the direction of the travel-time gradient, the travel-time gradient no longer serves as an indicator of the group velocity direction in extrapolating the travel-time field. Recently, a topography-dependent eikonal equation formulated in a curvilinear coordinate system has been established, which is effective for calculating first-arrival travel times in an isotropic model with an irregular surface. Here, we extend the above equation from isotropy to transverse isotropy (TI) by formulating a topography-dependent eikonal equation in TI media in the curvilinear coordinate system, and then use a fast sweeping scheme to solve the topography-dependent anisotropic eikonal equation in the curvilinear coordinate system. Numerical experiments demonstrate the feasibility and accuracy of the scheme in calculating P wave travel times in TI models with an irregular surface.     

Automating adjoint wave-equation travel-time tomography using scientific workflow

Recent advances in commodity high-performance computing technology have dramatically reduced the computational cost for solving the seismic wave equation in complex earth structure models. As a consequence, wave-equation-based seismic tomography techniques are being actively developed and gradually adopted in routine subsurface seismic imaging practices. Wave-equation travel-time tomography is a seismic tomography technique that inverts cross-correlation travel-time misfits using full-wave Fréchet kernels computed by solving the wave equation. This technique can be implemented very efficiently using the adjoint method, in which the misfits are back-propagated from the receivers (i.e., seismometers) to produce the adjoint wave-field and the interaction between the adjoint wave-field and the forward wave-field from the seismic source gives the gradient of the objective function. Once the gradient is available, a gradient-based optimization algorithm can then be adopted to produce an optimal earth structure model that minimizes the objective function. This methodology is conceptually straightforward, but its implementation in practical situations is highly complex, error-prone and computationally demanding. In this study, we demonstrate the feasibility of automating wave-equation travel-time tomography based on the adjoint method using Kepler, an open-source software package for designing, managing and executing scientific workflows. The workflow technology allows us to abstract away much of the complexity involved in the implementation in a manner that is both robust and scalable. Our automated adjoint wave-equation travel-time tomography package has been successfully applied on a real active-source seismic dataset.     

垂向非均匀介质的走时反演

本文提出了一种利用反射波走时曲线计算垂向非均匀介质速度和反射界面深度的方法。当在地球表面没有获得来自地下某一深度范围内介质的任何信息时,可以认为这一深度范围内的地震波速度具有连续性。利用来自其底部反射界面的反射波走时曲线,可以计算出该深度范围的地震波速度结构。对三种模型进行了理论计算,所得反演解与真实值较为一致,其中计算出的反射界面深度最为精确。利用本文提出的方法可以计算两个相邻反射面之间的垂向非均匀速度结构,如低速层等。     

三维复杂地壳结构非线性走时反演

中国大陆中西部乃至全球造山带普遍具有复杂地壳结构.随着矿产资源勘探和深部探测研究的深入,探测造山带及盆山耦合区下方地壳精细结构正逐渐成为当前面临的巨大挑战.人工源深地震测深方法正越来越清晰地揭示出不同构造域地壳速度结构的基本特征,然而传统的层状结构模型参数化方法难以准确描述复杂地质模型,通常情况下多忽略速度结构的精细间断面且采用层边界平滑处理,难以满足地壳精细结构成像的发展要求.针对上述困难,本文采用最近发展的块状结构建模方案构建三维复杂地壳模型,基于逐段迭代射线追踪正演走时计算方法,推导了走时对三角形界面深度以及网格速度的偏导数,开展了非线性共轭梯度走时反演方法研究.发展了利用直达波和反射波等多震相走时数据对界面深度和网格速度的多参数联合反演方法,并引人不同种类震相数据的权系数和不同类型参数偏导数归一化的方法.数值算例表明,基于块状结构的非线性共轭梯度走时反演方法适用于复杂地壳结构模型,在利用人工源走时数据反演复杂地壳精细结构领域具有良好的应用前景.     

基于虚拟偏移距方法的各向异性转换波保幅叠前时间偏移

In this paper, we use the method of pseudo-offset migration （POM） to complete converted wave pre-stack time migration with amplitude-preservation in an anisotropic medium. The method maps the original traces into common conversion scatter point （CCSP） gathers directly by POM, which simplifies the conventional processing procedure for converted waves. The POM gather fold and SNR are high, which is favorable for velocity analysis and especially suitable for seismic data with low SNR. We used equivalent anisotropic theory to compute anisotropic parameters. Based on the scattering wave traveltime equation in a VTI medium, the POM pseudo-offset migration in anisotropic media was deduced. By amplitude-preserving POM gather mapping, velocity analysis, stack processing, and so on, the anisotropic migration results were acquired. The forward modeling computation and actual data processing demonstrate the validity of converted wave pre-stack time migration with amplitude-preservation using the anisotropic POM method.     

