Surface Wave Inversion Research Articles

A Surface-Wave Inversion Method Based on FHLV Loss Function in LSTM

Surface-wave analysis methods have been widely applied to construct near-surface shear-wave velocity structures. Whether it is an active source or passive source, the near-surface shear-wave velocity structure is obtained by inverting the surface-wave dispersion curve. In order to solve the problems of low inversion efficiency and poor inversion results in traditional surface-wave exploration, we have studied the surface-wave inversion methods based on deep learning technology. In this study, we propose a long short-term memory (LSTM) surface-wave inversion method based on the first height last velocity (FHLV) loss function. The core of our proposed method is the FHLV loss function consisting of two parts: a speed loss and a thickness loss, which improves the overall prediction accuracy through optimizing the learning process of the thickness parameter by the network. To verify the accuracy of the proposed LSTM surface-wave inversion method based on the FHLV loss function, experiments are conducted on both synthetic and real datasets. The results show that our proposed method can efficiently and accurately invert the near-surface shear-wave velocity structure.

Separating Scholte Wave and Body Wave in OBN Data Using Wave-Equation Migration

The ocean bottom nodes (OBNs) acquire seismic data at a challenging depth to explore the subsurface structures. The recorded body wave and Scholte wave are highly mixed and are difficult to be separated. The strong body wave would influence the high-order modes extraction using the Scholte wave, while the Scholte wave would degrade the imaging of sedimentary structures using body wave. The lacking of effective methods for separating both waves prevents their application. We developed a migration-based method to accurately separate the Scholte wave and body wave in the OBN data. First, we use high-pass filtering to divide the original OBN data into three parts: background noise, high-frequency body wave, and the mixture of Scholte wave and low-frequency body wave. Then, we separate the Scholte wave and low-frequency body wave using migration and demigration based on the fact that they have different limits of reversible-migration velocity. Finally, we generate the separated body wave by subtracting the Scholte wave from the denoised OBN data. For the off-line data, the local orthogonalization method is required to retrieve the weak leakage of Scholte wave around the apices. Theoretical analyses and numerical experiments show that the proposed method can accurately separate Scholte wave and body wave without any visible artifacts while retaining most of their inherent properties. The separated body wave provides a high-quality input for imaging sedimentary structures, and the separated Scholte wave enables the extraction of high-order modes of dispersion curve that are crucial for high-resolution surface-wave inversion.

Direct shear-wave seismic survey in Sanhu area, Qaidam Basin, west China

The formation containing shallow gas clouds poses a major challenge for conventional P-wave seismic surveys in the Sanhu area, Qaidam Basin, west China, as it dramatically attenuates seismic P-waves, resulting in high uncertainty in the subsurface structure and complexity in reservoir characterization. To address this issue, we proposed a workflow of direct shear-wave seismic (S-S) surveys. This is because the shear wave is not significantly affected by the pore fluid. Our workflow includes acquisition, processing, and interpretation in calibration with conventional P-wave seismic data to obtain improved subsurface structure images and reservoir characterization. To procure a good S-wave seismic image, several key techniques were applied: (1) a newly developed S-wave vibrator, one of the most powerful such vibrators in the world, was used to send a strong S-wave into the subsurface; (2) the acquired 9C S-S data sets initially were rotated into SH-SH and SV-SV components and subsequently were rotated into fast and slow S-wave components; and (3) a surface-wave inversion technique was applied to obtain the near-surface shear-wave velocity, used for static correction. As expected, the S-wave data were not affected by the gas clouds. This allowed us to map the subsurface structures with stronger confidence than with the P-wave data. Such S-wave data materialize into similar frequency spectra as P-wave data with a better signal-to-noise ratio. Seismic attributes were also applied to the S-wave data sets. This resulted in clearly visible geologic features that were invisible in the P-wave data.

Near Surface Characterization by Seismic Surface Wave Inversion

Surface waves are usually treated as unwanted coherent noise and steps are taken to remove them using geophone receiver arrays during seismic survey acquisition. During early data processing, attempts are made to further remove the ground roll. Recently, there has been a growing interest in using surface waves for estimating the near surface velocity model. Surface waves propagate in the shallow subsurface and can be used to derive shear wave velocities of the near surface.

