Well-log Data Research Articles

Confined turbidite system of the upper Paleozoic Itararé Group, Paraná Basin, Brazil: Beyond the topographic control paradigm

Recent studies in the eastern margin of the Paraná Basin, about the initial deposition of the Taciba Formation (Itararé Group) revealed a preserved, partially exhumed glacial trough as indicated by the morphology of its basal nonconformity. In this work, we reinforce the importance of the glacially-carved topography controlling the subsequent sedimentation by emphasizing turbidity currents and their deposits. This glacial trough is large enough to confine part of the following deglacial sediments, including a sand-rich turbidite system that represent phases of more consistent sand supply, a result from high sedimentation rates associated either with direct influx of meltwater-derived jets and/or with a rapid delta progradation at the trough head, causing over-steepening, unsteadiness, and eventual failure of the delta-front deposits. This turbidite system consists of three sand-rich stages (Ts1 to Ts3) intercalated with three fine-grained intervals (F1 to F3). Ts stages form bedsets with an overall lenticular geometry, thinning-out toward the trough flanks where the individual sandstone beds pinch-out close to the metamorphic Precambrian bedrock. They are ascribed to sinuous, laterally extensive channelized bodies representing different phases of channel activity through time. In general, each stage consists of an initial deposition derived from long-lived turbidity currents triggered by flood-driven hyperpycnal flows followed by short-lived, surge-type turbidity currents. F1, F2 and F3 deposits consist of fine-grained intervals that separate Ts1, Ts2 and Ts3 stages by an abrupt change in grain size and bed thickness. They comprise thin, sand-to-mud couplets showing granule-to-boulder sized basement-affinity dropstones, and several minor slumps, which are interpreted as overbank/levee sedimentation taking place during the successive passage of turbidity currents. A paleogeographic reconstruction of the glacial landscape reveal some similarities with modern fjords of British Columbia, Canada, that can be used to geometries and facies distribution prediction as an analog to assist seismic and well-log data.

Improved prediction of shale gas productivity in the Marcellus shale using geostatistically generated well-log data and ensemble machine learning

This study proposes the application of geostatistically generated well-log data to predict well productivity in Marcellus shale reservoirs using ensemble machine learning (ESM). ESM was voting weighted by four individual machine learning methods: random forest (RF), gradient boosting, extreme gradient boosting, and light gradient boosting. The ESM model was trained based on 9921 wells that are qualified as training data. The well-log data had to be predicted using ordinary kriging for all 9921 wells because there were only 18 wells with well-log data for gamma-ray, resistivity, and density. Three cases were compared to analyze the influence of well-log data on machine learning performance. Case 1 comprised only basic well information, such as well location and fracturing-related parameters. Case 2 included the data from Case 1, and the well-log data of the closest well among the 18 wells were assigned to 9921 wells. In Case 3, to data of Case 1, the kriging-generated well-log data were added. In conclusion, Case 3 yielded the best performance among the machine learning algorithms, the highest coefficients of determination, and the lowest mean square errors. According to the RF models, the feature importance of the well-log data in Case 3 was double that in Case 2. The geostatistical method allowed us to make good use of the well-log data despite the restricted number of wells with the well-log compared to the enormous scale of the Marcellus shale. Moreover, basic well information is affordable compared to costly data such as permeability and porosity, even before the production start point. Thus, it can be applied to guide decision-makers when determining a drilling target. Furthermore, owing to its versatility in data affordability, the proposed method could be expanded to other shale basins, such as Barnett, Eagle Ford, and Scoop.

