Abstract Grid size has a significant influence on the computation efficiency and accuracy of finite-difference seismic modeling and can change the workload of reverse time migration (RTM) remarkably. This paper proposes a non-orthogonal analytical coordinate system, beam coordinate system (BCS), for the solution of seismic wave propagation and RTM. Starting with an optical Gaussian beam width equation, we expand the representation on vertically variable velocity media, which is the most common scenario in seismic exploration. The BCS based on this representation can be used to implement an irregular-grid increment finite-difference that improves the efficiency of RTM. Based on the Laplacian expression in Riemannian space, we derive the wave equation in the BCS. The new coordinate system can generate an irregular grid with increment increasing vertically along depth. Through paraxial ray tracing, it can be extended to non-analytical beam coordinate system (NBCS). Experiments for the RTM on the Marmousi model with the BCS demonstrate that the proposed method improves the efficiency about 52% while maintaining good image quality.
Ocean Bottom Node (OBN) is a seismic data acquisition technique, comprising a hydrophone and a three-component geophone. In practice, the vertical component is susceptible to high-amplitude, low-velocity, and low-frequency shear wave noise, which negatively impacts the subsequent processing of dual-sensor data. The most commonly used method is adaptive matching subtraction, which estimates shear wave noise in the vertical component by solving an optimization problem. Neural networks, as robust nonlinear fitting tools, offer superior performance in resolving nonlinear mapping relationship and exhibit computational efficiency. In this paper, we introduce a self-supervised shear wave suppression approach for 3D OBN seismic data, using a neural network in place of the traditional adaptive matching subtraction operator. This method adopts the horizontal components as the input to the neural network, and measures the output and the noisy vertical component to establish a loss function for the network training. The network output is the predicted shear wave noise. To better balance signal leakage and noise suppression, the method incorporates a local normalized cross-correlation regularization term in the loss function. As a self-supervised method, it does not need clean data to serve as labels, thereby negating the tedious work of obtaining clean field data. Extensive experiments on both synthetic and field data demonstrate the effectiveness of the proposed method on shear wave noise suppression for 3D OBN seismic data.
Road traffic noise is a special kind of high amplitude noise in seismic or acoustic data acquisition around a road network. It is a mixture of several surface waves with different dispersion and harmonic waves. Road traffic noise is mainly generated by passing vehicles on a road. The geophones near the road will record the noise while receiving the seismic signal. The amplitude of the traffic noise is much larger than the signal, which masks the effective information and degrades the quality of acquired data. At the same time, the traffic noise is coupled with the effective signal, which makes it difficult to separate them. Therefore, attenuating traffic noise is the key to improving the quality of the final processing results. In recent years, denoising methods based on convolution neural networks (CNN) have shown good performance in noise attenuation. These denoising methods can learn the potential characteristics of acquired data, thus establishing the mapping relationship between the original data and the effective signal or noise. Here, we introduce a method combining UNet networks with asymmetric convolution blocks (ACBs) for traffic noise attenuation, and the network is called the ACB-UNet. The ACB-UNet is a supervised deep learning method, which can obtain the distribution characteristics of noise and effective signal through learning the training data and then effectively separate the two to achieve noise removal. To validate the performance of the proposed method, we apply it to synthetic and real data. The data tests show that the ACB-UNet can obtain good results for high amplitude noise attenuation and is practical and efficient.
Ocean-bottom nodes (OBNs) are widely used because of their wide azimuth, long-offset, and low-frequency advantages. However, in the vertical component of the OBN geophone, a significant amount of S-wave induced noise may be recorded. This significantly impacts the quality of the vertical component data and may affect the follow-up merging of dual-sensor data. Some commercially available methods apply normal-moveout correction, which requires velocity data. We avoid this with an adaptive matching method, which relies solely on seismic data. Therefore, we develop a novel adaptive subtraction method for S-wave leakage suppression using the horizontal component as the noise model. Our method effectively handles the nonstationarity of the input seismic data in all time and space directions, mitigating instability caused by manual selection of the smoothing radius. A method is introduced for estimating a global nonstationary smoothing radius using the noise model. Compared with commercial and stationary smoothing methods, our method can better suppress S-wave noise while balancing residual noise and signal leakage more effectively. Synthetic and field data examples demonstrate significant improvement with our method.
