Seismic Amplitude Data Research Articles

A High-Resolution subtle fault mapping in Marlin oilfield, Brazil

Discriminating geological discontinuities is important for reservoir management because they influence storage capacity and fluid flow. Structural heterogeneities, such as faults and fractures, significantly impact the compartmentalization of the reservoir’s facies and affect the flow of hydrocarbons. These structures can act as the pore space for fluid storage, flow-control conduits, or barriers that allow the accumulation of exploitable volumes of oil and gas. Understanding these structures’ spatial and temporal development can significantly impact hydrocarbon exploration. Reservoir characterization is a critical process in the oil and gas industry that aims to comprehensively understand reservoir rocks and their fluid content and distribution. It is a multidisciplinary approach that combines geological, geophysical, and engineering datasets through statistical and deterministic mathematical models. Seismic datasets play a crucial role in reservoir characterization. Interpreters can use seismic data processing and imaging to interpret geological horizons and map faults, building complex geological models. Seismic inversion is also employed to estimate acoustic impedance and other petrophysical properties. Faults and fractures mapping are highly relevant themes in reservoir characterization. Identifying subtle faults and fractures is essential in specifying hydrocarbon migration paths and identifying bypassed oil accumulations. However, traditional seismic amplitude data interpretation should consider these structures more. In this study, we apply a high-resolution fault mapping workflow to map the subtle faults system at the Marlim field in the offshore portion of the Campos Basin. As an outcome, we imaged two separate fault trends inside and outside the Marlim reservoir.

A method for intelligent identification of faults in seismic using an attention-based ES-UNet network with model re-training learning

Accurate fault identification provides an important basis for well location deployment, oil and gas resource development. However, obtaining a large number of fault samples through manual labeling or interpretation is difficult. Conventional convolutional neural networks have certain challenges in finely characterizing low-order faults and fault continuity. This paper studies a 3D intelligent fault identification method that integrates deep learning with attention mechanism. Based on 3D-UNet network, the method constructs a dual attention fault identification model with an encoder-decoder structure, which focuses on enhancing low-order fault information and extracting key features. An adaptive hybrid loss function is proposed to address the imbalance problem of fault proportions in seismic amplitude data. The function introduces an adjustment coefficient to adaptively enhance the attention to high-order or low-order faults according to the actual situation of the fault, improving the accuracy and continuity of fault identification. Furthermore, a model re-training learning optimization method is presented to capture the missed fault information in the first training and repair the connectivity of low-order faults. Applying this method to the synthetic data set, the accuracy can be improved to 0.9732, and the loss can be converged to 0.0982. The results of shale reservoirs in the North Sea Basin and Kerry-3D indicate that the method can reduce the situation of fault missing and misidentification, effectively improve fault identification accuracy, and enhance low-order fault identification by obtaining more contextual fault features.

Using 3D Seismic Technology to Accurately Identify Faults and Reservoir Characteristics

With the accelerating development of the Chinese economy, the demand for oil and gas energy is increasing day by day. The accuracy and efficiency of fault interpretation are crucial for the exploration and development of oil and gas reservoirs. The presence of oil and gas in reservoirs will inevitably cause changes in the geophysical response characteristics, and different types of reservoirs have different response characteristics in logging curves such as sound, discharge, and electricity. 3D seismic technology is currently one of the most effective techniques for obtaining fault structural features and identifying small faults, but conventional seismic data processing techniques are unable to meet the accuracy requirements of current seismic exploration. With the development of computer technology, more and more scholars are applying deep learning (DL) algorithms to the identification of faults and reservoirs to further improve the recognition effect. This article is based on the DL algorithm and utilizes 3D seismic technology to design a method for automatic and accurate recognition of fault and reservoir features. Obtain labeled seismic amplitude faults, reservoir data, and seismic amplitude data to be identified from the 3D seismic amplitude data volume, and construct a CNN training model to identify the fault and reservoir data. The experimental results show that the method designed in this article can improve the efficiency and accuracy of fault and reservoir identification.