Prediction of Petrophysical Properties using Post Stack Seismic Inversion and Geostatistical Techniques over F-3 Block, Netherlands - A Comparative Study

To estimate petrophysical parameters, seismic inversion techniques have been frequently used for estimating attributes like P-impedance, elastic impedance, S-impedance, density, Vp / Vs ratio, and gamma-ray logs from seismic and well-log data. These characteristics enable us to comprehend subsurface lithology for geo-seismic analysis, including its extent and shape. Four different post-stack inversion techniques, including bandlimited inversion (BLI), colored inversion (CI), maximum likelihood sparse spike inversion (MLSSI), and model-based inversion (MBI), have been applied to the post-stack seismic data from the F3 block in the Netherlands in this study. The objective is to compare the efficacy of these inversion methods over F3 block seismic data. For post-stack inversion, the data was inverted into a very high-resolution P-impedance volume. The analysis depicts that all four inversion methods provide mutually consistent subsurface information with marginally better MBI results. Furthermore, geostatistical techniques have been used intensively for further testifying the results obtained from post-stack inversion methods. The geostatistical techniques use seismic data-derived attributes and inverted impedance-derived attributes as input to estimate P-wave velocity, porosity, and density away from the boreholes. Two geostatistical methods namely probabilistic neural network (PNN) and multilayer feed-forward neural network (MLFN) are used for the analysis. Porosity, density, and P-wave velocity have been predicted using both techniques which highlighted different characteristics of the subsurface with very detailed information. The derived results show that reservoir properties have been better estimated with the combination of MBI and PNN techniques for F3 block data over the other methods used in this study.

Spherical-wave elastic inversion in transversely isotropic media with a vertical symmetry axis

SUMMARY Although subsurface media are usually assumed to be isotropic, anisotropy is ubiquitous in crustal rocks and leads to the variation of seismic response with direction. Transversely isotropic media with a vertical symmetry axis (VTI media) are widely found in the real world, such as in textured shale reservoirs. Plane-wave reflection coefficients (PRCs) in VTI media have been widely exploited in amplitude variation with offset (AVO) inversion to estimate the elastic and anisotropy parameters of subsurface media. However, the PRCs in VTI media meet some fundamental problems, especially at near-critical or post-critical incidence angles where the spherical-wave effect is significant. To consider the wave front curvature, a complex spherical-wave reflection coefficient (SRC) in VTI media is derived. To better understand the spherical-wave seismic response in VTI media, we investigate the dependence of the complex SRC on frequency, reflector depth and Thomsen anisotropy parameters ($\varepsilon $ and $\delta $). Based on a complex convolution model, a spherical-wave AVO inversion approach in VTI media is proposed to estimate the vertical (symmetry-axis) compressional and shear wave velocities (P and S waves), density and Thomsen anisotropy parameters from observed seismic data with different incidence angle and frequency components. Synthetic data with Gaussian random noise are used to verify the robustness of the spherical-wave AVO inversion approach in VTI media. Field data examples show that the proposed approach can produce reasonable inversion results that match well with the well-logging data.

Deep learning-based upgoing and downgoing wavefield separation for vertical seismic profile data

The vertical seismic profile (VSP) considerably aids attenuation analysis and velocity calibration, enabling high-resolution seismic exploration. However, the imaging and interpretation of VSP data require pure upgoing and downgoing wavefields, and their separation is critical in processing and interpretation. Often, achieving high precision and efficient wavefield separation can be regarded as a challenging issue. To automate this process with higher accuracy, we develop a deep-learning high-precision intelligent separation method. First, to construct a physics-driven and data-generic training set, the well-log data from different geologic environments are used to simulate the upgoing and downgoing VSP data independently. Thereafter, we train a multitask-learning neural network to regress the network weights before and after separation to tune a mapping model, which is used to implement intelligent upgoing and downgoing wavefield separation. Validation data confirm the feasibility of our method. To further study the generalizability and robustness of the method, we apply the trained model to separate the modeled and real VSP data. By analyzing the separated profiles and frequency-wavenumber spectra, it is observed that the intelligent separation method is superior to conventional methods with a balance between accuracy and efficiency. It can be concluded that our method can separate upgoing and downgoing wavefields in VSP data with excellent performance, even in complex geologic environments. Our study indicates that the multitask-learning neural network combined with a physics-driven and data-generic training set is a new strategy to separate upgoing and downgoing wavefields and deserves to be popularized and applied.

