3D seismic data de-noising and reconstruction using Multichannel Time Slice Singular Spectrum Analysis

In the previous chapters, we have seen that time domain methods are good for denoising the seismic data with moderate noisy record. In order to separate true reflection amplitudes from the noisy seismic data, in this chapter, we discuss time slice singular spectrum analysis (TSSSA), a data adaptive signal decomposition based on SSA to alleviate such coloured and complex noise. This method is the time domain version of Cadzow filter but did not gain full attention until 2012. This method is applicable to pre-stack and post-stack data for denoising. By virtue of using seismic amplitudes corresponds to particular time (time slice) to benefit their strong correlation, the method was named as Time Slice SSA. It is a time domain method. In view of the above, in this chapter, Time Slice SSA (TSSSA) denoising algorithm was presented for denoising of 2D seismic reflection data. The first part of this chapter is devoted to provide the testing of the method on synthetic seismic data in comparison with f-x deconvolution (Canales 1984) and f-x SSA (Sacchi 2009) methods. In the second part, the method was applied to high-resolution seismic pre-stack and post data field data.

Seismic data incorporation in reservoir simulation models history matching (HM) studies has been continuously growing. 4D seismic data, in contrast with well production data, can provide a very good scenario of fluids arrangement along reservoir. In this work we describe how 3D and 4D seismic data gathered in acquisitions performed in Campos Basin was incorporated in Marlim Sul deep water field geological model reconstruction and in assisted HM (AHM). It is taken advantage of both 3D and 4D seismic data in several stages of the study, for instance, in the construction of a new porosity – most influential in impedance – model by using a methodology based on the inversion of synthetic seismic (calculated by petro-elastic model) in porosity through an optimization process that aims to reduce the difference between observed and synthetic impedance, and when defining influential parameters based on fluids displacement registered by seismic signal, by using a technique based on the creation of transmissibility multipliers parameters regions that considers the fluids displacement shown in 4D signal. Another relevant point is the use of information from reservoir and 3D seismic data when weighting the 4D data in the objective function. Combining the above mentioned techniques with the knowledge of the field – supported by the 3D seismic data – which allowed, for instance, identification of faults – where fault transmissibility multipliers were used as parameters in the HM process – a fairly good agreement on the observed well and seismic production data was achieved. HM studies using AHM tools have been shown a much more time-efficient technique when compared to manual HM. The incorporation of 4D seismic data can considerably improve the HM quality by improving the reservoir description, once it increases the ability of describing fluids arrangement and pressure distribution. The techniques successfully applied in the Marlim Sul field HM support these conclusions.

In some oilfield where 3D seismic survey has been done, the deeper structure cannot be discovered for poor deep seismic data. Layer stripping technique applied both seismic and gravity data is a solution, but it cannot get satisfied processing effect because the horizontal variation of formation density is ignored. To improve the gravity survey effect in 3D seismic survey area, this paper presents the formation separation technique. Based on 3D seismic depth data and the transformed density data from 3D seismic velocity data, the upper formation's gravity effects are calculated and removed from Bouguer gravity and then the formation‐separated gravity‐anomaly was obtained, which mainly reflects the deeper geological structure. In block XX, the seismic data of shallow formations are excellent but that of top of basement are poor. The formation‐separated gravity‐anomaly processed under the control of 3D seismic data fits well with the known seismic interpretation and wells. It makes the geological interpretation more reliable

