Velocity model updating using image gathers1

For successful prestack depth migration an accurate velocity model is needed. One method for model updating is based on image gather analysis. In an image gather all reflectors line up horizontally if the correct velocities are used for the depth migration. This is also true for dipping reflectors, as all traces of an image gather belong to the same surface coordinate. The images of the reflector in an image gather curve upwards if the velocity used for the migration is too low, or downwards if the velocity is too high. This deviation can be used for model updating. Curves which depend on depth, offset and a parameter which relates the estimated to the true model are fitted to the image. By calculating the coherence along the deviation curves, this parameter can be estimated and hence an update can be calculated. Formulae are derived for the deviation curves and the update of the velocity depth model for a multilayered model for both shot and common-offset migrated data, with and without gradients. The method is tested on synthetic data with satisfactory results.     

Migration to multiple offset and velocity analysis1

The stacking velocity best characterizes the normal moveout curves in a common-mid-point gather, while the migration velocity characterizes the diffraction curves in a zero-offset section as well as in a common-midpoint gather. For horizontally layered media, the two velocity types coincide due to the conformance of the normal and the image ray. In the case of dipping subsurface structures, stacking velocities depend on the dip of the reflector and relate to normal rays, but with a dip-dependent lateral smear of the reflection point. After dip-moveout correction, the stacking velocities are reduced while the reflection-point smear vanishes, focusing the rays on the common reflection points. For homogeneous media the dip-moveout correction is independent of the actual velocity and can be applied as a dip-moveout correction to multiple offset before velocity analysis. Migration to multiple offset is a prestack, time-migration technique, which presents data sets which mimic high-fold, bin-centre adjusted, common-midpoint gathers. This method is independent of velocity and can migrate any 2D or 3D data set with arbitrary acquisition geometry. The gathers generated can be analysed for normal-moveout velocities using traditional methods such as the interpretation of multivelocity-function stacks. These stacks, however, are equivalent to multi-velocity-function time migrations and the derived velocities are migration velocities.     

Establishing an effective offset-dependent velocity model by ray tracing and its application in Kirchhoff time migration

Conventional Kirchhoff prestack time migration based on the hyperbolic moveout can cause ambiguity in laterally inhomogeneous media, because the root mean square velocity corresponds to a one-dimensional model under the horizontal layer assumption; it does not include the lateral variations. The shot/receiver configuration with different offsets and azimuths should adopt different migration velocities as they contribute to a single image point. Therefore, we propose to use an offset-vector to describe the lateral variations through an offset-dependent velocity corresponding to the difference in offset from surface points to the image point. The offset-vector is decomposed into orthogonal directions along the in-line and cross-line directions so that the single velocity can be expressed as a series of actual velocities. We use a simple Snell's law-based ray tracing to calculate the travel time recorded at the image point and convert the travel time to an equivalent velocity corresponding to a pseudo-straight ray. The double-square-root equation using such an equivalent velocity in the offset-vector domain is non-hyperbolic and asymmetrical, which improves the accuracy of the migration. Numerical examples using the Marmousi model and a wide azimuth field data show that the proposed method can achieve reasonable accuracy and significantly enhances the imaging of complex structures.     

探地雷达图像的正演合成与偏移处理

本文讨论了射线追踪法实现二维地电断面的探地雷达图像的正演合成问题,其中包括二维地电断面的数字拟合、多层连续界面介质中高频电磁波的反射与透射统一的运动学边界方程;根据Huygen's-Fresnel原理导出了均匀介质中二维管状体模型的衍射雷达波的计算公式;给出了用射线追踪法合成的二维地电模型的探地雷达图像;用衍射波公式合成了二维管状模型的雷达图像,并与物理模拟作了比较.在此基础上,用Kirchhoff积分法对数字模拟与物理模拟记录的雷达图像作了偏移处理,并给出了实例.     