Characterization of silicification and dissolution zones by integrating borehole image logs and core samples: a case study of a well from the Brazilian pre-salt

Precise knowledge of the spatial distribution patterns of non-matrix porosity zones and the establishment of the geological factors controlling their evolution are crucial for building more accurate carbonate reservoir models and improving hydrocarbon production. The occurrence of intervals affected by significant carbonate dissolution may result in drilling fluid loss and time-consuming drawbacks during well construction. Vug or cave-rich reservoirs may exhibit excess permeability and extremely high initial flow rates. Similar situations have been reported in exploration activities in the Brazilian pre-salt plays, where evidence of dissolution and other burial diagenetic processes, such as severe silicification and dolomitization, are common. In this study, we investigate evidence of major post-depositional changes in the lacustrine carbonate reservoirs of the Barra Velha Formation, which comprises the most prolific hydrocarbon play in Brazil. Using a comprehensive database consisting of both core samples and well-log data from a vertical well in the Santos Basin, we have characterized, at multiple scales, reservoir zones affected by silicification and carbonate dissolution. In addition, we have performed a petrophysical evaluation of the reservoir to understand the impact of such processes on porosity and permeability development. The results suggested an intimate relationship between silicification and dissolution processes, which can be associated with late fluid percolation under a deep burial flow regime. The occurrence of silicified and vuggy beds, associated with specific zones and lithofacies, indicates an important degree of stratigraphic control on fluid percolation and lateral migration. Furthermore, the presence of fractures at discrete stratigraphic levels has preferentially influenced the development of high-permeability zones, including metric-scale fracture-related conduits. This study contributes to the general knowledge of carbonate reservoirs affected by silicification and dissolution while providing support for the recognition of such processes in partially- or non-cored wells.

Three-dimensional characterization of the Búzios Field reveals implications on the tectonostratigraphic evolution of the Santos Basin using combined seismic attributes and well-logs analysis

The Búzios Field is one of the most prolific petroleum fields along the Eastern Brazilian Margin. Despite significant research conducted restrictedly on the pre-salt reservoirs, few studies have focused on the three-dimensional characterization of the entire Búzios Field area. Therefore this work aims to spatially characterize a volume along the Búzios Field area using 3D seismic volume interpretation, seismic attributes, and well-log data. The combined use of these tools allowed the understanding of several parameters concerning the tectonostratigraphic evolution of the Santos Basin. Within the mapped area, we identified five units: Syn-Rift I, Syn-Rift II, Post-Rift I, Post-Rift II, Drift I, and Drift II. The 3D model reveals significant features such as the depth variation of the Syn-Rift I unit and the heterogeneous morphology of the Post-Rift II unit. The Syn-Rift I unit is deeper towards the N-NW and shallower towards the S-SE due to a regional half-grabens system. This system impacts the pre-salt geometry and thickness, as structural highs and lows are relevant factors in sediment accommodation. The Post-Rift I unit has a predominantly regular geometry and thickness but tends to thicken towards the S-SE and exhibits slight downward flexure towards the NE in the southeast of the Búzios Field. In three dimensions, the heterogeneous morphology of the Post-Rift II unit is evidenced by the variable thickness and irregular topography, accounting for the intense halokinesis in the Santos Basin. Above the evaporite layer, the thickness of the Drift I unit is highly conditioned by salt morphology, with thicker and deformed packages in areas of thinner salt section. On the other hand, the Drift II unit shows relatively constant thickness with minor variations near fault zones. Comparisons with other fields within the Santos Basin (Tupi and Sapinhoá/Lapa) reveal differences in layer thickness, mobilization, and rotation of the syn-rift blocks. The Post-Rift I layer (Barra Velha Fm.) and the Post-Rift II package (Ariri Fm.) are thinner and less mobilized in the Búzios Field area. Moreover, the syn-rift blocks are less rotated, and the internal reflectors have a lower inclination in the study area when compared to these other fields. Therefore, this research concludes that the lower mobilization of salt, the lower thickness of the post-rift package, and the minor syn-rift block rotation observed in the Búzios Field area are associated with the relative distance from the continental margin and, consequently, with the amount of displacement generated by the South Atlantic Rift. In this case, the Búzios Field is located in a more proximal portion than the Tupi and Sapinhoá/Lapa fields, exhibiting features suggestive of a lower displacement.

