For deep reservoir exploration, precise inversion of the deep target is important. However, the resolution of full waveform inversion (FWI) in the deeper region may not be as fine as in the shallow region due to the acquisition geometry and complex local structure. Besides, the cycle-skipping problem is critical and significantly influences the accuracy of the inversion in FWI. In order to increase the resolution for the deep region and reduce the cycle-skipping problem, we propose a local angle domain-based inversion method. We decompose the incident and scattered wavefields around a local target into the local angle domain. Then, by the simultaneous construction of the local angle filter and the local resolution function based on the wavefield decomposition, a local angle domain multiscale inversion method with illumination compensation is conducted. Based on the convergence criterion, we construct an auto-transitional misfit function that can avoid manual intervention for the multiscale inversion process. Numerical tests proved the feasibility of the proposed strategy.

The moving high-speed train (HST) generates strong vibrations in the railway roadbed, causing seismic waves to propagate through the subsurface medium. Consequently, moving HSTs can be considered as a novel seismic source for probing subsurface structures near high-speed railways (HSRs). An HST has several carriages, making it a typical combined moving source that induces a complex interference wavefield. Seismic interferometry (SI) is a commonly used method for generating virtual shot gathers based on background noise, and the phase-shifting method (PS) is commonly used to generate a dispersion spectrum based on the constructed virtual shot gathers. Therefore, SI and PS have been used for constructing virtual shot gathers and further generating the dispersion spectrum in HST-induced seismic signal processing. Although the HST-induced seismic wavefield exhibits complex interference features, it still maintains stable and strong amplitude characteristics. Therefore, we propose a method for directly extracting the dispersion spectrum from the HST-induced seismic signal through time-frequency decomposition and similarity-based velocity scanning. Compared to the commonly used procedure (SI + PS), the proposed method avoids the virtual shot gather construction procedure. The synthetic data example and real data example have shown the proposed method’s effectiveness.

Tunnel seismic prediction (TSP) is essential for guaranteeing the safety of tunnel construction. Reverse time migration (RTM) plays a vital role in providing precise visualization of the geology located in front of the tunnel. However, anomalies like karst caves cause signal reflection and attenuation, leading to blurred images and artifacts. The point spread function (PSF) characterizes the blurring effect of a specific observing system on an imaging point, and the migration result can be viewed as the convolution of the true reflectance model with the PSF. Thus, the ambiguity of the migration result can be eliminated by using the inverse of the PSF. In this paper, we utilize the PSF in the context of TSP. First, the wavefields from the source and receiver sides are broken down into angle domain components through the Poynting vector approach. Then, the PSF operator is obtained by calculating the local illumination matrix (LIM) and is further applied to image correction. We designed various models to simulate the complex geology in front of the tunnel. Numerical experiments show that the application of PSF can improve the imaging accuracy of complex structures in TSP. And the test results of actual tunnel seismic data also demonstrate the effectiveness of this method.

As a new type of seismic source, the high-speed train (HST) seismic source has great application potential in subsurface imaging in terms of high intensity, strong repeatability and wide distribution compared with traditional seismic sources. However, the HST seismic data are severely interfered in the time-space domain due to the complex mobile combination source. Besides, the body waves are almost swamped by strong surface waves. Therefore, it is difficult to directly use this kind of seismic data for subsurface exploration. Considering the characteristic differences between the surface and body waves in the frequency-wavenumber (F-K) domain, it is more conducive to separate the surface and body waves for the HST seismic data in the F-K domain, so as to use different wavefield information in sequence. Therefore, in this study, we attempt to apply the inversion method with F-K spectrum as the objective function to HST seismic data, which aims to more effectively utilize the surface and body waves therein. In addition, to alleviate the coupling effect between P-wave and S-wave in elastic waveform inversion, we introduce the divided-offset strategy to further improve the quality of the inverted P-wave and S-wave velocities. Numerical experiments indicate that the proposed inversion method is beneficial to make better use of the surface and body waves in HST seismic data, and is expected to provide more insights on the use of HST seismic data for subsurface monitoring and detection.

The ground vibration generated by running high-speed trains (HSTs) can excite seismic wavefields with high stability and good repeatability. When the seismic waves generated by the HST source propagate in the subsurface medium, information about the geological structure of the subsurface is embedded in the wavefields, which can be used for imaging and inversion. To make better use of the HST seismic data, it is necessary to focus on the propagation characteristic and carry out a forward modeling study of the seismic wavefield generated by the HST. Since the seismic waves generated by running HST propagate through 3-D space in practice, it is vital to explore the forward modeling of the HST seismic data in 3-D space, which can obtain more accurate simulation results of the seismic wavefield and helps for a better understanding of the wavefield characteristic of the HST seismic data. In this work, we conduct full-wave forward modeling of the wavefield excited by HSTs running on viaducts in 3-D space. The moving combination source function of the HST seismic data is provided. We use the 3-D isotropic elastic wave equation with constant density in this work and employ the free surface boundary condition to ensure the simulation for both body waves and surface waves. The 3-D finite-difference method with a staggered-grid technique is applied for the forward modeling. By comparing the simulated data with the real HST seismic data, we found that there is a good similarity between them. The 3-D forward modeling of the HST seismic data in this article lays the foundation for wavefield characterization and subsurface medium inversion in the future.