Estimation of Shear-Wave Velocity Structures in Taichung, Taiwan, Using Array Measurements of Microtremors

Near-surface S-wave velocity structures (VS) are crucial in site-effect studies and ground-motion simulations or predictions. We explored S-wave velocity structures in Taichung, the second-largest city in Taiwan by population, by employing array measurements of microtremors at a total of 53 sites. First, the fundamental-mode dispersion curves of Rayleigh waves were estimated by adopting the frequency–wavenumber analysis method. Second, the surface-wave inversion technique was used to calculate the S-wave velocity structures of the area. At many sites, observed phase velocities were almost flat, with a phase velocity of approximately 800–1300 m/s in the frequency range of 0.6–2 Hz. A high-velocity zone (VS of 900–1500 m/s) with a convex shape was observed at the shallow S-wave structures of these sites (depths of 50–500 m). On the basis of the inversion results, we constructed two-dimensional and three-dimensional contour maps to elucidate the variations of VS structures in Taichung. According to VS-contour maps at different depths, lowest S-wave velocities are found at the western coastal plain, whereas highest S-wave velocities appear on the eastern side. The S-wave velocity gradually decreases from east to west. Moreover, the S-wave velocity of the Tertiary bedrock is assumed to be 1500 m/s in the area. According to the depth-contour map (VS = 1500 m/s), the depths of the bedrock range from 250 m (the eastern part) to 1550 m (the western part). The thicknesses of the alluvium gradually decrease from west to east. Our results are consistent with the geology of the Taichung area.

Application of an Automatic Noise or Signal Removal Algorithm Based on Synchrosqueezed Continuous Wavelet Transform of Passive Surface Wave Imaging: A Case Study in Sichuan, China

Passive surface wave imaging based on noise cross-correlation has been a research hotspot in recent years. However, because randomness of noise is difficult to achieve in reality, prominent noise sources will inevitably affect the dispersion measurement. Additionally, in order to recover high-fidelity surface waves, the time series input during cross-correlation calculation is usually very long, which greatly limits the efficiency of passive surface wave imaging. With an automatic noise or signal removal algorithm based on synchrosqueezed continuous wavelet transform (SS-CWT), these problems can be alleviated. We applied this method to 1-h passive datasets acquired in Sichuan province, China; separated the prominent noise events in the raw field data, and enhanced the cross-correlation reconstructed surface waves, effectively improving the accuracy of the dispersion measurement. Then, using the conventional surface wave inversion method, the shear wave velocity profile of the underground structure in this area was obtained.

3D Crustal and Upper Mantle Model of East‐Central China From a Joint Inversion of Surface and Body Waves and Its Tectonic Implications

AbstractAs the junction of two important tectonic units, North China Craton and South China Block, East‐Central China has undergone multi‐period tectonic events, including the Triassic continental collision and the Mesozoic and Cenozoic subduction of the western Pacific plate, which forms unique crustal and upper mantle structures in East‐Central China. In this study, by jointly inverting body and surface waves, we construct a high‐resolution 3D Vs model of East‐Central China at the depths of 0–800 km. Our results show that along the eastern Qinling‐Dabie orogen formed by past continental collision, there are significant changes in velocity patterns in the upper mantle. From the eastern Dabie to the western Dabie, velocity features mainly vary from low velocities to high velocities, implying a westward weakening of the influence of the Paleo‐Pacific tectonic domain. The high‐velocity anomaly is partially missing beneath eastern Qinling, which may be attributed to the lower crustal and lithospheric mantle delamination. In addition, we find a significant low‐velocity anomaly in the upper mantle beneath the Lower Yangtze Craton, which may represent an upwelling of thermal fluids caused by the dehydration of stagnant slabs in the mantle transition zone. This low‐velocity anomaly extends upward into the shallow lithosphere and is consistent with the area of crustal and lithospheric thinning as well as the concentrated exposure of Cenozoic continental basalts. These connections reveal the reactivation of the lithospheric mantle by the dehydration of the stagnant slab and the asthenosphere upwelling during the Cenozoic.

