Real Seismic Data Research Articles

A time-varying wavelet phase extraction method using the wavelet amplitude spectra

ABSTRACTThe vertical resolution of seismic data is greatly affected by time-varying seismic wavelets, and the phase extraction is the key topic for the accurate wavelet extraction. The traditional methods are limited to the piecewise stationary and constant-phase assumptions, so that the extracted results are not satisfactory. In this paper, we try to explore a new time-varying wavelet phase extraction method using the wavelet amplitude information. The propagation and attenuation law of seismic wavelets is analyzed, and the relationship between the amplitude and phase spectra of time-varying wavelets is obtained by deriving the wave equation for a viscoelastic medium. We find that accurately estimating the amplitude spectra at different time points and the initial phase spectrum is necessary to extract accurate time-varying wavelet phase spectra. Based on this conclusion, the wavelet amplitude spectra and initial phase spectrum are estimated based on time–frequency spectral modelling and higher-order cumulants, respectively, so that the phase spectra of time-varying wavelets are estimated by applying the obtained relationship. The numerical simulation and real seismic data processing results demonstrate that the proposed method can improve the accuracy of time-varying wavelet phase extraction compared to the traditional method.

Regularisation parameter adaptive selection and its application in the prestack AVO inversion

The numerical solution of the inverse problem is usually obtained by solving a set of linear algebraic equations, while the system of equations may suffer from ill-posedness due to insufficient data. Regularisation is a technique for making the estimation problems well posed by adding indirect constraints on the estimated model, but the regularisation parameter selection is difficult. In geophysics, without explicit calculation methods and quantitative evaluation criteria, it is usually based on the experience of the inversion engineers to try to achieve the best inversion results by continuously modifying the regularisation parameter. For prestack amplitude variation with offset (AVO) inversion for real seismic data, fixed regularisation parameters cannot satisfy the optimisation conditions in the seismic data with different signal-to-noise (SNR) in one area. Besides, fixed regularisation parameter may cause that the model constraint misfit is too large or too small compared to data misfit, which may guide the inversion to generate undesirable results. Therefore, adaptive selection of regularisation parameter according to the seismic data can help guarantee a good inversion result. Based on the traditional L-curve criterion, we derive the theoretical formula of the adaptive computation of the regularisation parameter, which can be applied to any norm constraint. We proposed the application of this selection scheme in prestack AVO inversion. Model tests show that the improved L-curve method has better stability than its main competitor, the generalised cross-validation (GCV) method. Prestack AVO inversion on logging data and real seismic data demonstrate that the proposed method can improve the accuracy of the inversion, and it is more immune to strong noise.

A Distributed Principal Component Analysis Compression for Smart Seismic Acquisition Networks

This paper develops a new framework for data compression in seismic sensor networks by using the distributed principal component analysis (DPCA). The proposed DPCA scheme compresses all seismic traces in the network at the sensor level. First of all, the statistics of the seismic traces acquired at all sensors are represented by a mixture model of a number of probability density functions. Based on this mixture model, the DPCA finds the global PCs at the fusion center. These PCs are then sent back to all sensors so that each sensor projects its own traces over these PCs. This scheme does not require transmitting the original traces, here, leading to a low computational load and a high compression ratio, compared with compression obtained using the local PC analysis (LPCA). Furthermore, we develop an efficient communication solution for the DPCA implementation on practical sensor networks. Finally, the proposed scheme is evaluated using real and synthetic seismic data showing improved performance over the LPCA and the traditional 2-D discrete cosine transform (DCT-2-D) compression. Specifically, to preserve a given signal energy during the compression, the DPCA is shown to achieve a higher compression ratio than the LPCA and the DCT-2-D.