Source Rock Evaluation from Rock to Seismic Data: An Integrated Machine-Learning-Based Work Flow and Application in the Brazilian Presalt (Santos Basin)

The capacity to predict the occurrence and quality of source rocks in a sedimentary basin is of great economic importance in the evaluation of conventional and non-conventional petroleum resources. Direct laboratory examinations of rock samples are the most accurate way to obtain their geochemical properties. However, rock information is usually sparse, and source rocks are often sampled at positions that may not be representative of the average organic content and quality of oil kitchens. This work proposes a work flow supported by machine learning methods (random forest, DBSCAN, and NGBoost) to automate the source rock characterization process to maximize the use of available data, expand data information, and reduce data analysis time. From the automated quality control of the input data through the extrapolation of laboratory measurements to continuous well logs of geochemical properties, culminating in the 3D estimation of these properties, we generate volumes of total organic carbon (TOC) by applying machine learning techniques. The proposed method provides more accurate predictions, reducing uncertainties in the characterization of source rocks and assisting in exploratory decision making. This methodology was applied in the presalt source rocks from Santos Basin (Brazil) and allowed us to quantify the TOC distribution, improving the interpretation of the main source rock interval top and base based only on seismic amplitude data. The result suggests higher TOC values in the northern and western grabens of the studied area and a higher charge risk in the eastern area.

Application of continuous porosity-dependent decomposed density and velocity geophysical modelling for 4-D Seismic-based Quantification of hydrocarbon-bearing upper-Cretaceous stratigraphic systems, Southern Onshore, Pakistan

4D seismic was frequently used for assessing the migration of hydrocarbon-bearing fluids within the fluvial sedimentary systems. The current study focused on enhancing a detailed stratigraphic reservoir characterization of a transgressive system tract (TST) within a gas field in Pakistan. This system is dominated by ultra-thin-bedded mixed shale-sand layers with low-impedance contrast, which adversely attenuate the true amplitudes of seismic data. The continuous wavelet transforms (CWT)-based quantitative porosity-dependent decomposed density and velocity modelling (CPDV) is applied to predict the thickness and porosity based on a 4D seismic approach. A combination of 2D and 3D seismic surveys (A & B) were used in this research work. A 16.85 m thick fluvial point bar (PB) was resolved by 35 Hz CWT amplitude. The 34 Hz CWT amplitude revealed an 11.85 m thick shale-dominated channel sand-fills (CS). The CPDV assigned a porosity range of 21 to 27% for PBs. Meanwhile, the band-limited seismic attributes remained poor in predicting the actual stratigraphic prospect. The CPDV has predicted the exact orientation of the coarse-grained sandstone lens of PB, which pinch out to the east. The zone of erosion is predicted where the porosity decreases from 17 to 8%, the density increases from 2.3 to 2.5 g/c.c, and the velocity increases from 3500 to 4300 m/s. The depositional zone have experienced a velocity range of ∼2200 to 3000 m/s, density range of 2 to 2.2 g/c.c, and porosity range of 17 to 27 %, which can serve to ascertain the depleting stratigraphic plays for Worldwide Basins.

Seismic characterization of fault and fractures in deep buried carbonate reservoirs using CNN-LSTM based deep neural networks

Geologic structures such as faults and fractures play a vital role in reservoir development studies. Fault and fracture identification requires high-quality seismic amplitude data with good resolution and a high signal-to-noise ratio (SNR). An anticline fault-controlled carbonate structure in North China contains significant reserves of hydrocarbons. However, the heterogeneity of deeply buried carbonates (burial depth > 4000 m) results in poor resolution and low SNR of seismic amplitude data, making it challenging to distinguish fault edge characteristics under the buried structures. In this study, we apply an inverse Q-filtering method based on the Gabor transform and structure-oriented filtering to compensate for the weak signal of the seismic events and enhance the SNR under the deep buried carbonate rocks. We investigated fault and adjacent fracture zones in structurally complex deeply buried carbonate reservoirs by extracting seismic fracture attributes (fault likelihood, eigenvalue coherence, and curvature) and feedforward neural network spatial fracture distribution from high-resolution seismic amplitude data. Moreover, a fracture density prediction model based on seismic attributes, CNN (convolutional neural network), and LSTM (long short-term memory) is developed to predict spatial fracture density. The predicted fracture density results are verified with seismic fracture attributes, geological modeling, and FMI (formation micro-imager) logs. This integrated workflow can assist in identifying high-production zones in structurally complex and deeply buried carbonates reservoirs.