The high-speed train (HST), as a special kind of seismic source, can be simplified as a mobile combination source. The HST-induced seismic data (also known as HST seismic data) contains rich geological structural information. However, HST seismic data interfere highly in space and time and their spectra have narrow-band separation spectral characteristics in frequency domain. It is very difficult to effectively extract the geological structural information contained in HST seismic data. In this work, we try to use the elastic wave equation traveltime inversion (EWTI) method to explore the subsurface velocity information using the seismic data induced by HSTs running on viaducts, which is conducive to providing a more reliable initial model to serve full waveform inversion (FWI). However, as mentioned previously, HST seismic data are highly complex, which makes the EWTI method based on first-break picking no longer applicable to them. Therefore, we apply dynamic time warping (DTW) to obtain the time shifts between observed HST-induced seismic data and synthetic HST-induced seismic data. Based on the time shifts obtained by DTW, the subsurface velocity model can be updated by establishing the misfit function and calculating the gradient. Numerical experiments using the simple velocity anomaly model and the Marmousi2 model show that the DTW-based EWTI method is able to obtain the large-scale background model and provide a reliable initial model to serve FWI. The results in this work also demonstrate the considerable potential of underground exploration using HST-induced seismic data.

The high-speed train (HST) can create a new seismic source for subsurface exploration with the advantages of good repeatability, high strength, and wide distribution. However, the complexity of this brand new seismic data also brings challenges to its application. In this paper, we try the possibility of application of the elastic reverse time migration (RTM) to the HST seismic data. The decoupled equation system is utilized to obtain the P- and S-waves deduced from the HST. Then imaging conditions based on separated wave modes are used to get the interface information of the subsurface. Numerical experiments with the Marmousi2 velocity model indicated that the subsurface structures can be well acquired, which proves the feasibility of applying the elastic RTM on the seismic data induced by HST and illustrates the great potential in utilizing the HST seismic data for subsurface exploration.

Full waveform inversion (FWI) is, as an optimization problem, strongly nonconvex and is influenced intensely by the cycle skipping issue, especially for multiparameter inversions like the elastic case. When there are strong scattering heterogeneities in the target media, there will be more challenges for the inversion problem. The direct envelope inversion strategy uses the envelope Fréchet derivative to tackle the cycle skipping problem for strong scattering inversion and has been effectively used for the acoustic case. We extend the direct envelope inversion method to the elastic situation in this letter. We derive the elastic envelope Fréchet derivative and show how the strong scattering multiparameter elastic inversion is accomplished under the direct envelope inversion framework. Numerical tests with the SEG/EAGE salt velocity model proved the effectiveness of this method for the strong scattering elastic medium.

Full waveform inversion is often troubled by falling into a local minimum due to cycle-skipping problem when missing low-frequency seismic data. According to the dynamics of seismic waves, the travel-time information will not change due to the variation in the frequency band of seismic data, which is the working mechanism of travel-time inversion. However, the current travel-time inversion method often requires various assumptions for the convenience of calculation, resulting in limited inversion effects. We present a novel velocity building method based on travel-time information. First, we use the sparse-constrained deconvolution (SCD) to convert travel-time information of seismic data into reflection sequences, which greatly reduces the complexity of the travel-time inversion. Then, the phase-independent characteristic of the envelope is introduced into the SCD to deal with the phase shift of the seismic wave. The combination of SCD and envelope greatly improves the reconstruction accuracy of the reflection sequences. Finally, the reconstructed reflection sequences are convolved with the full-band source wavelet to obtain full-band seismic data, and thus, the envelope-based SCD (E-SCD) inversion method is proposed. The results of numerical experiments on the partial basic tracking (BP) model and SEG/EAGE overthrust model verify the performance of the E-SCD inversion method. The limitations of the method and the direction of future development are also briefly discussed.

In this study, we attempt to use the high-speed train (HST)-induced seismic data for subsurface elastic parameter reconstruction by the full waveform inversion (FWI) method. The HSTs provide a new kind of seismic source as the superposition of a series of moving subsources. We consider the situation of HST running on the viaduct and use a finite difference method to simulate the seismic wave excitation. The result shows that the HST-induced seismic data are highly complex in the time domain but appear as a series of discrete peaks in the frequency domain. Elastic FWI is used to retrieve the P- and S-wave velocities of the subsurface based on this new kind of seismic data. The alternative iteration strategy and multiscale inversion are used to improve the inversion results. Numerical examples with the Marmousi2 model show the feasibility and the great potential of using this new kind of seismic data in seismic imaging.