An inverseQfiltering approach based on a varying stabilization factor

The stable approach of inverse Q filtering based on wavefield continuation theory can simultaneously perform amplitude compensation and phase correction. The instability of conventional full inverse Q filtering is overcome through adding a stabilization factor to the wavefield continuation, and a good application effect is achieved. However, the method is likely to lead to noise amplification while compensating for the effective signal of the seismic data, resulting in the decrease of the signal-to-noise ratio (SNR) of seismic data. This phenomenon is due to the constant stabilization factor being added to the inverse Q filtering method. This paper presents an inverse Q filtering method based on a varying stabilization factor, which is a function of the propagation distance of a seismic wave, frequency of seismic data, and quality factor Q. This method can avoid the amplification of noise energy which can be caused by using the constant stabilization factor. The results of the synthetic and real seismic data show that compared with the compensation effect of the conventional inverse Q filtering, this method can avoid the amplification of the shallow and middle noise energy of seismic data, and effectively improves the SNR and resolution of the seismic data.

Ground-roll attenuation using curvelet downscaling

We have developed a ground-roll attenuation strategy for seismic records that adopts the curvelet transform. The curvelet transform decomposes the seismic events based on their dip and frequency content information. The curvelet panels that contain only either reflection or ground-roll energy can be used to alter the curvelet panels with mixed reflection and ground-roll energies. We build a curvelet-domain mask function from the ground-roll-free curvelet coefficients (high frequencies) and downscale it to the ground-roll-contaminated curvelet coefficients (low frequencies). The mask function is used inside a least-squares optimization scheme to preserve the seismic reflections and attenuate the ground roll. Synthetic and real seismic data examples show the application of the proposed ground-roll attenuation method.

Stochastic nonlinear inversion of seismic data for the estimation of petroelastic properties using the ensemble smoother and data reparameterization

We have developed a new stochastic nonlinear inversion method for seismic reservoir characterization studies to jointly estimate elastic and petrophysical properties and to quantify their uncertainty. Our method aims to estimate multiple reservoir realizations of the entire set of reservoir properties, including seismic velocities, density, porosity, mineralogy, and saturation, by iteratively updating the initial ensemble of models based on the mismatch between their seismic response and the measured seismic data. The initial models are generated using geostatistical methods and the geophysical forward operators include rock-physics relations and a seismic forward model. The optimization is achieved using an iterative ensemble-based algorithm, namely, the ensemble smoother with multiple data assimilation, in which each iteration is based on a Bayesian updating step. The advantages of the proposed method are that it can be applied to nonlinear inverse problems and it can provide an ensemble of solutions from which we can quantify the uncertainty of the model properties of interest. To reduce the computational cost of the inversion, we perform the optimization in a lower dimensional data space reparameterized by singular value decomposition. The proposed methodology is validated on a synthetic case in which the set of petroelastic properties is recovered with satisfactory accuracy. Then, we applied the inversion method to a real seismic data set from the Norne field in the Norwegian Sea.

A seismic fault recognition method based on ant colony optimization

Fault recognition is an important section in seismic interpretation and there are many methods for this technology, but no one can recognize fault exactly enough. For this problem, we proposed a new fault recognition method based on ant colony optimization which can locate fault precisely and extract fault from the seismic section. Firstly, seismic horizons are extracted by the connected component labeling algorithm; secondly, the fault location are decided according to the horizontal endpoints of each horizon; thirdly, the whole seismic section is divided into several rectangular blocks and the top and bottom endpoints of each rectangular block are considered as the nest and food respectively for the ant colony optimization algorithm. Besides that, the positive section is taken as an actual three dimensional terrain by using the seismic amplitude as a height. After that, the optimal route from nest to food calculated by the ant colony in each block is judged as a fault. Finally, extensive comparative tests were performed on the real seismic data. Availability and advancement of the proposed method were validated by the experimental results.