Detailing the pore structure of productive intervals within oil wells using 3D color imaging

The article describes an approach to the expansion of the methodology for applying hydraulic fracturing in oil fields by adding the possibilities of 3D modeling with color imaging of the pore structure within productive intervals of the wells. As an applied example, the geological and geophysical section of the productive level of one of the wells in Moscudinskoye oil field with known data on the integrated interpretation of the results of well logging and microcomputer tomography was chosen. According to well-logging data, the productive reservoir in the analyzed area of the section is characterized by a high degree of heterogeneity. The tomography studies of a full-size core enabled to identify four lithotypes here with different pore structure features. The consideration of the identified reservoir heterogeneity, as well as the data on the thickness and other characteristics of reservoir properties of individual lithotypes which make up the section, allowed to significantly increasing the detail of the final geological model of the wellbore section. A distinctive feature of this final geological model is the use of the method of enlargement of the initial data array by adding intermediate values which were calculated theoretically. The visibility of the final geological model of the borehole walls is provided by color 3D imaging of the calculated data of the enlarged massif and gives an idea on the presence of areas with good and weak fluid conductivity on the lateral surface of the borehole walls. According to this model, the intrastratal transverse and longitudinal fluid-conducting “corridors” are observed in the circumwell zone that define the hydro-dynamic movements of natural and artificial fluids in the medium of productive reservoirs.

Evaluation of reservoir subsidence due to hydrocarbon production based on seismic data

Environmental problems associated with depleted oil and gas reservoirs upon long-term production from them are likely to become important challenges in future decades. With the increasing trend of production from hydrocarbon reservoirs, more and more reservoirs across the world are reaching the second half of their life—a fact that places an emphasis on the necessity of investigating what is known as reservoir subsidence. Different analytical and numerical approaches have been introduced for analyzing the subsidence on the basis of the elasticity theory but in the form of case studies, leaving a comprehensive model yet to be proposed. In this work, a formulation was introduced for estimating reservoir subsidence by integrating the rock physics, rock mechanics, and thermo-poroelasticity theories. Then, a modified version of this formulation was developed to calculate compaction in an actively producing reservoir that is suspect of subsidence, as a case study. For this purpose, triaxial hydrostatic tests were carried out on core plugs obtained from the considered reservoir, and then, compaction parameters (i.e., compression index and coefficient of deformation) were obtained at a laboratory scale. In order to evaluate the subsidence at a reservoir scale, the laboratory-scale results and in situ reservoir properties were integrated with well-logging and 3D seismic data at well location to come up with 3D cubes of compaction information. Continuing with the research, time-dependent inelastic deformation was modeled considering continued production for different future periods. The field observations showed that the estimated compaction is not visible at the surface in the form of subsidence due to the high depth and stiffness of the studied reservoir. However, collapse of casing at some of wells drilled into the studied reservoir could be attributed to the reservoir subsidence. Finally, variations of compaction with pore pressure were investigated to propose a model for predicting the subsidence in future periods. Findings of this research can be used to forecast subsidence at well location to take the required measures for avoiding possible casing collapse and/or relevant environmental issues.

Sequence stratigraphy and reservoir quality of the Gulf of Suez syn-rift deposits of the Nukhul formation: Implications of rift initiation and the impact of eustacy and tectonic on deposition

