We develop a novel comprehensive theoretical framework, the Biot-patchy-spherical-squirt (BIPSSQ) model, for wave propagation in partially saturated dual-porosity media. This model simultaneously incorporates three key fluid flow mechanisms: macroscopic flow (Biot flow or global flow), mesoscopic flow and microscopic flow (3-D spherical squirt flow). The constitutive relations and fluid pressure expressions for the BIPSSQ model are first derived and then the governing wave equations are established using a Lagrangian approach based on the system's kinetic energy, potential energy and dissipation functions. Through plane wave analysis, we obtain the phase velocity and attenuation of the fast P wave. Numerical examples demonstrate that the BIPSSQ model predicts multiple dispersion transition bands and corresponding attenuation peaks, attributed to two squirt flows and two Biot global flows from two immiscible fluids. Furthermore, the presence of squirt flow significantly suppresses the mesoscopic patchy saturation effect, leading to the disappearance of mesoscopic dispersion transition band and attenuation peak. The influences of permeability, saturation, porosity, squirt-flow length and inclusion radius on velocity dispersion and attenuation are also analysed. Finally, excellent agreement between theoretical predictions and experimental measurements from an Aksu outcrop rock sample (800 kHz), a gas-water-saturated Estaillades limestone (1 kHz) and an oil-brine-saturated Vosgian sandstone (350 kHz), validates the applicability and effectiveness of the BIPSSQ model. Moreover, the BIPSSQ model can degenerate to other theories (i.e. Biot, BISSQ, BR) under certain conditions. Our proposed model provides a unified and robust tool for interpreting wave propagation phenomena in complex, partially saturated reservoir rocks.

Traveltime tomography based on the eikonal equation and adjoint-state method plays a crucial role in seismic velocity modeling. However, conventional methods have several inherent limitations, making it challenging to reconstruct stable and accurate P- and S-wave velocity models using only traveltime information. Reasonable prior constraints need to be introduced to stabilize the inversion process. To address these issues, we develop a P- and S-wave reflection traveltime tomography method that incorporates a modified fractional-order total variation (FOTV) regularization scheme. Unlike conventional FOTV methods that rely solely on fractional-order derivatives, this scheme jointly introduces both a Tikhonov term and an FOTV term into the objective function. Tikhonov regularization provides global smoothness constraints, while FOTV selectively preserves geological boundaries. This joint strategy achieves an optimal balance between data-fitting accuracy and model reliability. It effectively avoids data over-fitting and suppresses spurious structural artifacts. We formulate the P- and S-wave traveltime tomography as a new L2-norm minimization problem augmented by this modified FOTV regularization, and solve it with the alternating direction method of multipliers (ADMM). The ADMM is adopted to decompose the objective function into independent subproblems: two standard subproblems based on the eikonal equation and Tikhonov regularization, and two FOTV regularization subproblems. Each subproblem is solved through a stepwise optimization strategy. Experimental results on both synthetic and field datasets demonstrate that the proposed method achieves significantly higher modeling accuracy. Furthermore, this technique provides robust low-wavenumber P- and S-wave velocity models, which can serve as reliable initial models for waveform inversion and migration imaging.

Field seismic data are often irregular and incomplete due to environmental constraints, compromising the continuity and coherence of raw data. To achieve high-accuracy reconstruction, we propose a novel dual-regularization reconstruction method by incorporating a fractional-order total-variation (FOTV) regularization term into compressed sensing (CS). Compared with the traditional L1 regularization method, the proposed dual-regularization method based on CS and FOTV not only considers the sparsity of signals but also takes into account the complexity of subsurface media structures, enabling better reconstruction of seismic data. Specifically, the FOTV regularization term has the ability to balance edge preservation and noise suppression, making it suitable for processing noisy and incomplete data. We solve the proposed reconstruction model by using the alternating direction method of multipliers (ADMMs). The objective function is divided into two subproblems for stepwise optimization computation, and each subproblem is solved explicitly. Testing with both synthetic and field data demonstrates that the proposed method outperforms traditional CS approaches in reconstructing complex seismic structures, while exhibiting robust noise suppression. Notably, this method enables simultaneous reconstruction and denoising, making it particularly suitable for irregularly sampled and low signal-to-noise ratio (SNR) seismic data.