To realize the reduction of calculation time and memory usage for seismic wave modeling on large scale model, we propose a 3D trapezoid-grid finite difference time domain (FDTD) method. It adopts the size-increasing trapezoid-grid mesh to fits the increasing trend in depth of seismic wave velocity, which can significantly reduce the oversampling in high-velocity region. The trapezoid transformation is used to alleviate the difficulty of processing irregular grids. We derive the 3D acoustic equation with the Convolutional Perfectly Matched Layer (CPML) absorbing boundary condition in the trapezoid coordinate system. Numerical tests are given to verify the effectiveness of our method. With comparable accuracy, our method can achieve about 54.1% improvement on efficiency and 88.7% reduction on memory compared with regular-grid FDTD method.
The beam forming result is crucial for the speed and quality of the final beam migration image. We propose a beam migration algorithm based on the least square inversion beam forming technique. Marching pursuit is evoked to solve the least square problem. Preliminary test result demonstrates the proposed method achieves high resolution slant stack result in the τ-P domain. It shows that with the least square inversion slant stacking, the imaging results have a better performance on irregular data set and seismic data with random noise.
On land seismic exploration, especially in areas of complex topography such as mountainous regions, foothills, and thrust belts, implementing FWI faces great challenges because this technique is required to accurately solve the wave equation with complex free-surface boundaries of arbitrary shape and to effectively reconstruct a highly heterogeneous subsurface media. Conventional finite-difference-based FWI (FDFWI) cannot be successful in these challenging areas because it always suffers from the staircase effect inherent to the underlying Cartesian structured grid. Compared to FDFWI, the finite-element-based FWI with unstructured grids is more suitable for these complicated problems. In recent years, the discontinuous Galerkin (DG) finite-element method has received considerable attention due to the accurate discretization of complex topography boundaries and the use of adaptive interface-fitting tetrahedral (3D) or triangular (2D) meshes that can match local variations of heterogeneous medium properties. Here we use the DG method as forward modeling engine of acoustic FWI, establishing an acoustic DGFWI workflow for land data from areas of strong topography changes. We apply the DGFWI on a 2D land data set from a mountainous area in Tarim basin in China. The data with low frequency down to 6.0 Hz acquired from a routine geometry that is not specifically designed for FWI. We demonstrate the technical advantages of the DG method in forward modeling in areas of complex topography and show the capability of DGFWI in building a highly resolved near-surface velocity model. We also illustrate the effectiveness of low-frequency extrapolation in improving the DGFWI resolution on land data. Our work will provide guidance for the land FWI strategy in complex areas of strong surface topography variations. Presentation Date: Tuesday, October 13, 2020 Session Start Time: 9:20 AM Presentation Time: 9:20 AM Location: Poster Station 3 Presentation Type: Poster
An accurate seismic facies analysis (SFA) can provide insight into the subsurface sedimentary facies and has guiding significance for geological exploration. Many machine learning algorithms, including unsupervised, supervised, and deep learning algorithms, have been developed successfully for SFA over the past decades. However, SFA and facies classification are still challenging tasks due to the complex characteristics of geological and seismic data. A multiattribute SOM-K-means clustering algorithm, which implements a two-stage clustering by using multiple geological attributes, is proposed and applied for SFA. The proposed algorithm can effectively extract complementary features from the multiple attribute volumes and comprehensively use the different attributes to improve the recognition ability of seismic facies. Experimental results show that the proposed algorithm improves clustering accuracy and can be used as an effective and powerful tool for SFA.
Summary Slant stack Kirchhoff beam migration presented by Yonghe Sun et al. (2000) showed potential improvement in respect to computational efficiency and image quality. Patches of seismic data are first stacked into beam volume and then projected into imaging volume using traveltime originating in the patch center. In this work, we try to add first-hand experiences of this approach by implementing it to the 3D TTI SEAM model. The details are discussed and the final image, computation time and the crucial parameters are tested. Basically it can achieve high S/N ratio image comparing to Kirchhoff method. The tests show that there is trade-off between imaging quality and migration efficiency. The user defined parameters have huge effects on the image results and efficiency, and it is a tedious work to find the best parameter combinations.
Beam forming is one of the important steps for beam migration. Based on the alternating directions method of the augmented Lagrangian multiplier method (ADMM), a robust sparse linear radon transform (SLRT) for beam forming and migration is proposed in order to improve the representation of intermediate beam data and the quality of final image result. The ADMM splits sparse constrain the least-square problem into two main parts: updating least-square solutions and thresholding, which can handle sparse, near-sparse and even non-sparse optimization problems. We tailor it for SLRT of local seismic data through introducing the fast algorithm of a Hamiltonian matrix and recasting regularization parameters and threshold variable for soft-thresholding. Compared with matching pursuit (MP), it is faster with an average four iterations, increased robustness and improved performance for possible non-strict sparse situations. Experiments for beam migration show that the proposed method could obtain better results while maintaining good performance.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)