The Nukhul Formation represents the sediments deposited during the earliest phase of the syn-rift period in the Gulf of Suez rift. It is the primary hydrocarbon reservoir for fifteen fields in the region. However, the timing of rifting needs to be better constrained, and the influence of local tectonics and global eustatic sea-level fluctuations on reservoir quality and deposition is still being determined. This research aims to establish a stratigraphic sequence framework, determine the age of the Suez Rift commencement, use changes in paleobathymetry, and assess reservoir quality. The methods employed involve the analysis of four boreholes in the Abu Rudeis-Sidri field, incorporating biostratigraphy, well-logs, and 3D seismic data. The integrated paleo-structure and biostratigraphic analyses indicate that the beginning of rifting in the Gulf of Suez occurred at 23.53 Ma at the first common occurrence (FCO) of Trilobatus primordius. The paleobathymetric analysis of the Nukhul Formation reveals deeper paleo-water depths during certain intervals, indicating deposition during a phase of high subsidence and rift-climax. The Nukhul stratigraphic sequence reflects sea-level changes leading to shifts in accommodation, and facies changes, erosion, and sequence formation during the late Chattian and early Burdigalian stages. However, the reservoir characterization of the Nukhul Formation is significantly affected by various factors, including cementation, pore system properties, sediment sorting, rift system geometry, and relative sea-level fluctuations. These factors must be carefully considered in any regional exploration or development efforts. The study also reveals that tectonic activity played a role in forming fault-bounded basins, and eustatic sea-level changes influenced the rate and timing of sediment accumulation. These findings can aid in predicting the distribution of the Nukhul sedimentary deposits and improve the ability to explore and develop their resources.

Petrophysical log-driven kerogen typing: unveiling the potential of hybrid machine learning

The importance of characterizing kerogen type in evaluating source rock and the nature of hydrocarbon yield is emphasized. However, traditional laboratory geochemical assessments can be time-intensive and costly. In this study, an innovative approach was taken to bridge this gap by utilizing machine learning techniques to ascertain key parameters—Organic Oxygen Index (OI), Hydrogen Index (HI), and kerogen type—from petrophysical logs of a well in the Perth Basin, Western Australia. This approach assembled geochemical data from 138 cutting samples of the Kockatea and Woodada formations and petrophysical log data. Subsequently, six machine learning algorithms were applied to predict the OI and HI parameters. The efficacy of these methods was assessed using statistical parameters, including Coefficient of Determination (R2), Average Percentage Relative Error, Average Absolute Percentage Relative Error, Root Mean Square Error, and Standard Deviation. The Support Vector Machines method emerged as the standout performer, with an R2 of 0.993 for the OI and 0.989 for the HI, establishing itself as an optimal tool for predicting these indices. Additionally, six classifiers were employed to determine kerogen types, with accuracy tested using precision, recall, F1-Score, and accuracy parameters.The study's findings highlight the superiority of the Gradient Boosting method in kerogen-type classification, achieving an impressive accuracy rate of 93.54%. It is concluded that when utilized with petrophysical logs, machine learning methodologies offer a powerful, efficient, and cost-effective alternative for determining OI, HI, and kerogen type. The novelty of this approach lies in its ability to accurately predict these crucial parameters using readily available well-log data, potentially revolutionizing traditional geochemical analysis practices.Graphical abstract

Integrated well-log data and seismic inversion results for prediction of hydrocarbon source rock distribution in W segment, Pearl River Mouth Basin, China

Hydrocarbon source rocks possess a high concentration of organic matter and can generate significant quantities of hydrocarbons, which are integral components within hydrocarbon systems. It is critical for well deployment and hydrocarbon production enhancement to accurately anticipate the geographical dispersion of hydrocarbon source rocks in hydrocarbon-bearing basins. Total Organic Carbon (TOC) serves as a pivotal metric for assessing hydrocarbon source rocks, enabling the estimation of their spatial distribution through TOC prediction. In this study, we present a Machine Learning-based process for predicting the distribution of hydrocarbon source rocks, utilizing well-log data and seismic inversion results. Based on the TOC value of the well curve, the hydrocarbon source rock quality is divided into four categories: normal, medium, good, and excellent. We identify the sensitive parameters governing hydrocarbon source rock quality through cross-plot and correlation analysis. Machine Learning will predict the spatial distribution of hydrocarbon source rock by establishing a relationship between hydrocarbon source rock quality and sensitive parameters. The prediction results are evaluated by comparing the performance of Random Forest, Gradient Boosting Decision Tree (GBDT), eXtreme Gradient Boosting tree (XGBoost), and Histogram-Based Gradient Boosting Decision Tree (HGBDT) using confusion matrix and ROC curve. The findings consistently demonstrate that HGBDT exhibits superior prediction capability, accurately anticipating the spatial distribution of hydrocarbon source rocks. The prediction outcomes align well with the depositional properties of the research area and demonstrate excellent agreement with the well data.