Fault2SeisGAN: A method for the expansion of fault datasets based on generative adversarial networks

The development of supervised deep learning technology in seismology and related fields has been restricted due to the lack of training sets. A large amount of unlabeled data is recorded in seismic exploration, and their application to network training is difficult, e.g., fault identification. To solve this problem, herein, we propose an end-to-end training data set generative adversarial network Fault2SeisGAN. This network can expand limited labeled datasets to improve the performance of other neural networks. In the proposed method, the Seis-Loss is used to constrain horizon and amplitude information, Fault-Loss is used to constrain fault location information, and the Wasserstein distance is added to stabilize the network training to generate seismic amplitude data with fault location labels. A new fault identification network model was trained with a combination of expansion and original data, and the model was tested using actual seismic data. The results show that the use of the expanded dataset generated in this study improves the performance of the deep neural network with respect to seismic data prediction. Our method solves the shortage of training data set problem caused by the application of deep learning technology in seismology to a certain extent, improves the performance of neural networks, and promotes the development of deep learning technology in seismology.

Neural net detection of seismic features related to gas hydrates and free gas accumulations on the northern U.S. Atlantic margin

Bottom-simulating reflections (BSRs) that sometimes mark the base of the gas hydrate stability zone in marine sediments are often identified based on the reverse polarity reflections that cut across stratigraphic layering in seismic amplitude data. On the northern U.S. Atlantic margin (USAM) between Cape Hatteras and Hudson Canyon, legacy seismic data have revealed pronounced BSRs south of the deepwater extension of Hudson Canyon and more subtle ones from offshore Delaware south to Cape Hatteras, where the reflections sometimes follow stratigraphic layering. Using high-resolution seismic data acquired during the 2018 Mid-Atlantic Resource Imaging Experiment and a supervised neural net, we identify seismic features associated with gas hydrates and/or the top of gas between Hudson Canyon and Cape Hatteras. Using seismic attributes especially sensitive to the presence of gas, we train a neural network algorithm on seismic data from an area with strong BSRs and then apply the model to the rest of the data set. The results indicate that gas hydrate and/or shallow free gas are significantly more widespread on the northern part of the USAM than previously known. Seismic indicators of gas extend landward from the 2000 m isobath to the upper continental slope in sectors with (offshore Virginia) and, to a lesser extent, without (offshore New Jersey) pervasive upper slope methane seeps. Higher sand content and intermediate sediment thickness, factors related to the container size and gas charge in a petroleum systems framework, are associated with more robust gas indicators.

Utilization of seismic attributes for structural pattern detection In Bualawn, Dor Mansour fields, Western Sirt Basin, Libya

An integrated approach to the study of fault patterns carried out in the complex geological structures of Bualawn, Dor Mansour fields by using multiple seismic attributes of 3-D seismic data. Each type of geological structure event usually generates a unique seismic signature that can be recognized and identified. This paper highlights the practical importance of analyze integrated attributes and interpret them within the context of an appropriate structural deformation. Many of discontinuity attributes such as variance, 3DEE (3D edge enhancement), ant tracking, chaos and structure smoothing have been selected to delineate fault styles and their displacement effectively, which cannot be fully delineated using seismic amplitude data. Conjugate growth faults in NW-SE and NE-SW direction, dipping toward SW with antithetic faults dipping in opposite direction to the growth faults and other minor faults. The faults were identified manually and then tested by discontinuity attributes. Fault system is represented by a fracture system comprising long en echelon seismic faults and several discontinuities. This system of en echelon faults was initiated by left-lateral wrenching movements. This wrenching system has been identified by negative and positive flower structure along the releasing and restraining bends. The current understanding that these faults are strike-slip faults despite the absence of extensive horizontal displacements along them as shown on different time slices. A periodic strike-slip movement along deep-seated basement faults have developed many structural features, such as horst-graben styles, which is the orientation of the transtensional movements linked to the Upper Cretaceous Succession.