The exploration capacity of full waveform inversion (FWI) for deep targets is affected by the acquisition geometry, complexity of overlying strata and dip angle of the target, etc. It is of great importance to analyze the relationship between the dip angle of the target reflector and the incident/scattered angle of the wavefields near the target, so as to achieve the illumination distribution and improve the inversion quality accordingly. We propose an angle domain illumination compensated FWI strategy, which utilizes the local resolution function as preconditioning to the gradient of FWI in the local angle domain, in order to improve the inversion capability for deep targets. The local resolution function describes the inversion capacity for the local target in the target dip coordinate, which can be generated from the local illumination matrix that contains the illumination information for the target from different incident and scattering directions. The Marmousi model and the SEG/EAGE salt model are used to show the validity of this method. Results from the numerical experiments prove that the proposed method can effectively improve the inversion performance as well as increase the convergence of FWI.

Full waveform inversion (FWI) is a powerful tool to understand the real complicated earth model. As FWI is a highly nonlinear problem and depends strongly on the initial model, how to effectively retrieve the large-scale background model is critical for the success of FWI. For elastic FWI (EFWI), the inversion challenge increases because the P-wave and S-wave are coupled together if no mode separation technologies are applied. In this article, we develop a new EFWI strategy, where we simultaneously implement the angle decomposition and mode separation for the wavefield. Based on the analysis of radiation patterns of different parameters and the fact that small scattering angles correspond to large-scale model perturbations, we can retrieve the large-scale background model of the P-wave velocity with pure small scattering angle P-P mode wavefield. On the other hand, the pure small scattering angle S-S, S-P, and P-S mode wavefields are used to estimate the large-scale background model of the S-wave velocity. The correctly retrieved large-scale background models further guarantee the success of subsequent fine structure retrieving for the P- and S-wave velocity models by using different wave modes. The proposed method is able to reduce the cycle-skipping problem and the multiparameter crosstalk problem simultaneously. Numerical examples show that the proposed method provides much improved inversion results than the conventional EFWI, which demonstrates the validity of the proposed method.

The blended acquisition allows multiple sources to be simulated simultaneously in a narrow time interval, which can improve the acquisition efficiency and reduce the acquisition cost tremendously. However, the overlapped information from multiple sources poses challenges for traditional seismic data migration or inversion algorithms. Thus, accurate and efficient deblending should be implemented as a pre-requisite. Traditional inversion-based deblending algorithms can provide deblended data with a high computational burden, especially for a large volume of seismic data. As a deep learning strategy can match seismic data accurately in a nonlinear way through supervised learning, we propose a U-net-based accurate deblending algorithm, which incorporates transfer learning and an iterative strategy. A set of labeled synthetic data with a blending fold of 2 are classified into the training and validation data for U-net training and validation. Field data are regarded as the test data to assess the performance of the trained U-net. To guarantee the deblending performance of the field data to some extent, parts of field data with labels are used to fine-tune the trained U-net based on transfer learning. The fine-tuning procedure is relatively fast within several minutes. To further improve the deblending performance, we incorporate an iterative strategy with the fine-tuned U-net. The deblending performance is promising in the quality and computational efficiency compared with the curvelet-thresholding-based deblending method, which demonstrates the validity of the proposed intelligent deblending method.

Strong scattering perturbations like large-scale salt structures in the model parameters make the task of full-waveform inversion more difficult than the weak scattering inversion. The problem becomes even tougher when both the velocity and density are taken into consideration because the tradeoff among the multiparameters further influences the inversion. In order to accomplish effective estimation for both the velocity and density with strong scatterings, we introduce an angle domain direct envelope inversion method with the new Fréchet derivative. The direct envelope inversion method works well on salt structure recovery for the velocity model. However, it may not work well if the density parameter is considered. By introducing angle information into the inversion, the tradeoff between velocity and density can be greatly reduced. Numerical examples using the SEG/EAGE salt model show that by accomplishing the direct envelope inversion in the angle domain, both the velocity and density estimation with strong scattering perturbations are greatly improved, which demonstrates the validity of the proposed method.

Seismic reflections at an interface are often regarded as the variation of the acoustic impedance (product of seismic velocity and density) in a media. In fact, they can also be generated due to the difference in absorption of the seismic energy. In this work, we investigate the impacts of attenuation on seismic reflections based on the diffusive-viscous wave equation, which is used to investigate seismic attenuation and frequency-dependent seismic anomalies related to hydrocarbon reservoirs. The results show that the reflections are significantly affected by the diffusive attenuation but they are insensitive to the viscous attenuation in an acoustic dispersive medium. In an elastic dispersive medium, the attenuation parameter in P wave equation has a big impact on both PP and PS reflections, however, the attenuation parameter in S wave equation has little effect on the PP reflection but it strongly affects the PS reflections. Furthermore, the PP and PS reflections in the dispersive medium are dependent on the frequency, and the effect of attenuation on PP and PS reflections at lower frequencies is bigger than those at higher frequencies.