Using multicomponent GPR to monitor cracks in a historical building

We tested the usefulness of acquiring multicomponent GPR data to detect cracks in a historical building, and to monitor their dynamics, caused by a slowly and irregularly moving landslide. We used 2 GHz bipolar antennas in a configuration that allows for acquiring the in-line and cross-line electric field components with x- and y-directed antennas. The 2 × 2 data matrix was collected on a floor in the building along transects at four different times over a period of one year. The data were processed with a standard 2D scalar algorithm and with the latest 3D single component vector algorithm that corrects for antenna effects. We have implemented a 3D single component vector migration algorithm in a 2.5D sense to produce 2D slices of a 3D vector migration image by applying the algorithm on line data. This procedure allows for migrating single component line data taking into account all vector effects as well as three-dimensional wave propagation. We show that the 2.5D vector migration images have a much better in-line resolution than the migration images obtained by applying a standard 2D scalar migration algorithm.The GPR profiles agree with the a priori information about the structure of the floor. In particular, we detected two different types of anomalies, only a few of which can be due to utilities and to metallic mesh. Some shallower anomalies agree well with the cracks visible on the tiling, suggesting that some cracks can be directly detected using GPR. Visually there were no changes in the cracks on the floor and no clear changes in the GPR data could be attributed to possible subsurface changes in the cracks. The variations in the GPR images seemed primarily caused by changes in the coupling of the antennas with the investigated structure (floor) depending on the season when the measurements were made. For this reason the monitoring aspect of the survey is not successful.     

Layer-stripping reverse-time migration1

We present a layer-stripping method of migration for irregularly layered media in which first-order velocity discontinuities separate regions of constant or smoothly varying velocity. We use the reverse-time method to migrate seismic data layer by layer, from the surface downwards. As part of the migration of a given layer, the bottom boundary of the layer is defined based on power in the migrated signal, and a seismic section is collected along it. This new section serves as the boundary condition for migration in the next layer. This procedure is repeated for each layer, with the final image formed from the individual layer images. Layer-stripping migration consists of three steps: (1) layer definition, (2) wavefield extrapolation and imaging, and (3) boundary determination. The migration scheme when used with reverse-time extrapolation is similar to datuming with an imaging condition. The reverse-time method uses an explicit fourth-order time, tenth-order space, finite-difference approximation to the scalar wave equation. The advantages of layer-stripping reverse-time migration are: (1) it preserves the benefits of the reverse-time method by handling strong velocity contrasts between layers and steeply dipping structures; (2) it reduces computer memory and saves computation time in high-velocity layers, and (3) it allows interpretational control of the image. Post-stack layer-stripping reverse-time migration is illustrated with a synthetic CMP data example. Prestack migration is illustrated with a synthetic data set and with a marine seismic reflection profile across the Santa Maria Basin and the Hosgri Fault in central California.     

Double plane‐wave reverse‐time migration

We develop a new time‐domain reverse‐time migration method called double plane‐wave reverse‐time migration that uses plane‐wave transformed gathers. Original shot gathers with appropriate data acquisition geometry are double slant stacked into the double plane‐wave domain with minimal slant stacking artefacts. The range of plane‐wave components needed for migration can be determined by estimating the maximum time dips present in shot gathers. This reduces the total number of input traces for migration and increases migration efficiency. Unlike the pre‐stack shot‐profile reverse‐time migration where the number of forward propagations is proportional to the number of shots, the number of forward propagations needed for the proposed method remains constant and is relatively small even for large seismic datasets. Therefore, the proposed method can improve the efficiency of the migration and be suitable for migrating large datasets. Double plane‐wave reverse‐time migration can be performed for selected plane‐wave components to obtain subsurface interfaces with different dips, which makes the migration method target oriented. This feature also makes the method a useful tool for migration velocity analysis. For example, we are able to promptly obtain trial images with nearly horizontal interfaces and adjust velocity models according to common image gathers. Seismic signal coming from steeply dipping interfaces can be included into the migration to build images with more detailed structures and higher spatial resolution as better velocity models become available. Illumination compensation imaging conditions for the proposed method are also introduced to obtain images with balanced amplitudes.     