AVO study on a known geothermal reservoir located in the fractured carbonate formations of the pre-Cenozoic basement, Northwest Hungary

Amplitude Versus Offset (AVO) analysis has been successfully applied in the hydrocarbon industry for more than three decades. This effective tool of the joint seismic and well-log data processing enables to predict and analyze fluid saturated porous geological formations. AVO methodology is based on the anomalous behavior of the pre-stack reflected amplitudes observed from fluid bearing rocks. However, the potential of AVO methodology is still unexploited in geothermal exploration, although the lithology and rock physical properties are very similar. In this study, we summarize the theoretical backgrounds and calculate synthetic AVO responses of a known geothermal reservoir located in the fractured carbonates of the Mesozoic basement. We demonstrate that the AVO response of a deep geothermal reservoir can be quite different from the amplitude response observed from a hydrocarbon bearing clastic formation. AVO attributes of the investigated geothermal reservoir are presented, and the potential of its detection by seismic amplitude data are discussed.

Detection of Natural Tilted Fractures From Azimuthal Seismic Amplitude Data Based on Linear-Slip Theory

Tilted transverse isotropy (TTI) is common for naturally fractured reservoir when the fractures exist a tilted axis of symmetry. Seismic characterization of natural tilted fractures from observable azimuthal reflected amplitude data, however, is quite difficult. We focus on the detection of natural tilted fractures from azimuthal seismic amplitude data, especially for low- and high-angle tilted fractures. Using the linear-slip theory, we first express the effective elastic stiffness matrix of a TTI medium as a function of background elastic moduli, fracture density, and three angles, including the incident angle, the azimuth angle, and the tilt angle. Based on the scattering function and the first-order perturbations in stiffness components across a weak-contrast reflection interface separating two weak-anisotropy TTI media, we then derive a weak-anisotropy and linearized PP-wave reflection coefficient containing the tilt angle and fracture density. For low- and high-angle aligned fracture sets, we formulate the low- and high-angle approximate PP-wave reflection coefficient equations, respectively, and propose a Bayesian seismic inversion approach to estimate the tilt angle and the corresponding fracture densities using the azimuthal differences in seismic reflected amplitude data. Synthetic and real datasets are used to demonstrate the feasibility and reliability of the proposed inversion approach. Test results make us believe that our proposed inversion approach allow us to obtain the tilted fracture properties of hydrocarbon reservoirs in a more accurate manner than previous cases of vertically transverse isotropy (VTI) or horizontally transverse isotropy (HTI).

Seismic Data Consecutively Missing Trace Interpolation Based on Multistage Neural Network Training Process

Due to the constraints of natural environments, acquired prestack seismic data is usually not complete, which seriously affects subsequent seismic data processing. With the progress of deep learning, many neural networks with different structures have been applied to missing seismic data interpolation. Among them, U-net can efficiently interpolate the regularly and irregularly missing seismic traces with small gap. While for consecutively missing seismic traces with big gap, the interpolation results for low amplitude missing components need to be further improved. In this letter, we analyze the variation of interpolation results for consecutively missing seismic traces during the traditional U-net training process, and find that U-net tends to only interpolate the high amplitude missing components. Meanwhile, due to the distribution difference between low and high amplitude seismic data, one U-net model is insufficient to interpolate both high and low amplitude missing components with a wide amplitude range. To improve the interpolation results of single U-net, we propose a multistage training process to train multiple U-net models. Each U-net model focuses on interpolating different missing components with a small amplitude range. In this way, more accurate interpolation results for low amplitude missing components can be obtained. Comparison experiments conducted on synthetic and field seismic data show that, under the same number of training epochs, the proposed training process can produce more accurate interpolation results comparing with traditional single U-net.