Data Field-Based K-Means Clustering for Spatio-Temporal Seismicity Analysis and Hazard Assessment

Microseismic sensing taking advantage of sensors can remotely monitor seismic activities and evaluate seismic hazard. Compared with experts’ seismic event clusters, clustering algorithms are more objective, and they can handle many seismic events. Many methods have been proposed for seismic event clustering and the K-means clustering technique has become the most famous one. However, K-means can be affected by noise events (large location error events) and initial cluster centers. In this paper, a data field-based K-means clustering methodology is proposed for seismicity analysis. The application of synthetic data and real seismic data have shown its effectiveness in removing noise events as well as finding good initial cluster centers. Furthermore, we introduced the time parameter into the K-means clustering process and applied it to seismic events obtained from the Chinese Yongshaba mine. The results show that the time-event location distance and data field-based K-means clustering can divide seismic events by both space and time, which provides a new insight for seismicity analysis compared with event location distance and data field-based K-means clustering. The Krzanowski-Lai (KL) index obtains a maximum value when the number of clusters is five: the energy index (EI) shows that clusters C1, C3 and C5 have very critical periods. In conclusion, the time-event location distance, and the data field-based K-means clustering can provide an effective methodology for seismicity analysis and hazard assessment. In addition, further study can be done by considering time-event location-magnitude distances.

Nonstationary differential resolution: An algorithm to improve seismic resolution

High-resolution seismic data enables better well ties, structure delineation, stratigraphic mapping, and reservoir characterization. Differential resolution (DR) is a data-driven method to improve seismic resolution, but it can introduce a false spatial amplitude variation in the seismic output due to whole trace normalization. The newly proposed normalization technique decomposes the input seismic trace using a translating Gaussian window and then implements the DR algorithm on each window. The introduced weight factors give the interpreter control of the degree of spectral broadening. These developments enable the algorithm to account for the nonstationary properties of the seismic trace, reduce spurious spatial amplitude variation, and provide broader bandwidth seismic data for detailed analysis. We described the mathematical derivation of the nonstationary differential resolution (NSDR) algorithm and its implementation on synthetic and real seismic data. A comparison of NSDR with the original and DR shows better relative amplitude preservation.

Processing of large offset data: experimental seismic line from Tenerife Field, Colombia

ABSTRACT: Exploration seismology provides the main source of information about the Earth’s subsurface, which in many cases can be presented as a simple model of horizontal or near-horizontal layers. After the seismic acquisition step, conventional seismic processing of reflection data provides an image of the subsurface by using information about the reflections of these layers. The traveltime from a source to different receivers is adjusted using a hyperbolic function. This expression is used in the case involving an isotropic medium, which is a simplification of nature, whereas geologically complex media are generally anisotropic. A subsurface model that more closely resembles reality is the vertical transverse isotropy, which defines two parameters that are required to correct the traveltimes: the NMO velocity and the anellipticity parameter. In this paper, we reviewed the literature and methodology for velocity analysis of seismic data acquired from anisotropic media. A model with horizontal layers and anisotropic behavior was developed and evaluated. The anisotropic velocity was compared to the isotropic velocity, and the results were analyzed. Finally, the methodology was applied to real seismic data, i.e. an experimental landline from Tenerife Field, Colombia. The results show the importance of the anellipticity parameter in models with anisotropic layers.

From field analogues to realistic seismic modelling: a case study of an oil-producing andesitic sill complex in the Neuquén Basin, Argentina

Interpretation of seismic data has played a major role in recent advances in the studies of igneous sill complexes. Seismic modelling studies based on field analogues represent a promising tool to close the scale gap between observations from outcrops and seismic data and support seismic interpretation. Virtual outcrop models are commonly used to include high-resolution geological structures in models of seismic-scale field analogues. However, realistic seismic modelling requires not only detailed structural input, but also well-constrained elastic properties and an adequate seismic modelling technique. Here, we present a seismic modelling study of oil-producing andesitic sills in the Neuquén Basin, Argentina, which implements all modelling elements at high accuracy by combining virtual outcrop models, well data and a 2(3)D filtering method. Our results indicate that the modelled seismic signatures of intrusive bodies observed in field analogues are characterized by frequency-dependent interference and strong amplitude variations owing to highly variable elastic properties of both host rock and sills. We demonstrate that detailed waveform patterns observed in real seismic data can be linked to intrusive bodies below the traditionally assumed limit of resolution via realistic seismic modelling. This illustrates how an integrated modelling approach based on field analogues can aid seismic interpretation.