A committee machine-based estimation of shear velocity log by combining intelligent systems and rock-physics model using metaheuristic algorithms

Estimation of reservoir parameters such as porosity, lithology and determination of elastic moduli is based on well-logging data and seismic velocity information. However, the lack of shear velocity log in many wells, and the high costs of its acquisition have led to developing different approaches for shear velocity estimation. Currently, shear velocity estimation is accomplished by common industrial methods such as empirical equations, regression analysis, intelligent methods and rock-physics models (RPMs). The current study proposes that robust, accurate and cost-effective intelligent methods coupled with suitable RPM provide reasonable estimations of the shear velocity in an optimal processing time. Hybridizing intelligent models with RPMs in the framework of a committee machine is an optimal solution for shear velocity estimation of carbonate reservoirs. The intelligent methods used in this study include neural network (NN), fuzzy logic (FL) and an adaptive-neuro fuzzy inference system (ANFIS). In this research, for each individual intelligent model and a developed RPM, the optimal weights were allocated by Metaheuristic optimization algorithms including genetic algorithm (GA), ant colony optimization routing (ACOR) and simulated annealing (SA)). To obtain the best results with the lowest mean squared error (MSE), a total of 9408, 10801 and 12288 models were created and tested in MATLAB to optimally parametrize the GA, ACOR and SA, respectively. The results showed that the combination of the applied intelligent models with the developed RPMs leads to a higher correlation coefficient with reference to the measured Shear velocity in test dataset. The use of a combination of intelligent methods in the form of committee-machine instead of individual methods has led to a reduction in error and improved the results by up to 2% Considering the high sensitivity of shear wave velocity in geomechanical and reservoir characterization studies, even small improvements in its estimation can significantly reduce error in the characterization of hydrocarbon reservoirs.

Python Dash for Well Data Validation, Visualization, and Processing

Open-source Python libraries play a critical role in facilitating the digital transformation of the energy industry by enabling quick deployment of intelligent data-driven solutions. In this paper, we demonstrate an example of using Dash, a Python framework introduced by Plotly for creating interactive web applications. A fit-for-purpose software was tailored for an in-house research project in well-data validation, visualization, and processing. The application automates quality control of different sets of well-log data files (DLIS/LIS or LAS) for completeness, validity, and repeatability. For this tedious and critical process, a human expert is required to perform tasks using well-log interpretation software. A typical digital log file may contain hundreds or thousands of data channels that are difficult are difficult to visualize and validate manually. Sometimes it takes multiple iterations of communication between the data provider and the data receiver to achieve a final valid deliverable copy. By utilizing open-source Python libraries, such as DLISIO (Equinor ASA, 2022) and LASIO (Inverarity, 2023), a web interface based on Plotly-Dash is developed to visualize and check all data channels automatically and then produce a compliance summary report in PDF or HTML format. The time for validating one DLIS file that has hundreds of data channels is significantly reduced. Implementation of this automated data quality control workflow demonstrates that open-source Python libraries can significantly reduce the time from development to the deployment cycle. Quick implementation of intelligent software based on Python Plotly-Dash enables customized solutions or workflows that further improve both the effectiveness and efficiency of routine data quality control processes.