Seismic attributes extracted from full frequency seismic data always show weak lateral continuity and unclear distribution when used for identifying faults. Frequency decomposition technology can improve the accuracy of characterizing faults, while the accuracy depends on the time-frequency analysis (TFA) algorithms used. As a widely used TFA method, the generalized S-transform (GST) has the ability to perform multi-resolution analysis, although it still suffers from a problem of insufficient resolution. The time-reassigned synchrosqueezing theory has effectively addressed this problem. Based on this theory, we propose a new time-reassigned synchrosqueezing transform through deriving the group delay operator (GDO) of the GST. In order to achieve the best resolution, we use a parameter matching method to determine the optimal window parameters when calculating the GST spectrum. By performing multiple synchrosqueezing calculations on the obtained spectrum, we finally obtain the time-reassigned multisynchrosqueezing generalized S transform (TMGST). Synthetic signal tests show that TMGST not only exhibits significantly higher resolution than commonly used TFA methods, but also has high flexibility. We use TMGST to extract the frequency decomposition coherence attributes from the Kerry 3D seismic dataset, a publicly available marine seismic survey from New Zealand’s Taranaki Basin, for fault identification. The results show that the coherence anomalies at the fault locations are significantly enhanced and the faults are more clearly characterized by the proposed method in this paper.

First-arrival traveltime tomography has been widely used for near-surface velocity modeling, yet it often yields unsatisfactory inverted results in regions with insufficient ray coverage, abrupt velocity changes, or smoothly varying velocity structures. Conventional regularization techniques are commonly employed to address such limitations, yet they tend to blur sharp velocity interfaces and reduce inversion accuracy. Fractional order derivatives are capable of preserving sharp velocity boundaries and mitigating the staircase effect, but the selection and optimization of the order are computationally expensive. To overcome these drawbacks, we propose an adaptive fractional order total-variation (AFOTV) regularization framework for first-arrival traveltime tomography. The proposed method can not only avoid manual tuning of the fractional order but also ensure inversion accuracy, stability, and efficiency. The objective function is decomposed into two subproblems, which are efficiently solved by the adjoint-state method with preillumination compensation and the split Bregman algorithm, respectively. Both synthetic data examples and field data applications demonstrate that, compared with the conventional inverted method and other regularized methods, the AFOTV method delivers a higher-resolution velocity model with better preserved velocity boundaries and fewer artifacts. When used as an initial model for full-waveform inversion (FWI), it also accelerates convergence while maintaining comparable accuracy. The proposed method provides an effective tool for velocity modeling under more complex near-surface conditions.

Amplitude variation with offset (AVO) inversion, capable of deriving information on subsurface elastic parameters and fluid properties, represents a pivotal technique in reservoir prediction. Compared to PP-wave AVO inversion, the joint inversion incorporates PS-wave seismic data, enhancing the accuracy and reliability of inversion results, thus providing a more comprehensive portrayal of subsurface geological formations. Both PP-wave and joint inversions can be framed as least-squares optimization problems, with the adjoint-state method serving as an effective tool for computing gradients in optimization problems. We propose a joint inversion method based on the adjoint-state method and a multi-scale inversion strategy for the inversion of three key parameters: fluid factor, shear modulus, and density. Initially, the objective function for joint inversion is constructed based on the Russell approximation, and the adjoint-state equation is derived using the Lagrangian function to compute the gradient of the objective function. Combined with the limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) algorithm, efficient unconstrained optimization is achieved. To further enhance inversion accuracy, a multi-scale inversion strategy is introduced, which progressively optimizes model parameters from coarse to fine scales by utilizing data at different frequencies in stages, thereby improving the accuracy and lateral continuity of the inversion results. Synthetic data tests demonstrate that while the PP-wave inversion based on the adjoint-state method can invert elastic parameters, the joint inversion results are more accurate. The incorporation of the multi-scale inversion strategy further enhances the stability and accuracy of the inversion results, particularly demonstrating strong adaptability in complex geological conditions. Finally, the proposed method is applied to field seismic data, with the inversion results accurately characterizing subsurface media, thereby validating the effectiveness and practicality of the method.