单程波算子地震波入射角计算

基于单程波深度延拓方法,发展了一种地震波入射角度计算方法.入射角度的计算仅利用简谐波场,可得到整个成像区域内所有点的入射波波前面方向.该方法具有较高的计算效率,可服务于合成角道集等深度偏移方法;与偏移算法相比,其计算量几乎可以忽略.与射线法或基于走时梯度的入射角度计算方法相比,本文方法更稳健,避免了速度场的微小变化导致的入射角较大变化,因此更适用于实际偏移速度模型,也与波动方程深度偏移方法更匹配.数值算例表明,本文方法既有较高的计算效率又有很好的精度,且有很好的稳定性.     

Frequency-domain double-plane-wave least-squares reverse time migration

Least-squares reverse time migration is often formulated as an iterative updating process, where estimating the gradient of the misfit function is necessary. Traditional time domain shot-profile least-squares reverse time migration is computationally expensive because computing the gradient involves solving the two-way wave equation several times in every iteration. To reduce the computational cost of least-squares reverse time migration, we propose a double-plane-wave least-squares reverse time migration method based on a misfit function for frequency-domain double-plane-wave data. In double-plane-wave least-squares reverse time migration, the gradient is computed by multiplying frequency-domain plane-wave Green's functions with the corresponding double-plane-wave data residual. Because the number of plane-wave Green's functions used for migration is relatively small, they can be pre-computed and stored in a computer's discs or memory. We can use the pre-computed plane-wave Green's functions to obtain the gradient without solving the two-way wave equation in each iteration. Therefore, the migration efficiency is significantly improved. In addition, we study the effects of using sparse frequency sampling and sparse plane-wave sampling on the proposed method. We can achieve images with correct reflector amplitudes and reasonable resolution using relatively sparse frequency sampling and plane-wave sampling, which are larger than that determined by the Nyquist theorem. The well-known wrap-around artefacts and linear artefacts generated due to under-sampling frequency and plane wave can be suppressed during iterations in cases where the sampling rates are not excessively large. Moreover, implementing the proposed method with sparse frequency sampling and sparse plane-wave sampling further improves the computational efficiency. We test the proposed double-plane-wave least-squares reverse time migration on synthetic models to show the practicality of the method.     

ANALYSIS OF THE COMMON MIDPOINT GATHER BY DECOMPOSITION INTO ELEMENTARY WAVEFRONTS1

The imaginary wavefront of a common midpoint gather of seismic data is analysed in terms of elementary wavefronts. The wavefronts emitted by the sources are split into a sequence of elementary circular wavefronts. Each such wavefront is independently transmitted through the subsurface and received by a group of receivers associated with the sources. The theory is first demonstrated in the case of a one-layer medium and is then extended to the case of a multilayer medium. We show that pairs of imaginary stationary points F and F' occur in the first medium through which all the rays, corresponding to the group of sources and receivers associated with the elementary wavefront pass. The location of the points is determined by 2n equations with 2n unknowns; (2n - 1) of these equations correspond to n layers of the medium, and an additional equation is given by the geometry of the CMP gather. Applying Fermat's principle between the points F and F', we obtain an algorithm for CMP ray tracing. The wavefront sampling problem is considered but not theoretically solved. Results obtained in the general case (a multilayer medium with lateral velocity variation) are presented and compared with those obtained by other modelling techniques.     

Inverse kinematic problem for a random gradient medium in geometric optics approximation

Scattering at random inhomogeneities in a gradient medium results in systematic deviations of the rays and travel times of refracted body waves from those corresponding to the deterministic velocity component. The character of the difference depends on the parameters of the deterministic and random velocity component. However, at great distances to the source, independently of the velocity parameters (weakly or strongly inhomogeneous medium), the most probable depth of the ray turning point is smaller than that corresponding to the deterministic velocity component, the most probable travel times also being lower. The relative uncertainty in the deterministic velocity component, derived from the mean travel times using methods developed for laterally homogeneous media (for instance, the Herglotz-Wiechert method), is systematic in character, but does not exceed the contrast of velocity inhomogeneities by magnitude. The gradient of the deterministic velocity component has a significant effect on the travel-time fluctuations. The variance at great distances to the source is mainly controlled by shallow inhomogeneities. The travel-time flucutations are studied only for weakly inhomogeneous media.     

预条件共轭梯度法在地震数据重建方法中的应用

基于最小平方的Fourier地震数据重建方法最终转化为求解一个线性方程组, 其系数矩阵是Toeplitz矩阵,可以用共轭梯度法求解该线性方程组.共轭梯度法的迭代次数受系数矩阵病态程度的影响,地震数据的非规则采样程度越高,所形成的系数矩阵病态程度越高,就越难收敛和得到合理的计算结果.本文研究了基于Toeplitz矩阵的不同预条件的构造方法,以及对共轭梯度法收敛性的影响.通过预条件的使用,加快了共轭梯度法的迭代速度, 改进了共轭梯度算法的收敛性,提高了计算的效率.数值算例和实际地震数据重建试验证明了预条件共轭梯度法对计算效率有很大的提高.     