Asthenospheric flow from East Asia to the Philippine Sea Plate revealed by rj-MCMC inversion of surface waves

Asthenospheric flow from East Asia to the Philippine Sea Plate revealed by rj-MCMC inversion of surface waves

Removal of Crossed Artifacts from Multimodal Dispersion Curves with Modified Frequency–Bessel Method

ABSTRACT Frequency–Bessel (F-J) transform method can obtain higher-mode Rayleigh dispersion curves by multistation ambient noise data superposition (Wang et al., 2019). Because the dispersion curves of the overtones can provide more information compared with the single fundamental mode, the nonuniqueness of surface-wave inversion can be reduced. Because of the limited number of receivers, the integral in the process of transformation cannot be calculated precisely and there exists a kind of crossed artifacts which cuts off the real dispersion curves and contaminates the spectrum. Forbriger (2003) proposed to use the Hankel function instead of the Bessel function to conduct the transformation to remove the crossed artifacts. However, this method can reduce the resolution of the spectrum from ambient noise data. In this article, we give a complete workflow to deal with ambient noises which can eliminate the crossed artifacts without reducing the resolution. The Kramers–Kronig relations are used to obtain complete cross-correlation functions and a modified F-J transform is conducted to finally acquire the spectrum without crossed artifacts.

Multichannel Analysis of Surface Waves Accelerated (MASWAccelerated): Software for efficient surface wave inversion using MPI and GPUs

Multichannel Analysis of Surface Waves (MASW) is a technique frequently used in geotechnical engineering and engineering geophysics to infer 1D layered models of seismic shear wave velocities in the top tens to hundreds of meters of the subsurface. We aim to accelerate MASW calculations by capitalizing on modern computer hardware available in the workstations of most engineers: multiple cores and graphics processing units (GPUs). We propose new parallel and GPU accelerated algorithms for computing 1D MASW inversion, and provide software implementations in C using Message Passing Interface (MPI) and CUDA. These algorithms take advantage of sparsity that arises in the problem, and the work balance between processes considers typical data trends. We compare our methods to an existing open source Matlab MASW tool. Our serial C implementation achieves a 2x speedup over the Matlab software, and we continue to see improvements by parallelizing the problem with MPI. We see nearly perfect strong and weak scaling for uniform data, and improve strong scaling for realistic data by repartitioning the problem to process mapping. By utilizing GPUs available on most modern workstations, we observe an additional 1.3x speedup over the serial C implementation on the first use of the method. We typically repeatedly evaluate theoretical dispersion curves as part of an optimization procedure, and on the GPU the kernel can be cached for faster reuse on later runs. We observe a 3.2x speedup on the cached GPU runs compared to the serial C runs. This work is the first open-source parallel or GPU-accelerated software tool for MASW imaging, and should enable geotechnical engineers to fully utilize all computer hardware at their disposal.

Multiparameter viscoelastic full-waveform inversion of shallow-seismic surface waves with a pre-conditioned truncated Newton method

SUMMARY 2-D full-waveform inversion (FWI) of shallow-seismic Rayleigh waves has become a powerful method for reconstructing viscoelastic multiparameter models of shallow subsurface with high resolution. The multiparameter reconstruction in FWI is challenging due to the potential presence of cross-talk between different parameters and the unbalanced sensitivity of Rayleigh-wave data with respect to different parameter classes. Accounting for the inverse Hessian using truncated Newton methods based on second-order adjoint methods provides an effective tool to mitigate cross-talk caused by the coupling between different parameters. In this study, we apply a pre-conditioned truncated Newton (PTN) method to shallow-seismic FWI to simultaneously invert for multiparameter near-surface models (P- and S-wave velocities, attenuation of P and S waves, and density). We first investigate scattered wavefields caused by these parameters to evaluate the coupling between them. Then we investigate the performance of the PTN method on shallow-seismic FWI of Rayleigh wave for reconstructing all five parameters simultaneously. The application to spatially correlated and uncorrelated models demonstrates that the PTN method helps to mitigate the cross-talk and improves the resolution of the multiparameter reconstructions, especially for the weak parameters with small sensitivity such as attenuation and density parameters. The attenuation of Pwaves cannot be inverted reliably due to its negligible sensitivity on the Rayleigh wave. The comparison with the classical pre-conditioned conjugate gradient method highlights the improved performance of the PTN method and thus the benefit of accounting for the information included in the Hessian.