Reverse time migration based on the wave equation can effectively image subsurface structures, but it often suffers from amplitude imbalance and limited resolution because it relies on the adjoint operator rather than the true inverse operator. Least-squares reverse time migration (LSRTM) alleviates these issues but is computationally expensive in the data domain. In this study, we propose a hybrid-regularization-based qP-wave image-domain LSRTM method for vertically transversely isotropic (VTI) media. The Hessian operator is approximated using space-variant point-spread functions constructed via Born modeling and migration. A hybrid regularization strategy combining Tikhonov and total variation terms is employed, and the optimization problem is efficiently solved using the Split Bregman method. Numerical results on synthetic and field data demonstrate that the proposed method improves imaging resolution, amplitude balance, and structural continuity, while maintaining high computational efficiency. This approach provides an effective solution for target-oriented seismic imaging in VTI media.

Accurate separation of quasi-compressional (qP) and quasi-shear (qSV) waves in anisotropic media remains challenging because wave propagation and polarization directions are generally not aligned. To address this issue, we develop a qP- and qSV-wave decoupled acoustic–elastic coupling equations (DAECEs) for vertically transversely isotropic (VTI) media, enabling direct spatial domain decoupling of vector wavefields through numerical solutions. Starting from the elastic wave equations in VTI media, polarization analysis is employed to derive model-weighting parameters that jointly incorporate anisotropic properties and wave propagation directions. These parameters are used to construct first-order velocity decoupled equations (FOVDEs), which are subsequently coupled with acoustic-elastic coupling equations (AECEs) to establish the qP- and qSV-wave DAECEs. To determine the propagation directions required for wavefield decoupling, a two-step strategy is proposed. Initial qP- and qSV-wavefields are first obtained using simplified first decoupled equations (FDEs) independent of propagation directions. Optical-flow vectors derived from these wavefields are then used to estimate group angles, from which phase angles and propagation directions are determined through an inverse-function approach. The estimated propagation directions are subsequently incorporated into the DAECEs to achieve adaptive secondary decoupling. Numerical solution schemes and stability conditions are derived and implemented using a staggered-grid finite-difference method. Numerical experiments on homogeneous, layered, and complex VTI models demonstrate that the proposed equations accurately simulate wave propagation in seabed fluid–solid coupled media and yields high-fidelity qP and qSV vector wavefields from multicomponent seismic data. The proposed framework provides a basis for anisotropic wavefield separation and ocean-bottom seismic imaging.

Time-frequency analysis (TFA) methods are widely used techniques for local frequency analysis and seismic attribute extraction, playing a significant role in hydrocarbon-bearing reservoir prediction and geological structure interpretation. The general linear Chirplet transform (GLCT) has been widely adopted in TFA due to its superior time resolution. However, the GLCT suffers from the limitation of a fixed time window width, which restricts its adaptability to signals with varying frequency components. Furthermore, GLCT can only maintain optimal time-frequency resolution within localized frequency bands and is prone to introducing artifacts in low-frequency regions. To further enhance the time-frequency resolution, we propose a new TFA method named deconvolutive optimized GLCT (DOGLCT). The methodology consists of two main steps: first, a four-parameter-controlled Gaussian window function is introduced as the time window for optimized GLCT (OGLCT). The width of this four-parameter Gaussian window adaptively adjusts in response to frequency variations, thus ensuring guaranteed time-frequency resolution across arbitrary frequency bands. Subsequently, the time-frequency spectrum (TFS) obtained from OGLCT undergoes deconvolution with the point spread function (PSF), ultimately yielding the refined TFS of DOGLCT. The TFA results of synthetic seismic signals demonstrate that DOGLCT outperforms the conventional GLCT in terms of TF resolution and effectively addresses the issue of low-frequency artifacts. Tests on noisy seismic data validate the superior noise robustness of DOGLCT. Finally, the application of DOGLCT to field data for hydrocarbon detection confirms the enhanced time resolution of the proposed method, enabling precise identification and delineation of hydrocarbon-bearing reservoir distributions.