Inflating Shallow Plumbing System of Bezymianny Volcano, Kamchatka, Studied by InSAR and Seismicity Data Prior to the 20 December 2017 Eruption

Explosive eruptions at steep-sided volcanoes may develop with complex precursor activity occurring in a poorly-understood magma plumbing system so that timelines and possible interactions with the geologic surrounding are often unresolved. Here we investigate the episode prior to the energetic December 20, 2017 eruption at Bezymianny volcano, Kamchatka. We compare degassing activity inferred from time-lapse camera images, seismicity and real-time seismic amplitude (RSAM) data derived from a temporary station network, as well as high-resolution InSAR displacement maps. Results show that the first changes can be identified in low-frequency seismicity and degassing at least 90 days before the eruption, while the first volcano-tectonic (VT) seismicity occurred 50 days before the eruption. Coinciding with significant changes of the RSAM, surface displacements affect the volcanic flanks at least 9 days prior to the eruption. Inversion modeling of the pre-eruptive surface deformation as well as deflation-type, co-eruptive surface changes indicate the presence of a shallow and transient reservoir. We develop a conceptual model for Bezymianny volcano initiating with deep seismicity, followed by shallow events, rockfalls, steaming and an inflating reservoir. The eruption is then associated with subsidence, caused by deflation of the same reservoir. This sequence and conceivable causality of these observations are providing a valuable contribution to our understanding of the shallow magma plumbing system beneath Bezymianny and may have relevance for volcano monitoring and early warning strategies at similar volcanoes elsewhere.

Which seismic attributes are best for subtle fault detection?

Subtle fault detection plays a vital role in reservoir development studies because faults may form baffles or conduits that significantly control how a petroleum reservoir is swept. Small-throw faults are often overlooked in interpreting seismic amplitude data. However, seismic attributes can aid in mapping small faults. Over the years, dozens of seismic attributes have been developed that offer additional features for interpreters with associated caveats. Using the Maui 3D seismic data acquired in the Offshore Taranaki Basin, New Zealand, we have generated seismic attributes that are typically useful for fault detection. We find that multiattribute analysis provides greater geologic information than would be obtained by the analysis of individual attribute volumes. We extract the geologic content of multiple attributes in two ways: interactive corendering of different seismic attributes and the unsupervised machine learning algorithm self-organizing maps (SOM). Corendering seismic attributes that are mathematically independent but geologically interrelated provides a well-integrated structural image. We suggest eight combinations of 16 various attributes useful for a human interpreter with interest in fault and fracture detection. Current interpretation display capabilities constrain corendering to only four attribute volumes. Therefore, we use principal component analysis and SOM techniques to efficiently integrate the geologic information contained within many attributes. This approach gathers the data into one classification volume based on the interrelationships between seismic attributes. We show that our resulting SOM classification volume better highlights small faults that are difficult to image using conventional seismic interpretation techniques. We find that SOM works best when a fault exhibits anomalous features for multiple attributes within the same voxel. However, human interpreters are more adept at recognizing spatial patterns within various attributes and can place them in an appropriate geologic context.

Unsupervised machine learning using 3D seismic data applied to reservoir evaluation and rock type identification