Seismic data offer valuable spatial information that contributes to the development strategy, such as drilling decisions and well planning. Inverting 3D seismic reflectivity data to impedance unlocks the potential of using seismic data to estimate the lateral variability and to predict reservoir geomechanical behaviour. The geomechanical model contributes for the evaluation of wellbore stability, frackability and drilling direction. The objective of this work is to illustrate the use of seismic derived models for reservoir and geomechanical properties, such as total porosity, total organic carbon (TOC) and rock rigidity. These properties are instrumental to characterize tight reservoirs and useful tools for defining a horizontal well trajectory.The study area has an exploration potential of unconventional reservoirs, targeting tight carbonates from Upper Jurassic. To date, only one well has been drilled in the area of the interest and high quality 3D seismic data is available. The process consists of conducting seismic inversion, using the well and the 3D PSTM seismic data, and resulting in the generation of 3D subsurface impedance model. The well-log relationships between impedance and reservoir quality (effective porosity, source richness) and completion quality (brittleness) were determined and showed high correlation. Hence, 3D rock quality and geomechanical volumes were derived to characterize the behaviour of the reservoir target.Seismic derived properties models yield a better description of the reservoir heterogeneity than using seismic reflectivity data only. The horizontal well trajectory was planned taking into account the observed and significant lateral variability revealed by the 3D models estimated based on seismic inverted data.The results obtained using this methodology demonstrate the importance of integrating geophysical characterisation as a relevant supporting factor in the drilling decision making process. In particular, it validates the use of seismic derived reservoir and completion quality models as an additional tool in the definition of a horizontal well trajectory in a tight/unconventional deep gas target.

Interpretation of three-dimensional (3D) seismic data from the northern part of central Saudi Arabia allowed mapping of the Permo-Carboniferous Unayzah Formation's aeolian sand deposits and fluvial channels. These types of depositional environments have been associated frequently with better reservoir-quality sandstones within Saudi Arabia. The depositional setting of the Unayzah reservoirs can be highly variable and range from glacial, to lacustrine, to meandering fluvial, to aeolian, to braided stream. Seismic amplitude data were instrumental in mapping a truncation of the upper Unayzah reservoir against the overlying Pre-Khuff Unconformity. The use of seismic attributes, such as coherence and curvature, indicated the existence of faulting and a fluvial channel network at the top of the Unayzah Formation. The integration of well and seismic amplitude data, via acoustic impedance inversion, revealed the direction of potentially higher porosity fairways within the channel system. The structural and depositional architecture depicted in this study is being appraised through drilling of an exploration well.

Relevance of the work. The paper considers challenging problems related with outlining of intervals with oil and gas presence in the mature Khylly field by use of latest 3D seismic survey techniques in order to gain larger crude resources base. The purpose of this research is to discover the most promising intervals of target horizons with relatively high reservoir properties outlined by 3D seismic data. The subjects of research are 3D seismic survey data, downhole seismic survey – Vertical Seismic Profiling (VSP) and well logging diagrams. The object of research is the Khylly deposit. The paper describes in brief geological and geophysical characteristics, stratigraphic and lithological features of rocks making the section. It is noted that despite repeated surveys by use of various geological and geophysical techniques, the field setting is not thoroughly studied and it has been covered by 3D seismic survey in 2012. Research results. 3D seismic survey applied across Khylly area is resulted in drawing of 4 structural maps for III and I horizons of Productive Series (PS), Akchagyl and Lower Absheron suites. Taking into account the relevance of structural planes of various stratigraphic levels and III horizon of PS being one of the major reference horizons the paper gives description of structural map drawn for this horizon. The detailed velocity model is designed based on VSP data with wide use of velocity analysis data. It has been made clear that Khylly area has block structure and each block has been described in detail. Based on acquired data it has been recommended to drill exploratory well R-1. Conclusion. Processing and interpretation of seismic data are aimed at solving some geological problems; the main task was to obtain results that ensure the study of the geological structure in the seismic survey area, including tracing of seismic horizons, faults and outlining the areas and section intervals which may be of interest due to possible oil and gas presence. VSP data acquired in well 2012 and velocity analysis made it possible to design velocity model of the section under the study, with the use of which the temporary 3D cube was transformed into a depth cube. The quality of seismic data is good and made it possible to solve the tasks set for this research.