利用矩阵分解理论的双程波方程叠前深度偏移方法

叠前深度偏移理论及方法一直是地震数据成像中研究的热点问题.业界对单程波叠前深度偏移方法和逆时深度偏移开展了深入的研究,但对双程波方程波场深度延拓理论及成像方法的研究还鲜有报道.本文以地表记录的波场值为基础,利用单程波传播算子估计波场对深度的偏导数,为在深度域求解双程波方程提供充分的边界条件,并提出利用矩阵分解理论实现双程波方程的波场深度外推.通过对强速度变化介质中传播波场的计算,与传统的单程波偏移方法相比,本文提出的偏移方法计算的波场与常规有限差分技术计算的波场相一致,证明了本方法计算的准确性.通过对SEAM模型的成像,在相同的成像参数下,与传统的单程波偏移算法和逆时深度偏移算法方法相比,本文提出的偏移方法能够提供更少的虚假成像和更清晰的成像结果.本文所提偏移算法具有深度偏移和双程波偏移的双重特色,推动和发展了双程波叠前深度偏移的理论和实践.     

TI介质局部角度域射线追踪与叠前深度偏移成像

研究与实践表明,对于长偏移距、宽方位地震数据,忽略各向异性会明显降低成像质量,影响储层预测与描述的精度.针对典型的横向各向同性(TI)介质,本文面向深度域构造成像与偏移速度分析的需要,研究基于射线理论的局部角度域叠前深度偏移成像方法.它除了像传统Kirchhoff叠前深度偏移那样输出成像剖面和炮检距域的共成像点道集,还遵循地震波在成像点处的局部方向特征、基于扩展的脉冲响应叠加原理获得入射角度域和照明角度域的成像结果.为了方便快捷地实现TI介质射线走时与局部角度信息的计算,文中讨论和对比了两种改进的射线追踪方法:一种采用从经典各向异性介质射线方程演变而来的由相速度表征的简便形式;另一种采用由对称轴垂直的TI(即VTI)介质声学近似qP波波动方程推导出来的射线方程.文中通过坐标旋转将其扩展到了对称轴倾斜的TI(即TTI)介质.国际上通用的理论模型合成数据偏移试验表明,本文方法既适用于复杂构造成像,又可为TI介质深度域偏移速度分析与模型建立提供高效的偏移引擎.     

华北地区上地幔速度结构

本文选用了北京台网记录的来自不同方位并通过华北地区的140个地震,使用平滑法和dT/d△法,得到P波慢度曲线;采取走时和慢度结合使用的方法,得到走时曲线;从Gerver-Markushevich公式出发,导出一个易于计算的简捷形式,并由此反演初始速度模型;通过模型理论走时曲线的计算,选出与实测数据符合得较好的模型。平均模型中,低速层出现的深度较浅,约60公里;在360-420公里和660-680公里深处,存在两个速度急剧增加的梯度层;500公里处可能还有一个较小的梯度层。华北的东南和西南方向上,相当于660公里梯度层的深度的差异还较为显著。     

基于高斯束传播算子的成像域走时层析成像方法

层析反演是速度建模中最重要的方法之一,结合偏移成像在成像域进行走时层析速度反演是当前比较成熟有效且广泛应用的技术.本文从高斯束偏移成像条件出发,在波动方程的一阶Born近似和Rytov近似下,推导了成像域走时扰动与速度扰动的线性关系,建立了成像域走时层析方程及其显式表达的层析核函数.该核函数的本质是有限频层析核函数,利用该核函数替换常规射线层析核函数可以明显提高层析反演精度.该核函数的计算关键是背景波场格林函数的计算,本文利用高斯束传播算子计算格林函数进而得到走时层析核函数,实现方式灵活高效且计算精度较高.基于高斯束传播算子的偏移成像与层析成像相结合进行深度域建模迭代,体现了速度建模与偏移成像一体化的思想.数值计算及实际数据应用证明了基于高斯束传播算子的成像域走时层析方法的有效性.     

Wavefield tomography based on local image correlations

The estimation of a velocity model from seismic data is a crucial step for obtaining a high‐quality image of the subsurface. Velocity estimation is usually formulated as an optimization problem where an objective function measures the mismatch between synthetic and recorded wavefields and its gradient is used to update the model. The objective function can be defined in the data‐space (as in full‐waveform inversion) or in the image space (as in migration velocity analysis). In general, the latter leads to smooth objective functions, which are monomodal in a wider basin about the global minimum compared to the objective functions defined in the data‐space. Nonetheless, migration velocity analysis requires construction of common‐image gathers at fixed spatial locations and subsampling of the image in order to assess the consistency between the trial velocity model and the observed data. We present an objective function that extracts the velocity error information directly in the image domain without analysing the information in common‐image gathers. In order to include the full complexity of the wavefield in the velocity estimation algorithm, we consider a two‐way (as opposed to one‐way) wave operator, we do not linearize the imaging operator with respect to the model parameters (as in linearized wave‐equation migration velocity analysis) and compute the gradient of the objective function using the adjoint‐state method. We illustrate our methodology with a few synthetic examples and test it on a real 2D marine streamer data set.