Multicomponent seismic exploration technology comprehensively utilizes the dynamics and kinematics characteristics of seismic waves, which can reduce the non-uniqueness of seismic imaging and predict hydrocarbon-bearing reservoirs with high accuracy. However, multicomponent seismic data are often incomplete due to the constraints from field environment and acquisition costs. To address the problem of reconstructing multicomponent seismic data, conventional compressed sensing (CS)-based algorithms reconstruct each component independently, failing to exploit implicit relationships between different components. By mapping different components to the real and imaginary parts of complex numbers, we can quantify their intrinsic correlations through the mathematical structure of complex numbers and preserve the cross-information between different components. Therefore, under the framework of CS, we construct the objective function of joint reconstruction of two-component seismic data and propose a joint reconstruction method of two-component seismic data based on CS and complex Curvelet transform (CCT). First, we introduce complex numbers to establish the implicit relationship between different components by taking them as real and imaginary parts. The complex numbers constructed from different components are then subjected to the CCT as a whole. In this process, the joint sparsity of the two components is used as a priori information for data reconstruction. Finally, the proposed reconstruction model is solved using an improved fast projection onto convex sets. Experiments with synthetic and field data demonstrate that the proposed method effectively achieves the joint reconstruction of two-component seismic data. Compared with reconstruction methods only using a single component of seismic data, the proposed method exhibits higher computational efficiency and accuracy without increasing data dimensions.

Elastic least-squares reverse time migration (ELSRTM) can enhance imaging resolution and interpret multicomponent seismic data. However, traditional ELSRTM only considers primary scattered waves and cannot accommodate secondary scattered waves in the observed records. This limitation leads to inadequate imaging of steeply dipping and complex structures, where secondary scattered waves are present. To address this issue, we propose dual-scattering elastic least-squares reverse time migration (DS-ELSRTM). We construct the objective function of DS-ELSRTM under the second-order Born approximation and derive its gradient, forming a corresponding computational algorithm and implementation steps. Additionally, we introduce improved DS-ELSRTM strategies to address the weak amplitude matching of secondary scattered waves and the non-stationary gradient issue that arises during the nonlinear process of DS-ELSRTM. By comparing the imaging results of DS-ELSRTM and conventional ELSRTM in numerical experiments with the vertical fault model and Marmousi model, it is demonstrated that the DS-ELSRTM method has advantages in imaging steeply dipping structures and complex geological structures. DS-ELSRTM can produce higher-precision images than conventional ELSRTM.

The heterogeneity of rock fabric and fluid distribution induces wave-induced fluid flow (WIFF), a mechanism that significantly affects the velocity dispersion and energy attenuation of seismic waves. We integrate pore structure (intralayer) and fluid saturation (interlayer) heterogeneity into a unified framework to establish a layered double-porosity patchy-saturated (LDPPS) porous model and obtain the potential, kinetic, and dissipation functions via the generalized Biot theory. Based on Hamilton's principle, we derive the new governing equations for the LDPPS model. Five P-waves and one S-wave are achieved through the plane wave analysis. We investigate the properties of fast P-wave and S-wave through changing relevant rock parameters. The results show that there are three attenuation peaks for fast P-wave in the whole frequency band, corresponding to mesoscopic fluid flow (MFF) (fluid heterogeneity), microscopic fluid flow (fabric heterogeneity), and macroscopic fluid flow (Biot flow). Furthermore, it is found that the model exhibits strong sensitivity to water saturation, permeability, fluid viscosity, and heterogeneity scale. Finally, we compare the dispersion and attenuation predicted by our theory with experimental data, and the results show that the theoretical predictions are consistent with reality. Since our proposed theory can account for multiscale fluid flow in the whole frequency band, it is of great significance for fluid evaluation in reservoirs.

High-precision P-wave and S-wave velocity models are very important for seabed exploration. Elastic full-waveform inversion is an effective way to obtain the velocity models. Full-waveform inversion is strongly dependent on source wavelet, and inaccurate wavelet estimation will severely affect the inversion results. Furthermore, elastic full-waveform inversion is a highly non-linear problem and has larger calculation and memory cost. Based on the deconvolution method, we develop a multi-scale source-independent elastic full-waveform inversion method in a hybrid domain to alleviate the non-linearity, to improve computation efficiency and decrease memory cost and to reduce the influence of source wavelet. We reconstruct the deconvoluted objective function and further derive the adjoint source and gradient formulas for elastic full-waveform inversion with the adjoint-state method. We can obtain good inversion results by using this objective function, adjoint source and gradient even if the source wavelet is unknown. In order to illustrate the influence of different reference traces on the objective function, we provide the theoretical formula of the objective function changing with the variation of reference traces. Experiments show that the deconvolution objective function cannot completely eliminate the influence of inaccurate source wavelet on the inversion, and the inversion effect depends on the selection of reference trace. Different inversion results of the Marmousi2 model indicate that the minimum offset trace is the best reference trace. Comparison of the inverted Marmousi2 model by the proposed method and the conventional elastic full-waveform inversion method using correct wavelet shows that the two inversion results are very close, which proves the effectiveness of the proposed method and indicates a potential application of the method in seabed exploration.