Joint tomographic inversion of crustal structure beneath the eastern Tibetan Plateau with ambient noise and gravity data

SUMMARY We have developed a joint tomographic inversion method with seismic surface wave dispersion and gravity data for obtaining more reliable crustal 3-D shear wave structures. We take the eikonal-based direct surface wave tomographic method and adaptive gravity modelling method in spherical coordinates in the inverse problem. Based on the empirical relations between seismic velocity and density parameters, our method combines surface wave dispersion curves (i.e. surface wave traveltimes at different periods) and Bouguer gravity anomaly data together to invert for 3-D shear wave velocity structures. In our method, off-great-circle propagation of the surface wave and the earth's curvature is considered in the forward modelling. Synthetic tests suggest that the joint tomographic method could improve the reliability and obtain more convincing results than individual seismic surface wave tomography. The gravity data can provide more constraints into the model resolution and help restore the crustal anomalies better. The inversion results in the eastern Tibetan Plateau and Sichuan basin indicate complex distributions of low-velocity zone in the mid-crust of the eastern Tibetan Plateau and a craton-like basement of the Sichuan basin, which supports the crust channel flow model. Although both the 3-D shear wave velocity model from joint inversion and the individual seismic surface wave inversion can fit the surface wave data almost equally well, the joint inversion model can better match the gravity data We also found that the 3-D model from joint inversion in this study shows similar structural characteristics with the surface wave tomographic model, which indicates the icing on the cake effects of gravity data.

Two-Dimensional Full-Waveform Joint Inversion of Surface Waves Using Phases and Z/H Ratios

Surface-wave dispersion and the Z/H ratio are important parameters used to resolve the Earth’s structure, especially for S-wave velocity. Several previous studies have explored using joint inversion of these two datasets. However, all of these studies used a 1-D depth-sensitivity kernel, which lacks precision when the structure is laterally heterogeneous. Adjoint tomography (i.e., full-waveform inversion) is a state-of-the-art imaging method with a high resolution. It can obtain better-resolved lithospheric structures beyond the resolving ability of traditional ray-based travel-time tomography. In this study, we present a systematic investigation of the 2D sensitivities of the surface wave phase and Z/H ratio using the adjoint-state method. The forward-modeling experiments indicated that the 2D phase and Z/H ratio had different sensitivities to the S-wave velocity. Thus, a full-waveform joint-inversion scheme of surface waves with phases and a Z/H ratio was proposed to take advantage of their complementary sensitivities to the Earth’s structure. Both applications to synthetic data sets in large- and small-scale inversions demonstrated the advantage of the joint inversion over the individual inversions, allowing for the creation of a more unified S-wave velocity model. The proposed joint-inversion scheme offers a computationally efficient and inexpensive alternative to imaging fine-scale shallow structures beneath a 2D seismic array.

ScanArray—A Broadband Seismological Experiment in the Baltic Shield

Abstract The ScanArray international collaborative program acquired broadband seismological data at 192 locations in the Baltic Shield during the period between 2012 and 2017. The main objective of the program is to provide seismological constraints on the structure of the lithospheric crust and mantle as well as the sublithospheric upper mantle. The new information will be applied to studies of how the lithospheric and deep structure affect observed fast topographic change and geological-tectonic evolution of the region. The program also provides new information on local seismicity, focal mechanisms, and seismic noise. The recordings are generally of very high quality and are used for analysis by various seismological methods, including P- and S-wave receiver functions for the crust and upper mantle, surface wave and ambient noise inversion for seismic velocity, body-wave P- and S-wave tomography for upper mantle velocity structure using ray and finite frequency methods, and shear-wave splitting measurements for obtaining bulk anisotropy of the upper and lowermost mantle. Here, we provide a short overview of the data acquisition and initial analysis of the new data, together with an example of integrated seismological results obtained by the project group along a representative ∼1800-km-long profile across most of the tectonic provinces in the Baltic Shield between Denmark and the North Cape. The first models support a subdivision of the Paleoproterozoic Svecofennian province into three domains, where the highest topography of the Scandes mountain range in Norway along the Atlantic Coast has developed solely in the southern and northern domains, whereas the topography is more subdued in the central domain.