Reconstruction of Missing Well-Logs Using Facies-Informed Discrete Wavelet Transform and Time Series Regression

Summary Geophysical logging is widely used in lithofacies identification, reservoir parameter prediction, and geological modeling. However, it is common to have well-log sections with low-quality and/or missing segments. Repeating the well-log measurements is not only expensive but might also be impossible depending on the condition of the borehole walls. In these situations, reliable and accurate well-log prediction is, therefore, necessary in different stages of the geomodeling workflow. In this study, we propose a time series regression model to predict missing well-log data, incorporating facies information as an additional geological input and using discrete wavelet transform (DWT) to denoise the input data set. The main contributions of this work are threefold: (i) We jointly use facies information with well logs as the input data set; (ii) we use DWT to denoise the input data and consequently improve the signal-to-noise ratio of the input data; and (iii) we regard the depth domain as the time domain and use a time series regression algorithm for log reconstruction modeling. We show a real application example in two distinct scenarios. In the first, we predict missing well-log intervals. In the second, we predict complete well logs. The experimental results show the ability of the proposed prediction model to recover missing well-log data with high accuracy levels.

New interpretation approach of well logging data for evaluation of Kern aquifer in South California

Today, various methods are offered to effectively understand the subsurface layers. Meanwhile, the effective interpretation of different data can be effective in understanding the subsurface layers. In this research, a new strategy for understanding geological structures is proposed. Interpretation of geophysical data by considering geodynamic facts is a method that can improve the process of understanding subsurface structures. Geophysical measurement, especially well-logging data, gives us direct information from the subsurface layers. But geodynamics introduces us to the potential of natural resources such as water, oil, and mineral resources. In fact, geodynamics examines the possibility of the formation and existence of natural resources and examines the suitable structures for the exploration of natural resources. In this research, the Kern region of California state selected for investigation. Kern region is very interesting in terms of the existence of natural resources, but the issue of water shortage is investigating in this region. The well-logging information in this area was interpreted considering the tectonics and geodynamics of Kern and California in general, and the depth of the aquifers determined. This interpretation method is able to use in the exploration of various natural resources. Using different information sources makes it easier to interpret and understand subsurface structures.

Electro-facies classification based on core and well-log data

Facies studies represent a key element of reservoir characterization. In practice, this can be done by making use of core and petrophysical data. The high cost and difficulties of drilling and coring operations coupled with the time-intensive nature of core studies have led researchers toward using well-log data as an alternative. In the Teapot Dome Oilfield, where core data are limited to those from only a single well, we used well-log data for reservoir electro-facies (EF) studies via two unsupervised clustering methods, namely multi-resolution graph-based clustering (MRGC) and self-organizing map (SOM). Satisfactory results were obtained with both methods, distinguishing seven electro-facies from one another, where MRGC had the highest discriminatory accuracy. The best reservoir quality was exhibited by electro-facies 1, as per both methods. Our findings can be used to avoid some time-intensive steps of conventional reservoir characterization approaches and are useful for prospect modeling and well location proposal.

Sequence Stratigraphy and Implications for Shale Gas Exploration in the Southern Sichuan Basin, South China

In contrast to the widely used sequence stratigraphic models for passive continental margins, the stacking patterns of strata within epeiric seas, which are influenced by regional tectonic activity, may display opposing characteristics during the same geological period. These variations serve as a record of basin evolution and also affect the accumulation of hydrocarbons within the strata. Our study investigated the development potential of the deep Longmaxi Shale in the southern Sichuan Basin by examining the sequence stratigraphy and sedimentary fill patterns. Using a combination of core observation, well-logging data analysis, and 3D seismic profile interpretation, we aimed to gain an understanding of the sedimentary fill history of the Longmaxi Shale during the Early Silurian. Our analysis revealed that deglaciation and regional tectonic events affected the sequence stratigraphy, resulting in unconformities that were identifiable using seismic data and wireline logs. Through an analysis of thirty wireline logs and two seismic profiles, we identified two third-order sequences suggested in the Lower Longmaxi Formation. Within the two third-order sequences were five systems tracts, with the first exhibiting a complete cycle of sea-level change and the second cycle being incomplete due to regional tectonic events. The graptolite succession on the upper Yangtze Platform provided a temporal view of the sequence stratigraphy and sedimentation rates of the Longmaxi Shale. The thickness trends of the systems tracts reflected the interplay of short-term eustasy fluctuations, subsidence, and uplift. Our analysis suggests that regional subsidence played a significant role in the deposition of the second transgressive systems tract (TST) in the Weiyuan and Luzhou areas, which represents a promising target for shale gas exploration, in addition to the first TST. However, the Changning area experienced a relative sea-level decrease due to the intense uplift of the Qianzhogn Paleo-uplift and the increased supply of sediment and is interpreted as a highstand systems tract (HST); it is not considered to have shale gas exploration potential.