Time-lapse (4D) seismic inversion is the leading method to quantitatively monitor fluid-flow dynamics in the subsurface, with applications ranging from enhanced oil recovery to subsurface CO2 storage. The process of inverting 4D seismic data for reservoir properties is a notoriously ill-posed inverse problem due to the band-limited and noisy nature of seismic data and inaccuracies in the repeatability of 4D acquisition surveys. Consequently, ad-hoc regularization strategies are essential for the 4D seismic inverse problem to obtain geologically meaningful subsurface models and associated 4D changes. Motivated by recent advances in the field of convex optimization, we propose a joint inversion-segmentation algorithm for 4D seismic inversion that integrates total variation and segmentation priors as a way to counteract missing frequencies and present noise in 4D seismic data. The proposed inversion framework is designed for poststack seismic data and applied to a pair of seismic volumes from the open Sleipner 4D seismic data set. Our method has three main advantages over state-of-the-art least-squares inversion methods. First, it produces high-resolution baseline and monitor acoustic models. Second, it mitigates nonrepeatable noise and better highlights real 4D changes by leveraging similarities between multiple data. Finally, it provides a volumetric classification of the acoustic impedance 4D difference model (4D changes) based on user-defined classes (i.e., percentages of speedup or slowdown in the subsurface). Such advantages may enable more robust stratigraphic/structural and quantitative 4D seismic interpretation and provide more accurate inputs for dynamic reservoir simulations. Alongside presenting our novel inversion method, we introduce a streamlined data preprocessing sequence for the 4D Sleipner poststack seismic data set that includes time-shift estimation and well-to-seismic tie. Finally, we provide insights into the open-source framework for large-scale optimization that we used to implement the proposed algorithm in an efficient and scalable manner.

Seismic interpretation is an essential factor in the success of field development projects because it allows geologists and engineers to better understand subsurface formations and identify potential resources; however, noisy seismic data, especially in regions with complex geological structures, frequently hinder this process. Such noise emerging from equipment limitations, environmental interference, and inherent geological complexities can obscure essential geological features, making accurate subsurface imaging challenging and reducing the effectiveness of traditional interpretation methods. This paper presents a machine learning (ML) method to enhance the conditioning of 2D and 3D post-stack seismic data by targeting noise and migration artifacts. These issues are often responsible for compromising the quality of seismic data sets. The method this paper presents leverages convolutional neural networks (CNNs) to identify and correct spatial inconsistencies within the data. With minimal preprocessing, requiring only the conversion of data into a 3D NumPy array format, this approach simplifies the workflow while improving data clarity and accuracy. Early statistical analysis of outliers and data normalization provides the ML model with stable inputs, allowing it to iteratively train on raw data and infer the original signal patterns in an unsupervised manner without the need for labeled data sets. The model outputs a conditioned version of the seismic data, significantly reducing noise while preserving critical geological features. The results demonstrate that ML seismic conditioning improves fault imaging, horizon continuity in areas with low signal-to-noise ratios (SNR), and localized smoothing without compromising resolution. Furthermore, the method presented in this paper retains signal fidelity and relative amplitude strengths, ensuring that geological interpretations remain accurate and reliable. By automating the traditional manual task of data conditioning, the proposed solution optimizes interpreter workflows, allowing them to focus on higher-level geological analysis and decision-making. This approach ultimately increases the accuracy of subsurface interpretations and resource exploration, improving the overall quality and efficiency of seismic interpretation projects. In addition, for 2D seismic, traditional methods of denoising rely on expert judgement using tools such as f-x deconvolution. 2D CNNs offer an efficient alternative, especially when training data can be synthesized and leveraged in supervised learning. This method can help the CNN identify and remove noise while preserving pertinent subsurface structures. This paper presents a 2D CNN architecture containing a model trained with synthetic seismic data to improve SNR of field seismic data; hence, interpretation.