起伏地表下的直接叠前时间偏移

提出了一种新的叠前时间偏移方法和流程,可不必应用野外静校正,直接对起伏地表采集的地震数据进行叠前时间偏移.本文采用输入道成像方式,通过基于稳相点原理给出单道数据的走时和振幅计算方法,发展了一个表驱动的叠前时间偏移算法.偏移方法可依据同相轴是否被拉平确定叠加速度和修正近地表速度模型,也可依据拟成像的构造倾角,自适应地确定偏移孔径;后者既减少了偏移计算量,也压制了偏移噪声.文中用二维起伏地表的断陷盆地模型的理论数据验证了所发展方法的成像效果.     

Application of sub-image multiresolution analysis of Ground-penetrating radar data in a study of shallow structures

Fourier-based algorithms originally developed for the processing of seismic data are applied routinely in the Ground-penetrating radar (GPR) data processing, but these conventional methods of data processing may result in an abundance of spurious harmonics without any geological meaning. We propose a new approach in this study based essentially on multiresolution wavelet analysis (MRA) for GPR noise suppression. The 2D GPR section is similar to an image in all aspects if we consider each data point of the GPR section to be an image pixel in general. This technique is an image analysis with sub-image decomposition. We start from the basic image decomposition procedure using conventional MRA approach and establish the filter bank accordingly. With reasonable knowledge of data and noise and the basic assumption of the target, it is possible to determine the components with high S/N ratio and eliminate noisy components. The MRA procedure is performed further for the components containing both signal and noise. We treated the selected component as an original image and applied the MRA procedure again to that single component with a mother wavelet of higher resolution. This recursive procedure with finer input allows us to extract features or noise events from GPR data more effectively than conventional process.To assess the performance of the MRA filtering method, we first test this method on a simple synthetic model and then on experimental data acquired from a control site using 400 MHz GPR system. A comparison of results from our method and from conventional filtering techniques demonstrates the effectiveness of the sub-image MRA method, particularly in removing ringing noise and scattering events. Field study was carried out in a trenched fault zone where a faulting structure was present at shallow depths ready for understanding the feasibility of improving the data S/N ratio by applying the sub-image multiresolution analysis. In contrast to the conventional methods, the MRA sub-image filtering technique provides an overall improvement in image quality of the data as shown in the field study.     

Removal of overburden velocity anomaly effects for depth conversion

A method of compensating for the presence of discrete overburden velocity anomalies during depth conversion of time horizons interpreted on conventional, post-stack time-migrated seismic data is presented. Positive and negative time delays are estimated either from the push-down or pull-up of reflectors directly beneath the anomalies or from interpreted time thickness of the anomalous body and interval velocities estimated from well data. The critical steps are pre‐stack simulation of seismic acquisition across the velocity anomalies, incorporating the effects of a Fresnel volume which changes its width as a function of depth, and simulation of common-midpoint (CMP) stacking using a linear regression of time delay, Δ t , versus offset-squared, X ². The time-correction method predicts the time distortion for any target horizon and the distortion is removed as a correction in time. Depth conversion is then performed using a background velocity function. The final average velocity map is calculated from the resulting depth structure and the raw times at the target horizon. The method is implemented by manipulating time grids within an industry-standard mapping package. The final average velocity map shows steep lateral velocity gradients which are constrained by the interpreted boundaries of the velocity anomalies.     