Viscoelastic wave equations based on the constant- Q (CQ) model can accurately describe the amplitude dissipation and phase distortion of waves in anelastic media. However, only three velocity or displacement components can be obtained directly by solving such equations. Starting from the time-domain second-order displacement viscoelastic wave equation, we derived the decoupled P- and S-wave displacement vector viscoelastic wave equation by using the polarization difference of P- and S-wave propagation in isotropic media. The equation can be transformed into the velocity-dilatation-rotation viscoelastic wave equation containing the first-order temporal derivative and fractional Laplacian operators, which can be solved directly by using the staggered-grid finite-difference and pseudospectral methods. We use the low-rank decomposition method to approximate the derived mixed space-wavenumber domain fractional Laplacian operators for modeling wave propagation in heterogeneous attenuating media. We also demonstrated the precision of our equation by comparing the numerical solutions with the analytical solutions. Furthermore, compared with the conventional velocity-stress viscoelastic wave equation, experimental results demonstrate that our equation can separate the pure P and S waves from the mixed wavefield during wavefield continuation. In addition, it can be separated into an equation containing predominantly an amplitude attenuation or a phase distortion term.

The propagation direction of the wavefield is particularly important for migration imaging in the reverse-time migration (RTM) of elastic waves in transversely isotropic (TI) media. However, due to the problem of the computational instability of the Poynting vector, the wave field propagation direction estimated based on the Poynting vector method has errors and cannot accurately indicate the real propagation direction of the elastic wavefield. To solve this problem, a method for calculating the optical flow vector of elastic waves in TI media is developed to obtain the propagation direction. The optical flow vector of elastic waves in TI media is determined by applying the spatial and temporal derivatives of the wavefield at each time step under the assumption that the wavefields are almost the same at subsequent time steps and are smooth in the spatial direction. As the additional smoothing item is added and the multiple iterative algorithm is introduced in calculating the optical flow vector, the direction is calculated more accurately than the Poynting vector. Based on the optical flow vectors, we can separate the source wavefield and receiver wavefield into four directions: up-going, down-going, left-going, and right-going wavefields, respectively, and finally perform elastic reverse-time migration (ERTM) imaging based on the optical flow vector traveling-wave separation. We use a layered model and the BP model to test our method. The testing results demonstrate that the optical flow vector can overcome the Poynting vector limitations and obtain more accurate and reliable information regarding the direction of elastic wave propagation in TI media, as well as precisely separate the wavefields. The separated wavefields for migration effectively improve the quality of the ERTM.

Due to the limitations of the actual acquisition environment in the field, especially in ocean bottom seismometer (OBS) acquisition, the acquired seismic data are often irregular and incomplete, which affects the subsequent data imaging, interpretation and hydrocarbon-bearing reservoir prediction. The interpolation reconstruction algorithm based on the compressive sensing theory can reconstruct the data without the limitation of Nyquist sampling interval. However, the reconstruction accuracy and effect are different for different sparse representations of the data. On the basis of compressed sensing theory, we propose a Bregman iterative seismic data reconstruction method based on the sparse decomposition of discrete orthonormal Coiflets and Symlets wavelet transforms. First, the discrete orthonormal matrix is constructed by using the above two wavelet functions to make the original seismic data sparse, then the Bregman iterative algorithm is used to reconstruct the sparse coefficients in the discrete wavelet domain, and finally the recovery matrix is used to reconstruct the seismic data. The discrete orthonormal Coiflets and Symlets wavelet transforms have good sparse representation ability and can compensate for the problem that discrete Fourier transform cannot well sparse representation of the data. After numerical experiments on horizontal-layered model, Marmousi2 model and actual data, it is verified that the proposed method can reconstruct the missing seismic data under the regular observation system and under the irregular observation system with non-uniform OBS distribution. Furthermore, this method can hardly bring the interference of random noise, and the reconstruction result is of high accuracy.