Inversion of vehicle-induced signals based on seismic interferometry and recurrent neural networks

Vehicle-induced vibrations provide useful signals for passive seismic exploration. Such signals are repeatable and environmentally friendly; hence, they can provide an economical way to analyze subsurface structures. We have developed a new workflow to monitor roads or railways by producing 1D subsurface S-wave velocities in real time. This workflow consists of two steps: seismic interferometry and recurrent neural networks (RNN). Seismic interferometry can efficiently retrieve surface waves by crosscorrelating vehicle-induced vibrations. RNN was designed to first encode the picked dispersion curve into a fixed-length vector and then decode the vector into 1D S-wave velocities. To simulate the railway vibrations, we first analyze the time-dependent characteristic of the high-speed-train source and verify its mathematical expression by comparing the frequency spectrums of the real and synthetic data. We then evaluate the RNN-based surface-wave dispersion inversion method and validate the designed network structure using the 3D SEG/EAGE overthrust model. Finally, seismic interferometry and RNN-based surface-wave inversion are applied to a synthetic train-induced data set, a 33 min field record of railway vibrations and a 76 min field data of road vibrations, respectively. The synthetic and field data tests indicated that our workflow can be a feasible and cost-effective tool for real-time monitoring of subsurface media along roads and railways.

Combining Petrology and Seismology to Unravel the Plumbing System of a Typical Arc Volcano: An Example From Marapi, West Sumatra, Indonesia

AbstractMarapi in Sumatra is characterized by frequent short‐lived explosions and small eruptions (Volcanic Explosivity Index 1–2) and in the past 250 years, the volcano has erupted >60 times. Recent volcanic bombs and the presence of broadband seismic stations lead us to reconstruct the plumbing system of the volcano through an interdisciplinary study. A petrologic study of the summit bombs uses pyroxene, plagioclase and glass compositions to obtain pressures and temperatures of magma storage as well as identify pre‐eruptive processes. Two‐pyroxene geothermobarometry provides pre‐eruptive crystallization pressure estimates of 4–7 kbar (∼15–26 km). Compositional and textural analyses of plagioclase and pyroxene crystals indicate that mafic magma recharge was followed by mixing with the resident magma and renewed crystallization prior to eruption. In order to further image the magma reservoir, we performed a joint inversion of teleseismic receiver functions and surface waves (H/V ratio). The inversion reveals a low velocity zone (LVZ) at depths of 15–21 km that has 7 ± 3% shear velocity reduction, corresponding to an estimated melt fraction of 5 ± 2%. The depth of this LVZ overlaps with the depth of magma storage estimated from petrology, constraining it with high confidence. Such a combined interdisciplinary study provides valuable information for evaluating future periods of unrest, laying out a framework for the interpretation of incoming monitoring data and signals to look out for in order to improve eruption forecasts.

Aquifer Monitoring Using Ambient Seismic Noise Recorded With Distributed Acoustic Sensing (DAS) Deployed on Dark Fiber

AbstractGroundwater is a critical resource for human activities worldwide, and a vital component of many natural ecosystems. However, the state and dynamics of water‐bearing aquifers remain uncertain, mostly due to the paucity of subsurface data at high spatial and temporal resolution. Here, we show that analysis of infrastructure‐generated ambient seismic noise acquired on distributed acoustic sensing (DAS) arrays has potential as a tool to track variations in seismic velocities (dv/v) caused by groundwater level fluctuations. We analyze 5 months of ambient noise acquired along an unused, 23 km‐long telecommunication fiber‐optic cable in the Sacramento Valley, CA, a so‐called “dark fiber." Three array subsections, ∼6 km apart, are processed and the stretching technique is applied to retrieve daily dv/v beneath each location. Near the Sacramento river, dv/v variations in the order of 2%–3% correlate with precipitation events and fluctuations in river stage of ∼1.5 m. In contrast, regions away (2.5 km) from the river do not experience large dv/v variations. These observations reveal short‐scale spatial variability in aquifer dynamics captured by this approach. Dispersion analysis and surface wave inversion of noise gathers reveal that seismic velocity perturbations occur at depths of 10–30 m. Rock physics modeling confirms that observed dv/v are linked to pore pressure changes at these depths, caused by groundwater table fluctuations. Our results suggest that DAS combined with ambient noise interferometry provides a means of tracking aquifer dynamics at high spatial and temporal resolutions at local to regional scales, relevant for effective groundwater resource management.