Machine learning approaches for formation matrix volume prediction from well logs: Insights and lessons learned

Determining the formation rock type and their petrophysical properties using well-log data is necessary for resource assessment. Formation porosity, shale content, and saturations must be estimated to quantify hydrocarbon reserves. Interpreting well logs could be challenging, especially when the formation is highly heterogeneous. Conventional physics-based interpretation can be tedious and time-consuming. Machine learning (ML) models can automate the well-log interpretation process, saving time and reducing the need for the expertise of trained engineers. In this research, we utilized five supervised ML algorithms incorporating multiple well-log parameters to identify patterns and behaviours in the well-log dataset. Our objective is to provide accurate, real-time predictions of sandstone, carbonate, and shale mineral volumes in the formation. We used the data from open-hole well logs of four wells in the Volve field and trained ML models using Random Forest, Gradient Boosting, Support Vector Regression, K-Nearest Neighbour, and Artificial Neural Networks. We used a robust framework that included random and grid search and cross-validation techniques to find the best hyperparameters and minimize bias and variance in the ML models. ML models were built by training and testing data from two wells and then validated on unseen data from the next two wells.All the ML models were able to predict mineral volume both on testing data as well as on the validation set. K-Nearest Neighbour, Random Forest, and Gradient Boosting outperform the other ML models. The overall prediction accuracy is similar for these three models, and the Random Forest model can be taken as the best model based on Taylor's plot. Random Forest can reduce the variance of the model to minimize the overfitting of decision trees. Gradient boosting builds trees sequentially and focuses on weak predictors to minimize a loss function and achieve better accuracy. We observed that all predictive ML models could easily quantify shale content but have difficulties differentiating sandstone and carbonate. The KNN model results are better than the other ML models for predicting carbonate volumes. One challenge when using ML techniques in the current state is that the accuracy of the model is limited to wells with similar drilling conditions as the training data, without significant changes in geological conditions. This means that a domain expert is needed to assist with predictive modelling using machine learning algorithms. Despite this challenge, our results show that using a machine-learning framework for reservoir characterization can be effective. However, ML models should not be viewed as a replacement for traditional interpretation methods but rather as an additional tool that can be used by engineers involved in petrophysical evaluations.

VeerNet: Using Deep Neural Networks for Curve Classification and Digitization of Raster Well-Log Images.

Raster logs are scanned representations of the analog data recorded in subsurface drilling. Geologists rely on these images to interpret well-log curves and deduce the physical properties of geological formations. Scanned images contain various artifacts, including hand-written texts, brightness variability, scan defects, etc. The manual effort involved in reading the data is substantial. To mitigate this, unsupervised computer vision techniques are employed to extract and interpret the curves digitally. Existing algorithms predominantly require manual intervention, resulting in slow processing times, and are erroneous. This research aims to address these challenges by proposing VeerNet, a deep neural network architecture designed to semantically segment the raster images from the background grid to classify and digitize (i.e., extracting the analytic formulation of the written curve) the well-log data. The proposed approach is based on a modified UNet-inspired architecture leveraging an attention-augmented read-process-write strategy to balance retaining key signals while dealing with the different input-output sizes. The reported results show that the proposed architecture efficiently classifies and digitizes the curves with an overall F1 score of 35% and Intersection over Union of 30%, achieving 97% recall and 0.11 Mean Absolute Error when compared with real data on binary segmentation of multiple curves. Finally, we analyzed VeerNet's ability in predicting Gamma-ray values, achieving a Pearson coefficient score of 0.62 when compared to measured data.