Net reservoir discrimination and rock type identification play vital roles in determining reservoir quality, distribution, and identification of stratigraphic baffles for optimizing drilling plans and economic petroleum recovery. Although it is challenging to discriminate small changes in reservoir properties or identify thin stratigraphic barriers below seismic resolution from conventional seismic amplitude data, we have found that seismic attributes aid in defining the reservoir architecture, properties, and stratigraphic baffles. However, analyzing numerous individual attributes is a time-consuming process and may have limitations for revealing small petrophysical changes within a reservoir. Using the Maui 3D seismic data acquired in offshore Taranaki Basin, New Zealand, we generate typical instantaneous and spectral decomposition seismic attributes that are sensitive to lithologic variations and changes in reservoir properties. Using the most common petrophysical and rock typing classification methods, the rock quality and heterogeneity of the C1 Sand reservoir are studied for four wells located within the 3D seismic volume. We find that integrating the geologic content of a combination of eight spectral instantaneous attribute volumes using an unsupervised machine-learning algorithm (self-organizing maps [SOMs]) results in a classification volume that can highlight reservoir distribution and identify stratigraphic baffles by correlating the SOM clusters with discrete net reservoir and flow-unit logs. We find that SOM classification of natural clusters of multiattribute samples in the attribute space is sensitive to subtle changes within the reservoir’s petrophysical properties. We find that SOM clusters appear to be more sensitive to porosity variations compared with lithologic changes within the reservoir. Thus, this method helps us to understand reservoir quality and heterogeneity in addition to illuminating thin reservoirs and stratigraphic baffles.

Poisson impedance for lithology characterization

Tyumen formation is a formation with very considerably lateral heterogeneity. It influences on process of exploration oil and gas traps. No one can deny that seismic survey is a main method for prediction lateral heterogeneity. Nowadays, geoscientists can model different useful seismic attributes, apart from structural maps. Seismic inversion is most effective tool for transformation amplitude seismic data into elastic properties. However, sometimes result of seismic inversion is very hard to understand and analyze. In this work, authors will show additional methods for better understanding results of seismic inversion using Impedance Poisson. Area of research is located in the Western Siberia.

A new deeper channel system predicted using seismic attributes in scarab gas field, west delta deep marine concession, Egypt

The west Delta deep marine (WDDM) concession is one of the most gas-producing provinces in Egypt, due to its widely distributed channel systems. This paper aims to illustrate the role of seismic attributes analysis to improve tracing of subtle stratigraphic features, such as shallow fluvial channel systems. The study is carried out in Scarab gas field, located in the WDDM area, which embodies deep marine slope channels. Two vertically overlapped channels, channel-1 and channel-2 (of Pliocene age), are considered the chief channels constituting reservoirs of this area. The methodology implemented here is to analyze the available seismic data volume, using the spectral decomposition and root mean square (RMS) attributes, to highlight the occurrence and extension of channel features on the stratigraphic surface of interest and within a certain time interval slice. These subtle features are difficult to investigate using traditional amplitude seismic data. The use of spectral decomposition tool and root mean square attribute is proved to help identify and highlight the channels and bright spot features, which indicate the gas accumulation. In addition, faults are easily recognized using these attributes. Below the total depths of the present wells, a new deeper channel is expected to be apparent below channel-2 (below the total depths of wells Dc) and to the west of channel-16 at the time slice −2460 ms (about −2035 m depth). The major part of this channel is meandering and trends from south to north where it is wider to the north forming an area of a bright spot. This channel may be filled with gas-prone sand, as deduced by the used seismic attributes. This expectation is proposed depending on the high similarity between the characteristics of the used seismic attributes of the drilled channels (1 and 2) and that of the anticipated ones. More detailed studies and drilling to a depth of approximately −2000 m at a proposed well, located about 1 km to the north of well Dc, are recommended for exploration of the detected channel.

Application of Artificial Neural Network in seismic reservoir characterization: a case study from Offshore Nile Delta