Improving Kirchhoff migration with repeated local plane-wave imaging? A SAR-inspired signal-processing approach in prestack depth imaging

A local plane-wave approach of generalized diffraction tomography in heterogeneous backgrounds, equivalent to Kirchhoff summation techniques when applied in seismic reflection, is re-programmed to act as repeated synthetic aperture radar (SAR) imaging for seismic prestack depth migration. Spotlight-mode SAR imaging quickly provides good images of the electromagnetic reflectivity of the ground via fast Fourier transform (FFT)-based signal processing. By calculating only the Green's functions connecting the aircraft to the centre of the illuminated patch, scattering structures around that centre are also recovered. SAR technology requires us to examine seismic imaging from the local point of view, where the quantity and quality of the available information at each image point are what are important, regardless of the survey geometry. When adapted to seismics, a local image of arbitrary size and sampling is obtained by FFT of seismic energy maps in the scattering wavenumber domain around each node of a pre-calculated grid of Green's functions. These local images can be used to generate a classic prestack depth-migrated section by collecting only their centres. However, the local images also provide valuable information around the centre, as in SAR. They can therefore help to pre-analyse prestack depth migration efficiently, and to perform velocity analysis at a very low cost. The FFT-based signal-processing approach allows local, efficient and automatic control of anti-aliasing, noise and resolution, including optimized Jacobian weights. Repeated local imaging could also be used to speed up migration, with interpolation between local images associated with a coarse grid of Green's functions, as an alternative to interpolation of Green's functions. The local images may, however, show distortions due to the local plane-wave approximation, and the velocity variations across their frame. Such effects, which are not necessarily a problem in SAR, should be controlled and corrected to further enhance seismic imaging. Applications to realistic models and to real data show that, despite the distortion effects, the local images can yield similar information to prestack depth migration, including common-image-point gathers for velocity analyses and AVO/AVA effects, at a much lower cost when a small target is considered.     

Migration methods for imaging different-order multiples

Multiples contain valuable information about the subsurface, and if properly migrated can provide a wider illumination of the subsurface compared to imaging with VSP primary reflections. In this paper we review three different methods for migrating multiples. The first method is model-based, and it is more sensitive to velocity errors than primary migration; the second method uses a semi-natural Green's function for migrating multiples, where part of the traveltimes are computed from the velocity model, and part of the traveltimes (i.e., natural traveltimes) are picked from the data to construct the imaging condition for multiples; the third method uses cross-correlation of traces. The last two methods are preferred in the sense that they are significantly less sensitive to velocity errors and statics because they use “natural data” to construct part of the migration imaging conditions. Compared with the interferometric (i.e., crosscorrelation) imaging method the semi-natural Green's function method is more computationally efficient and is sometimes less prone to migration artifacts. Numerical tests with 2-D and 3-D VSP data show that a wider subsurface coverage, higher-fold and more balanced illumination of the subsurface can be achieved with multiple migration compared with migration of primary reflections only. However, there can be strong interference from multiples with different orders or primaries when multiples of high order are migrated. One possible solution is to filter primaries and different orders of multiples before migration, and another possible solution is least squares migration of all events. A limitation of multiple migration is encountered for subsalt imaging. Here, the multiples must pass through the salt body more than twice, which amplifies the distortion of the image.     

Offset continuation (OCO) ray tracing using OCO trajectories

Offset continuation (OCO) is a seismic configuration transform designed to simulate a seismic section as if obtained with a certain source-receiver offset using the data measured with another offset. Since OCO is dependent on the velocity model used in the process, comparison of the simulated section to an acquired section allows for the extraction of velocity information. An algorithm for such a horizon-oriented velocity analysis is based on so-called OCO rays. These OCO rays describe the output point of an OCO as a function of the Root Mean Square (RMS) velocity. The intersection point of an OCO ray with the picked traveltime curve in the acquired data corresponding to the output half-offset defines the RMS velocity at that position. We theoretically relate the OCO rays to the kinematic properties of OCO image waves that describe the continuous transformation of the common-offset reflection event from one offset to another. By applying the method of characteristics to the OCO image-wave equation, we obtain a raytracing-like procedure that allows to construct OCO trajectories describing the position of the OCO output point under varying offset. The endpoints of these OCO trajectories for a single input point and different values of the RMS velocity form then the OCO rays. A numerical example demonstrates that the developed ray-tracing procedure leads to reliable OCO rays, which in turn provide high-quality RMS velocities. The proposed procedure can be carried out fully automatically, while conventional velocity analysis needs human intervention. Moreover, since velocities are extracted using offset sections, more redundancy is available or, alternatively, OCO velocities can be studied as a function of offset.     