Defining the root zone of mud volcanoes (MVs), structural interpretation, and geologic modeling of their body is a problematic task when only seismic data are available. We have developed a strategy for integration of gravity and seismic data for better structural interpretation. Our strategy uses the concept of the normalized full gradient (NFG) for integration of gravity and seismic data to define geometry and the root zone of MVs in the southeast onshore of the Caspian Sea. Our strategy will increase the resolution of the seismic envelope compared with the conventional Hilbert transform. Prior to interpretation, we applied the NFG method on field gravity data. First, we perform a forward-modeling step for accurate NFG parameter definition. Second, we estimate the depth of the target, which is the root zone of the MV here. Interpretation of field gravity data by optimized NFG parameters indicates an accurate depth of the root zone. Subsequently, we apply the NFG method with optimized parameters on a 2D seismic data. Application of our strategy on seismic data will enhance resolution of the seismic image. The depth of the root zone and the geometry of the MV and mud flows was interpreted better on the enhanced image. It also illustrates the complex structure of a giant buried MV, which was not well-interpreted on conventional seismic image. Interpretation of the processed data reveals that the giant MV had lost its connection to its reservoir, whereas the other MV is still connected to the mud reservoir. The giant MV is composed of complex bodies due to pulses in the mud flows. Another MV in the section indicates narrow neck with anticline and listric normal faults on its top. Thus, application of the NFG concept on seismic image could be considered as an alternative to obtain enhanced seismic image for geologic interpretation.

The W block shale in the study area is located at the West Canada basin and belongs to the low-porosity and ultra-low permeability gas-condensate reservoirs. The sandwiched layers in the shale are extremely important for reservoir reconstruction. As the single-layer thickness of the carbonate interbeds in the W block shale is very thin and have multi-stage superposition characteristics, the use of conventional geostatistical inversion cannot effectively and accurately predict the thin sweet spots and interbeds in the shale. This article uses three-dimensional probability constraints as pre-stack geostatistical inversion to replace traditional one-dimensional and two-dimensional face constraint inversion and achieved quantitative prediction of interbeds using well logging and 3D seismic data, so that prediction outcomes fully integrates well logging, seismic data and established a set of 3D probability constraint pre-stack geostatistical inversion process for thin interbeds and sweet spots in the shale. The process mainly includes the following crucial steps: (1) Well logging evaluation of shale reservoir parameters, analysis of shale rockphysical parameters and analysis of petrophysical characteristics of shale sweet spots and carbonate interbeds from well data, in order to evaluate and use seismic data to predict the possibility of shale sweet spots and interbeds. The characteristics of shale sweet spots is low Poisson ratio, moderate Young's modulus, high total organic content (TOC) and high brittleness. The characteristics of carbonate interbeds is high Poisson ratio, high Young's modulus, low TOC and high density. (2) Pre-stack gather optimization to establish six reliable partial sub-offset seismic data volumes. (3) Application of waveform difference simulation for prediction of shale interbeds, analysis of the vertical and horizontal variation characteristics of the interbeds based on post-stack seismic data and the three-dimensional probability volume of carbonate interbeds was estabilished according to the results of seismic waveform difference simulation and well logs. (4) High resolution 3D lithologic probability constrained pre-stack geostatistical inversion are performed and effectively predicted the spatial variation characteristics of shale thin interbeds and sweet spots.