To clarify the tectonic evolution of M15 block in the Andaman Sea, we perform a delicate study of fault geometry and dynamics using a 3D seismic data. The data reveal eight sequence interfaces from the Early Oligocene to the Quaternary, large scale and multi angle extensional strike-slip faults, and a series of normal faults. The two large scale faults F1 and F2 start in the Eocene and end in the Quaternary, controlling the regional structure. The NNE-SSW strike-slip F1 fault belongs to the South Sagaing fault and the NNE-SSW strike-slip F2 is the eastern Andaman fault, the strike-slip movement of which are controlled by the impact of the collision between the Indian plate and the Eurasian plate. Through the analysis of the fault development history by the method of the ancient drop and the growth index, we find that most of the large or secondary scale faults reach the maximum drop and growth index in the Miocene, indicating that the Miocene is a significant period of plate collision enhancing and faults generating. The regional stress field is dominated by E-W tension. The continental crust has expanded rapidly from the Oligocene to the Miocene which results in the rapid subsidence of the crust. This regional stress intensity becomes weak after the Miocene. The activities of the faults caused a large difference in terrain height between the west and the east in the study area, forming a pattern of the western depression and the eastern terrace. Many NNE-SSW, NE-SW or NEE-SWW trend strike-slip faults and minor faults develop in the Miocene. It echoes the event that the convergence and subduction of the Indian plate from SW to NE direction led to the right rotation and N-NNE strike-slip of the West Myanmar block in the Miocene, thus forming a regional large strike-slip fault. All of the faults affect the structure of the region.

Seismic signals are usually nonlinear and nonstationary. The Fourier transform (FT) based on stationary signal processing theory cannot depict the frequency components at any moment. However, the time–frequency analysis (TFA) methods have the capability of describing the partial features of signal both in time and frequency domains. S transform (ST), as a common TFA method, has great time–frequency (TF) combination characteristics, but the changing trend of the window function is fixed and the TF resolution cannot be adjusted. In addition, for seismic signals, the peaks of the frequency distribution in the TF spectrum bias the actual Fourier spectrum, which will affect the accuracy of data analysis. Therefore, we propose a new TFA method called the deconvolutive improved S transform (DIST). The DIST introduces one parameter to the window function other than multiple parameters to improve the flexibility in the application process. The normalization factor is also removed from the window function to avoid the frequency bias. Moreover, the deconvolution in DIST can further improve the accuracy of TF representation. The comparison of the TFA results of synthetic seismic signals shows that the DIST has better TF resolution and energy aggregation than other TFA methods in this article. By adding different degrees of noise to synthetic seismic signals, we conclude that DIST has better noise robustness. Finally, we apply DIST to different field data for hydrocarbon detection, and the results are basically consistent with the drilling data.

Seismic signals are typical non-stationary signals, and time-frequency analysis method is a powerful tool for processing non-stationary signals, which is also widely used in seismic signal processing. S-transform is a frequently used time-frequency analysis method. However, the time and frequency resolution of S-transform is limited due to Heisenberg uncertainty principle. The post-processing method of time-frequency analysis can solve this problem by further calculation on the basis of the original time-frequency spectrum of the signal. In order to improve the time and frequency resolution, we propose a new time-frequency analysis post-processing method -time synchroextracting of generalized S-transform algorithm. We first deduce the group delay operator expression of the generalized S-transform, and then remove a large amount of fuzzy energy by extracting only the time-frequency coefficients at the group delay operator in the time-frequency spectrum to improve the time-frequency resolution. The comparative analysis of synthetic signals shows that the proposed algorithm has better time-frequency aggregation and can more accurately describe the time-frequency characteristics of seismic signals. We apply the proposed method to the fault identification of the field seismic data, and the results show that the coherent attribute slice extracted by the time-synchroextracting of the generalized S-transform can well describe the fault.

Time-frequency (TF) analysis is an important tool for seismic signal analysis, and through traditional methods, it is difficult to achieve high TF resolution and energy aggregation. In this letter, we propose the adaptive time-synchroextracting S transform (ATSEST) for seismic data processing and interpretation. The method first calculates the scale parameter of the window function through the Fourier spectrum of the signal to obtain the adaptive S transform spectrum. After that, the final result is obtained by extracting the TF coefficients at the group delay (GD) and removing a large amount of fuzzy energy in the TF spectrum. The results of the synthetic example TF analysis show that the method can improve the localization ability of transient features of seismic signals. We apply the method to the fault identification of field data, and the results show that the coherent attribute slices extracted by using ATSEST can well characterize the fault.