Piecewise‐continuous velocity inversion

We suggest a new method to determine the piecewise‐continuous vertical distribution of instantaneous velocities within sediment layers, using different order time‐domain effective velocities on their top and bottom points. We demonstrate our method using a synthetic model that consists of different compacted sediment layers characterized by monotonously increasing velocity, combined with hard rock layers, such as salt or basalt, characterized by constant fast velocities, and low velocity layers, such as gas pockets. We first show that, by using only the root‐mean‐square velocities and the corresponding vertical travel times (computed from the original instantaneous velocity in depth) as input for a Dix‐type inversion, many different vertical distributions of the instantaneous velocities can be obtained (inverted). Some geological constraints, such as limiting the values of the inverted vertical velocity gradients, should be applied in order to obtain more geologically plausible velocity profiles. In order to limit the non‐uniqueness of the inverted velocities, additional information should be added. We have derived three different inversion solutions that yield the correct instantaneous velocity, avoiding any a priori geological constraints. The additional data at the interface points contain either the average velocities (or depths) or the fourth‐order average velocities, or both. Practically, average velocities can be obtained from nearby wells, whereas the fourth‐order average velocity can be estimated from the quartic moveout term during velocity analysis. Along with the three different types of input, we consider two types of vertical velocity models within each interval: distribution with a constant velocity gradient and an exponential asymptotically bounded velocity model, which is in particular important for modelling thick layers. It has been shown that, in the case of thin intervals, both models lead to similar results. The method allows us to establish the instantaneous velocities at the top and bottom interfaces, where the velocity profile inside the intervals is given by either the linear or the exponential asymptotically bounded velocity models. Since the velocity parameters of each interval are independently inverted, discontinuities of the instantaneous velocity at the interfaces occur naturally. The improved accuracy of the inverted instantaneous velocities is particularly important for accurate time‐to‐depth conversion.     

Vertical image waves in elliptically anisotropic media

By reparameterization of the kinematic expressions for remigration in elliptically anisotropic media using a new ellipticity parameter, we derive a new image-wave equation in elliptically anisotropic media describing the position of the reflector as a function of the medium ellipticity. This image wave equation, which is a kind of medium-dependent one-way wave equation, can be used for automatically stretching a time-migrated image in depth until wells are tied or other given geologic criteria are met. In this way, it provides a useful means to use a priori depth information for finding an estimate of the vertical velocity, which cannot be detected from time processing only. Simple numerical examples confirm the validity of the image-wave equation even for nonconstant velocity.     

基于Hilbert变换的全波场分离逆时偏移成像

逆时偏移方法利用双程波算子模拟波场的正向和反向传播,通常采用互相关成像条件获得偏移剖面,是一种高精度的成像方法.但是传统的互相关成像条件会在偏移结果中产生低频噪声;此外,如果偏移速度中存在剧烈速度变化还可能进一步产生偏移假象.为了提高逆时偏移的成像质量,可在成像过程中先对震源波场和检波点波场分别进行波场分离,然后选择合适的波场成分进行互相关成像.本文基于Hilbert变换,推导了可在偏移过程中进行上下行和左右行波场分离的高效波场分离公式以及相应的成像条件,结合Sigsbee 2B合成数据,给出了不同波场成分的互相关成像结果.数值算例结果表明,采用本文提出的高效波场分离算法以及合理的波场成分互相关成像条件可以获得高信噪比的成像结果.     