Well log and seismic data analysis for complex pore-structure carbonate reservoir using 3D rock physics templates

The pore structure in heterogeneous carbonate rock is usually very complex. This complex pore system makes the relationship between the velocity and porosity of the rock highly scattered, so that for the classical two-dimensional rock physics template (2D RPT) it is not enough to accurately describe the quantitative relationship between the rock elastic parameters of this kind of reservoir and its porosity and water saturation. Therefore it is possible to attribute the effect of pore type to that of the porosity or water saturation, and leads to great deviations when applying such a 2D RPT to predict the porosity and water saturation in seismic reservoir prediction and hydrocarbon detection. This paper first presents a method to establish a new three-dimensional rock physics template (3D RPT) by integrating the Gassmann equations and the porous rock physics model, and use it to characterize the quantitative relation between rock elastic properties and the reservoir parameters including the pore aspect ratio, porosity and water saturation, and to predict these parameters from the known elastic properties. The test results on the real logging and seismic inversion data show that the 3D RPT can accurately describe the variations of elastic properties with the porosity, water saturation and pore-structure parameters, and effectively improve the accuracy of reservoir parameters prediction.

The design of L1-norm visco-acoustic wavefield extrapolators

Explicit depth frequency–space (f − x) prestack imaging is an attractive mechanism for seismic imaging. To date, the main focus of this method was data migration assuming an acoustic medium, but until now very little work assumed visco-acoustic media. Real seismic data usually suffer from attenuation and dispersion effects. To compensate for attenuation in a visco-acoustic medium, new operators are required. We propose using the L1-norm minimization technique to design visco-acoustic f − x extrapolators. To show the accuracy and compensation of the operators, prestack depth migration is performed on the challenging Marmousi model for both acoustic and visco-acoustic datasets. The final migrated images show that the proposed L1-norm extrapolation results in practically stable and improved resolution of the images.

Synthetic seismic model of a Permian biosiliceous carbonate – carbonate depositional system (Spitsbergen, Svalbard Archipelago)

Abstract The non-unique correspondence between seismic and subsurface geology is a fundamental problem when interpreting seismic data sets. Synthetic seismic models of outcrop analogs are commonly constructed to cover the gap between the small-scale outcrop observations and low-resolution seismic data. The Permian biosiliceous carbonate – carbonate sediments on Spitsbergen are characterized by a wide variability in lithologies and microfacies determining the petrophysical properties (e.g., porosity, acoustic properties) that consequently complicate seismic interpretation. This study uses 1D and 2D synthetic seismic modelling techniques (at different resolutions) to gain an understanding of how seismic reflectors are expressed with respect to the sediment distribution, aiming to facilitate real seismic interpretation. In the study area, nine microfacies were defined that were used to produce a geological model displaying small-scale microfacies variations within a well-defined sequence stratigraphic framework. Laboratory derived petrophysical properties (Vp, ρBulk and AI) were assigned to each predefined microfacies body in the geological models in order to construct acoustic impedance models that were used to produce synthetic seismograms. The appearance of the seismic reflectors in the synthetic seismic profiles is primarily controlled by changes in mineral composition and link to spatial microfacies distributions within the depositional sequences. Differences in acoustic impedance and the origin of synthetic seismic reflections within a single microfacies type are mainly caused by porosity contrasts (varying between 5 and 20%) and diagenetic modifications such as chertification and cementation. This detailed information cannot be derived from the low-frequency seismograms (25–100 Hz) resulting in changes in seismic expression when seismic resolution diminishes. Comparison with time equivalent, real and synthetic seismic data of the Finnmark Platform reveals similarities with the synthetic seismic reflection patterns of Spitsbergen. In both areas, the pronounced seismic traces follow abrupt microfacies transitions, which are coherent with cycle boundaries and timelines.