The Prediction of the reservoir characteristics from seismic amplitude data is a main challenge. Especially in the Nile Delta Basin, where the subsurface geology is complex and the reservoirs are highly heterogeneous. Modern seismic reservoir characterization methodologies are spanning around attributes analysis, deterministic and stochastic inversion methods, Amplitude Variation with Offset (AVO) interpretations, and stack rotations. These methodologies proved good outcomes in detecting the gas sand reservoirs and quantifying the reservoir properties. However, when the pre-stack seismic data is not available, most of the AVO-related inversion methods cannot be implemented. Moreover, there is no direct link between the seismic amplitude data and most of the reservoir properties, such as hydrocarbon saturation, many assumptions are imbedded and the results are questionable. Application of Artificial Neural Network (ANN) algorithms to predict the reservoir characteristics is a new emerging trend. The main advantage of the ANN algorithm over the other seismic reservoir characterization methodologies is the ability to build nonlinear relationships between the petrophysical logs and seismic data. Hence, it can be used to predict various reservoir properties in a 3D space with a reasonable amount of accuracy. We implemented the ANN method on the Sequoia gas field, Offshore Nile Delta, to predict the reservoir petrophysical properties from the seismic amplitude data. The chosen algorithm was the Probabilistic Neural Network (PNN). One well was kept apart from the analysis and used later as blind quality control to test the results.

Q-compensated reverse time migration in tilted transversely isotropic media

Anisotropy and absorption are critical to the modeling and analysis of seismic amplitude,phase, and traveltime data. To neglect any of these phenomena, which are often bothoperating simultaneously, degrades the resolution and interpretability of migrated images.However, a full accounting of anisotropy and anelasticity is computationally complex andexpensive. One strategy for accommodating these aspects of wave propagation, while keepingcost and complexity under control, is to do so within an acoustic approximation. Weset up a procedure for solving the time-domain viscoacoustic wave equation for tilted transverselyisotropic (TTI) media, based on a standard linear solid model and, from this, developa viscoacoustic reverse time migration (Q-RTM) algorithm. In this approach, amplitudecompensation occurs within the migration process through a manipulation of attenuationand phase dispersion terms in the time domain differential equations. Specifically, theback-propagation operator is constructed by reversing the sign only of the amplitude lossoperators, but not the dispersion-related operators, a step made possible by reformulatingthe absorptive TTI equations such that the loss and dispersion operators appear separately.The scheme is tested on synthetic examples to examine the capacity of viscoacoustic RTM to correct for attenuation, and the overall stability of the procedure.

3D seismic fault detection using the Gaussian process regression, a study on synthetic and real 3D seismic data

Faults are natural tectonic events as planar features along which rock units are moved. The patterns of these planes on seismic sections include linear or curvilinear features that are called edge in image processing, where amplitudes sharply change. In seismic data, these edge features made by fault traces are usually detected by seismic attributes, which need complex mathematical calculation such as a dip-steered cube. In this study, we introduce faults as global anomalies which disturb the normal interaction of seismic reflectors. Here, the normal interaction means the absence of intensive changes in reflectors trend in seismic sections. Fault detection as a global anomaly is done by the Gaussian process regression model (GPR), which is a nonparametric probabilistic model based on Bayesian statistics supporting noisy (Gaussian noise) and sparse features in data. In data mining, anomaly detection identifies items or events that do not match an expected pattern in a dataset. Global anomalies are sparse and affect a wide range of normal trends of data which are also held by fault features in seismic data. In this study, the GPR-based anomaly detection algorithm was implemented on the 3D seismic data of the Gulf of Mexico containing normal growth faults and a salt dome to detect salt boundary and fault traces in seismic data. In this respect, the reflections from rock units and fault features were taken into account as normal interactions and global anomalies, respectively, because faults disrupted the normal trend of reflectors in seismic sections. To detect fault locations, the location of a voxel in voxel grid where it is a part of fault in seismic data, after smoothing the seismic data, a Gaussian process (GP) model was trained on seismic data, attempting to describe the seismic amplitude data as a multivariate Gaussian model. However, GP regression fails to describe the seismic data at fault locations. Thus, the failure of the GP in the regression step was analyzed to separate the probable fault points, highlighted by calculating the variance of the GPR results. Finally, the detected probable fault points were improved and separated from background results by implementing a consistent reconstruction morphological algorithm. The results were validated using a similar structural index method, mean square error, and power signal to noise ratio indices in comparison with interpreted faults, implying the superiority of the proposed method in comparison with seismic attributes. The faults detected by the proposed method have the most structural similarity to faults interpreted. This similarity has improved by 22% compared to used attributes.