Subsurface offset behaviour in velocity analysis with extended reflectivity images

Migration velocity analysis with the constant‐density acoustic wave equation can be accomplished by the focusing of extended migration images, obtained by introducing a subsurface shift in the imaging condition. A reflector in a wrong velocity model will show up as a curve in the extended image. In the correct model, it should collapse to a point. The usual approach to obtain a focused image involves a cost functional that penalizes energy in the extended image at non‐zero shift. Its minimization by a gradient‐based method should then produce the correct velocity model. Here, asymptotic analysis and numerical examples show that this method may be too sensitive to amplitude peaks at large shifts at the wrong depth and to artefacts. A more robust alternative is proposed that can be interpreted as a generalization of stack power and maximizes the energy at zero‐subsurface shift. A real‐data example is included.     

在成像空间中衰减多次波方法研究

在偏移后成像空间中的共成像点道集中可以对多次波进行衰减,对于给定的偏移速度模型,一次波与多次波在叠前偏移后的共成像点道集中具有不同的动校时差,这样我们就可以使用类似于偏移前衰减多次波的方法将一次波和多次波进行分离.本文在成像空间中应用抛物Radon变换分离多次波和有效波,由于每个共成像点道集都包含了复杂三维波场传播效应,所以本文方法具有处理三维数据和复杂地下构造的能力.相比于SRME以及传统Radon变换衰减多次波方法,本文方法能够在保持较小的计算量的同时,保证了衰减多次波的准确性.通过对模型数据试算和对实际数据的处理验证了本文方法在叠前时间偏移后衰减多次波的能力,并取得了很好的成像效果.     

联合速度建模及其在反射波法隧道超前预报中的应用

We propose a combined migration velocity analysis and imaging method based on Kirchhoff integral migration and reverse time migration, using the residual curvature analysis and layer stripping strategy to build the velocity model. This method improves the image resolution of Kirchhoff integral migration and reduces the computations of the reverse time migration. It combines the advantages of efficiency and accuracy of the two migration methods. Its application in tunnel seismic prediction shows good results. Numerical experiments show that the imaging results of reverse time migration are better than the imaging results of Kirchhoff integral migration in many aspects of tunnel prediction. Field data show that this method has efficient computations and can establish a reasonable velocity model and a high quality imaging section. Combination with geological information can make an accurate prediction of the front of the tunnel geological structure.     

基于全波形反演的探地雷达数据逆时偏移成像

逆时偏移成像(RTM)常用来处理复杂速度模型,包括陡倾角及横向速度变化剧烈的模型.与常规偏移成像方法(如Kirchhoff偏移)相比,逆时偏移成像能提供更好的偏移成像结果,近些年逆时偏移成像越来越广泛地应用到勘探地震中,它逐渐成为石油地震勘探中的一种行业标准.电磁波和弹性波在动力学和运动学上存在相似性,故本文开发了基于麦克斯韦方程组的电磁波逆时偏移成像算法,并将其应用到探地雷达数据处理中.时间域有限差分(FDTD)用于模拟电磁波正向和逆向传播过程,互相关成像条件用于获得最终偏移结果.逆时偏移成像算法中,偏移成像结果受初始模型影响较大,而其中决定电磁波传播速度的介电常数的影响尤为重要.本文基于时间域全波形反演(FWI)算法反演获得了更为精确的地下介电常数模型,并将其反演结果作为逆时偏移成像的初始介电常数模型.为了验证此算法的有效性,首先构建了一个复杂地质结构模型,合成了共偏移距及共炮点探地雷达数据,分别应用常规Kirchhoff偏移算法及逆时偏移成像算法进行偏移处理,成像结果显示由逆时偏移成像算法得到的偏移结果与实际模型具有较高的一致性;此外本文在室内沙槽中进行了相关的物理模拟实验,采集了共偏移距及共炮点探地雷达数据,分别应用Kirchhoff和叠前逆时偏移成像算法进行处理,结果表明叠前逆时偏移成像在实际应用中能获得更好的成像效果.