Feasibility Study of SQp and SQs Attributes Application for Facies Classification

Formation evaluation is a critical requirement in oil and gas exploration and development projects. Although it may be costly, wireline logs need to be acquired to evaluate and understand the subsurface formation. Gamma ray and resistivity are the two main well-log data used for formation evaluation purposes. However, outside the well, formation evaluation becomes difficult, as these logs are not available. Hence, it is important to have other data equivalent to the gamma ray or resistivity logs, which can be derived from other technique, such as seismic data. As a consequence, the dependency on well-log data can be avoided. Thus, the complexity in formation evaluation outside the well, such as the determination of facies, lithology, and fluid content, as well as petrophysical properties can be solved accurately even without well-log data. The objective of this paper was to demonstrate an application of the SQp and SQs attributes for facies classification. These attributes were derived from attenuation attributes through rock physics approximation by using basic elastic properties: P-wave, S-wave, and density. A series of tests were carried out to show the applicability of these attributes on well-logs and real seismic data from offshore the Malaysia Peninsular. Simultaneous inversion was used in the data sets to produce the three-dimensional (3D) SQp and SQs attributes required as inputs of a neural network engine in defining the facies distribution. The results showed that the SQp attribute was very similar to the gamma ray, while the SQs attribute was similar to the resistivity responses even in different reservoir conditions, including low resistivity low contrast and coal masking environment. In conclusion, the SQp motif, which is similar to the gamma ray motif, can potentially be used for facies classification/identification. Together with the SQs attribute, the SQp attribute can be used as input for the facies classification workflow. The application of the SQp and SQs attributes successfully identified the gas sand distribution and separated it clearly from the brine distribution in an offshore Malaysian field.

Multiscale Fractures Characterization Based on Ant Colony Optimization and Two-Dimensional Variational Mode Decomposition

Hydrocarbon exploration in fractured reservoirs requires accurate estimates of fracture characteristics such as fracture orientations and fracture development degrees. The coherence method, which measures the degree of similarity between the multichannel seismic data, can image fractures by detecting variation anomalies in the seismic signals. However, it is weak in precisely positioning fractures and is sensitive to noise. We use an improved ant colony optimization (ACO) algorithm that adopts both coherence data and edge detection data as heuristic factor to refine the coherence data. As a distinctive time–frequency analysis method, variational mode decomposition (VMD) is a flexible method with excellent time–frequency resolution. We propose a measure factor combining mutual information and energy entropy to judge how many intrinsic mode functions to be decomposed are most suitable. Since fracture characterization is similar to image analysis, we propose a novel method combining ACO-coherence with a two-dimensional VMD (2D-VMD) to meticulously depict fractures at different scales and different directions. First, the seismic coherence data are obtained via a dip scanning coherence algorithm. Then, the coherence data are refined by the improved ACO algorithm. We extract the slice along the target stratum from the ACO-coherence data, and the slice is decomposed into different frequency band images using the 2D-VMD method. After applying the method to real seismic data from a typical fractured reservoir, it is clear that the method we propose is capable of revealing fractures at different scales and different directions with high precision.

A new Fourier azimuthal amplitude variation fracture characterization method: Case study in the Haynesville Shale