Integration of 3D and time-lapse 4D seismic data into reservoir modelling and history matching processes poses a significant challenge due to the frequent mismatch between the initial reservoir model, the true reservoir geology, and the pre-production (baseline) seismic data. A fundamental step of a reservoir characterisation and performance study is the preconditioning of the initial reservoir model to equally honour both the geological knowledge and seismic data. In this paper we analyse the issues that have a significant impact on the (mis)match of the initial reservoir model with well logs and inverted 3D seismic data. These issues include the constraining methods for reservoir lithofacies modelling, the sensitivity of the results to the presence of realistic resolution and noise in the seismic data, the geostatistical modelling parameters, and the uncertainties associated with quantitative incorporation of inverted seismic data in reservoir lithofacies modelling. We demonstrate that in a geostatistical lithofacies simulation process, seismic constraining methods based on seismic litho-probability curves and seismic litho-probability cubes yield the best match to the reference model, even when realistic resolution and noise is included in the dataset. In addition, our analyses show that quantitative incorporation of inverted 3D seismic data in static reservoir modelling carries a range of uncertainties and should be cautiously applied in order to minimise the risk of misinterpretation. These uncertainties are due to the limited vertical resolution of the seismic data compared to the scale of the geological heterogeneities, the fundamental instability of the inverse problem, and the non-unique elastic properties of different lithofacies types.

Three-dimensional (3D) seismic data visualization is a critical technique in 3D seismic data interpretation. In the paper, the characteristics of seismic data, the technology of 3D data sets visualization and the technology of 3D graphics construction and display using interactive data language (IDL) were discussed. The author proposed new solutions for 3D seismic data visualization, moreover expounded how to reorganize 3D seismic data and how to realize direct volume rendering and surface rendering using IDL. Using the interface provided by IDL based on COM standard, the author applied the new solutions to practice and built a 3D Seismic Data Visualization Platform..

In the study area, southeast of Ordos Basin in China, thick lacustrine shale/mudstone strata have been developed in the Triassic Yanchang Formation. Aiming to study these source/reservoir rocks, a 3D full-azimuth, high-density seismic survey was acquired. However, the surface in this region is covered by a thick loess layer, leading to seismic challenges such as complicated interferences and serious absorption of high frequencies. Despite a specially targeted seismic processing workflow, the prestack Kirchhoff time-migrated seismic data were still contaminated by severe noise, hindering seismic inversion and geologic interpretation. By taking account of the particular data quality and noise characteristics, we have developed a cascade workflow including three major methods to condition the poststack 3D seismic data. First, we removed the sticky coherent noise by a local pseudo [Formula: see text]-[Formula: see text]-[Formula: see text] Cadzow filtering. Then, we diminished the random noise by a structure-oriented filtering. Finally, we extended the frequency bandwidth with a spectral-balancing method based on the continuous wavelet transform. The data quality was improved after each of these steps through the proposed workflow. Compared with the original data, the conditioned final data show improved interpretability of the shale targets through geometric attribute analysis and depositional interpretation.

Deep learning has been successfully applied to image denoising. In this study, we take one step forward by using deep learning to suppress random noise in poststack seismic data from the aspects of network architecture and training samples. On the one hand, poststack seismic data denoising mainly aims at 3-D seismic data. We designed an end-to-end 3-D denoising convolutional neural network (3-D-DnCNN) that takes raw 3-D cubes as input in order to better extract the features of the 3-D spatial structure of poststack seismic data. On the other hand, denoising images with deep learning require noisy-clean sample pairs for training. In the field of seismic data processing, researchers usually try their best to suppress noise by using complex processes that combine different methods, but clean labels of seismic data are not available. In addition, building training samples in field seismic data has become an interesting but challenging problem. Therefore, we propose a training sample selection method that contains a complex workflow to produce comparatively ideal training samples. Experiments in this study demonstrate that deep learning can directly learn the ability to denoise field seismic data from selected samples. Although the building of the training samples may occur through a complex process, the experimental results of synthetic seismic data and field seismic data show that the 3-D-DnCNN has learned the ability to suppress the Gaussian noise and super-Gaussian noise from different training samples. Moreover, the 3-D-DnCNN network has better denoising performance toward arc-like imaging noise. In addition, we adopt residual learning and batch normalization in order to accelerate the training speed. After network training is satisfactorily completed, its processing efficiency can be significantly higher than that of conventional denoising methods.

Imaging Pleistocene volcanic edifices along the Egyptian Red Sea margin: Insights from reflection seismics and 3D constrained inversion of gravity and magnetic data