Seismic Site Classification from Surface Wave Data to Vs,30 without Inversion

Surface wave methods have been extensively employed to estimate time-averaged shear-wave velocity (Vs) to 30 m (Vs,30) for seismic site classification. However, the inversion of surface wave dispersion data for Vs profiling is inherently computationally intensive. To overcome the challenge involved in surface-wave inversion, a new method is proposed to directly estimate Vs,30 from surface wave fundamental-mode dispersion data without inversion. The new method is based on the concept that the phase velocity of surface waves (VR) at a given frequency (f) is proportional to the average shear-wave velocity of soils within one-half of the wavelength (λ=VR/f). The proposed method starts from calculating the Vs of the top layer as it can be directly estimated from the phase velocity of surface waves whose half wavelengths are smaller than the thickness of the top layer. Likewise, the Vs of the second layer can be deduced from the top layer Vs along with the phase velocity of surface waves whose half wavelengths are within the thickness range of the second layer. Through such a repetition of sequence-deduction, the Vss of the other layers can be calculated. This method is verified with five case studies that cover typical geotechnical sites. Results indicate that the proposed method can obtain a similar Vs,30 with the same site classification for seismic design in comparison with invasive tests and inversion techniques.

The High-Speed Inversion of Rayleigh Wave and its Microtremor Application Analysis

The application of Rayleigh waves in geological prospecting and engineering has been widely tested and used as a new method of shallow seismic exploration. The previous method of surface wave inversion needed to calculate the corresponding phase velocity that satisfied the dispersion function. This process increases the computation amount and reduces the surface wave inversion efficiency. This paper proposes a method to solve the problem by using different definition methods of the objective function. The method greatly reduces the amount of computation while results in approximately equal results with traditional method. This cuts the inversion time to 1% of conventional methods and provides favorable conditions for multiple inversions to suppress multiple solutions. We set artificial initial parameters with linear constraints in our velocity increasing model test to accelerate the convergence rate of the target function and obtain accurate results quicker. Finally, this method is applied to the microtremor dispersion data with a strong disturbance, and the inversion results are consistent with the high-density electrical profile and geological survey data.

Geodynamic tomography: constraining upper-mantle deformation patterns from Bayesian inversion of surface waves

SUMMARY In the Earth’s upper mantle, seismic anisotropy mainly originates from the crystallographic preferred orientation (CPO) of olivine due to mantle deformation. Large-scale observation of anisotropy in surface wave tomography models provides unique constraints on present-day mantle flow. However, surface waves are not sensitive to the 21 coefficients of the elastic tensor, and therefore the complete anisotropic tensor cannot be resolved independently at every location. This large number of parameters may be reduced by imposing spatial smoothness and symmetry constraints to the elastic tensor. In this work, we propose to regularize the tomographic problem by using constraints from geodynamic modelling to reduce the number of model parameters. Instead of inverting for seismic velocities, we parametrize our inverse problem directly in terms of physical quantities governing mantle flow: a temperature field, and a temperature-dependent viscosity. The forward problem consists of three steps: (1) calculation of mantle flow induced by thermal anomalies, (2) calculation of the induced CPO and elastic properties using a micromechanical model, and (3) computation of azimuthally varying surface wave dispersion curves. We demonstrate how a fully nonlinear Bayesian inversion of surface wave dispersion curves can retrieve the temperature and viscosity fields, without having to explicitly parametrize the elastic tensor. Here, we consider simple flow models generated by spherical temperature anomalies. The results show that incorporating geodynamic constraints in surface wave inversion help to retrieve patterns of mantle deformation. The solution to our inversion problem is an ensemble of models (i.e. thermal structures) representing a posterior probability, therefore providing uncertainties for each model parameter.