We have evaluated a novel fracture characterization technique using azimuthal amplitude variations (AVAz) present in 3D seismic data, and we implemented it using synthetic and real seismic data targeting the Haynesville Shale. The method we evaluated overcomes many common AVAz limitations and differs from standard AVAz approaches in the following ways: (1) It was explicitly designed to model vertically fractured transverse isotropic (VFTI) media; (2) it can correctly resolve the fracture strike azimuth without a 90° ambiguity and uses a new magnitude-based method that is invariant to the sign of seismic reflectivity [Formula: see text]; and (3) it incorporates advanced inversion techniques to estimate a novel fracture density proxy that responds linearly to crack density. Our method is based on a newly derived relationship that relates seismic reflectivity directly to rock/fracture properties in VFTI media. We validated our method through rigorous testing on more than 400 synthetic seismic data sets. These synthetic tests indicate that our method excels at estimating fracture azimuth and fracture density from surface seismic data with overall success rates around 80%–85% for noisy data and 90%–95% for noise-free data. Applying our method to field data from the Haynesville Shale indicates that the dominant fracture set is oriented at approximately [Formula: see text] relative to geodetic north, i.e., rotated slightly counterclockwise of east–west. We assume a constant azimuth of 80° throughout our relatively small 20 square miles study area, and our method clearly identifies a general area with unusually high fracture density as well as several smaller subzones of dense fracturing. These smaller features appear to be connected by a pervasive large-scale fracture network covering the area with dominant features aligned at roughly parallel and perpendicular to our calculated fracture azimuth. Although we could not directly confirm these fracture characteristics, our results largely agree with previously published information about fracturing in our study area.

Seismic Attribute Analysis with Saliency Detection in Fractional Fourier Transform Domain

Most image saliency detection models are dependent on prior knowledge and demand high computational cost. However, spectral residual (SR) and phase spectrum of the Fourier transform (PFT) models are simple and fast saliency detection approaches based on two-dimensional Fourier transform without the prior knowledge. For seismic data, the geological structure of the underground rock formation changes more obviously in the time direction. Therefore, one-dimensional Fourier transform is more suitable for seismic saliency detection. Fractional Fourier transform (FrFT) as an improved algorithm for Fourier transform, we propose the seismic SR and PFT models in one-dimensional FrFT domain to obtain more detailed saliency maps. These two models use the amplitude and phase information in FrFT domain to construct the corresponding saliency maps in spatial domain. By means of these two models, several saliency maps at different fractional orders can be obtained for seismic attribute analysis. These saliency maps can characterize the detailed features and highlight the object areas, which is more conducive to determine the location of reservoirs. The performance of the proposed method is assessed on both simulated and real seismic data. The results indicate that our method is effective and convenient for seismic attribute extraction with good noise immunity.

Inversion-based data-driven time-space domain random noise attenuation method

Conventional time-space domain and frequency-space domain prediction filtering methods assume that seismic data consists of two parts, signal and random noise. That is, the so-called additive noise model. However, when estimating random noise, it is assumed that random noise can be predicted from the seismic data by convolving with a prediction error filter. That is, the source-noise model. Model inconsistencies, before and after denoising, compromise the noise attenuation and signal-preservation performances of prediction filtering methods. Therefore, this study presents an inversion-based time-space domain random noise attenuation method to overcome the model inconsistencies. In this method, a prediction error filter (PEF), is first estimated from seismic data; the filter characterizes the predictability of the seismic data and adaptively describes the seismic data’s space structure. After calculating PEF, it can be applied as a regularized constraint in the inversion process for seismic signal from noisy data. Unlike conventional random noise attenuation methods, the proposed method solves a seismic data inversion problem using regularization constraint; this overcomes the model inconsistency of the prediction filtering method. The proposed method was tested on both synthetic and real seismic data, and results from the prediction filtering method and the proposed method are compared. The testing demonstrated that the proposed method suppresses noise effectively and provides better signal-preservation performance.

Time-domain wavefield reconstruction inversion

Wavefield reconstruction inversion (WRI) is an improved full waveform inversion theory that has been proposed in recent years. WRI method expands the searching space by introducing the wave equation into the objective function and reconstructing the wavefield to update model parameters, thereby improving the computing efficiency and mitigating the influence of the local minimum. However, frequency-domain WRI is difficult to apply to real seismic data because of the high computational memory demand and requirement of time-frequency transformation with additional computational costs. In this paper, wavefield reconstruction inversion theory is extended into the time domain, the augmented wave equation of WRI is derived in the time domain, and the model gradient is modified according to the numerical test with anomalies. The examples of synthetic data illustrate the accuracy of time-domain WRI and the low dependency of WRI on low-frequency information.