To address the high cost and low efficiency of lithology identification for intrusive rocks in oil and gas exploration, this study proposes a hierarchical identification method based on petrochemical big data and machine learning. By integrating global geochemical databases, a standardized sample set covering ultramafic to felsic intrusive rocks was constructed, and a three-level classification system (“group–subgroup–specific lithology”) was established. Visual discrimination charts and an automatic identification model were developed using Linear Discriminant Analysis and Multilayer Perceptron, respectively. The results show that the method achieves over 90% accuracy in the first- and second-level classifications, effectively identifying major rock types such as gabbro, diorite, and granite. Although the accuracy fluctuates in the third-level classification due to compositional overlap and data quality issues, the method exhibits good interpretability and generalizability. This approach provides a low-cost and efficient technical solution for rapid lithology identification and reservoir evaluation, demonstrating potential application in deep and unconventional hydrocarbon exploration.

How to cite: Quan, Z. and the Quan Zhou: Lithology identification of intrusive rocks based on petrochemical big data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6040, https://doi.org/10.5194/egusphere-egu26-6040, 2026.

Detecting fluid contacts, such as oil-water interfaces, remains a significant challenge for conventional seismic exploration when the acoustic impedance contrast is weak. However, the seismoelectric effect, which depends on electrical conductivity and electrokinetic coupling, provides a promising method for interface detection. In this study, we investigate the radiating seismoelectric electromagnetic (EM) reflection and transmission coefficients generated at an interface between two fluid-saturated porous media under fast P-wave incidence.

We analytically derive the reflection and transmission coefficients for all generated wave modes (fast P, slow P, SV, and EM waves) using Pride’s theory and Helmholtz decomposition. To verify the validity of our derivation, we calculate the energy fluxes of the wavefields using the electrokinetic Poynting vector and confirm that energy conservation is strictly satisfied.

Analyzing the frequency- and angle-dependent behavior of the reflection and transmission coefficients, we show that seismoelectric conversion is directly governed by the contrast in elastic and electric properties of the reservoir and the fluids therein, at the interface. Building on this, we compare the response of an "elastic interface" (strong mechanical contrast) with an "electric interface" (strong conductivity contrast but weak mechanical contrast, representing an oil-water contact). The analysis shows that while the seismic reflection energy at the electric interface is negligible, the reflected seismoelectric EM wave energy is comparable to the seismic signal generated at a strong lithological interface. These findings suggest that reflected seismoelectric waves are a reliable tool for identifying oil-water interfaces that are effectively invisible to conventional seismic surveys.

How to cite: Liu, Y. and Smeulders, D.: Seismoelectric Radiating Reflection and Transmission at Porous–Porous Interfaces: Angle- and Frequency-Dependent Coefficients, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17981, https://doi.org/10.5194/egusphere-egu26-17981, 2026.

Shale reservoirs, characterized by intricate fluid occurrence,elevated clay and organic matter (OM) contents, and diverse pore structures, present complexities that obscure the primary controlling factors of lacustrine shale conductivity and render existing saturation models insufficient in accuracy. Taking the medium-high maturity lacustrine shale of the Qing1 Member in the Changling Sag, Southern Songliao Basin as an example,  we use 2D nuclear magnetic resonance and pressure-maintained and sealed coring techniques to obtain the original fluid information of shale reservoirs. By combining geochemical, mineralogical, and geophysical properties, we investigate the impact of mineral components, physical properties, and fluid occurrence on the electrical conductivity properties of shale reservoirs. The clay and carbonate minerals play a primary role in  influencing the conductivity of shale, whereas effective porosity plays a secondary role. Conductivity variations within single lithologies are affected by total organic carbon and fluid types. A novel resistivity-saturation interpretation model, which is a function of saturation, for mature lacustrine shale (MLS) is developed based on lithologic distinctions and the influence of OM. This model identifies two primary conductive pathways: free water conduction in matrix pores and additional conduction from clay-bound water. The bound water cementation indexmwb and conductive bound water fraction Swb are introducedto reflect the impact of OM on clay additional conductivity. Compared with the Archie model developed for clay-free rockand the Indonesian model used for shales, the MLS model offers a more accurate calculation of oil saturation in MLS. Our approach makes a step ahead toward reducing uncertainty in  the evaluation of MLSs as potentially economic oil reservoirs. 

How to cite: Li, Z.: Conducting mechanism and saturation model of mature lacustrine shales: A case study of the first member of the Qingshankou Formation in the Changling Sag, Southern Songliao Basin , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1767, https://doi.org/10.5194/egusphere-egu26-1767, 2026.

Recent shale gas drilling wells in southern China confirms significant resource potential within the Lower Carboniferous marine shale of the Yaziluo Rift Trough, yet exploration efficiency is hindered by poorly understood sedimentological heterogeneity and gas accumulation mechanisms. This study integrates major and trace elements, high-resolution field emission scanning electron microscopy, and adsorption analyses (low-pressure N₂/CO₂, high-pressure CH₄) to systematically characterize sedimentary facies, reservoir properties, and gas enrichment patterns. The trough exhibits an asymmetric "platform-basin" model with three distinct sedimentary facies. Basin facies comprise siliceous shale formed in deep-water anoxic settings, yielding the highest total organic carbon (TOC) content (average 3.63%) controlled by redox conditions and paleoproductivity. Lower slope facies consist of mixed shale in dysoxic environments, with moderate TOC (average 1.80%), dominated by redox conditions. Upper slope facies are calcareous shale in shallow, weakly reducing settings, showing the lowest TOC (average 0.99%), influenced by clay-mediated organic preservation. Reservoir analysis reveals that basin facies are dominated by organic pore, whereas lower slope facies display reduced organic pores but increased inorganic pores and micro-fractures, and upper slope facies shift predominantly to inorganic pores and microfractures. Moving from basinward to upper slope, increasing carbonate content expands dissolution pore networks, yet declining TOC diminishes organic pore development, promoting organic-clay complexes and weakening pore structure. Additionally, three shale associations classified by shale-to-argillaceous limestone ratios correspond to specific sedimentary facies. The lower slope shale association demonstrates optimal gas preservation due to high TOC, argillaceous limestone interlayers acting as direct caps, and fracture-enhanced porosity facilitating gas migration. The upper slope association shows promise for self-sealing bodies via acid fracturing despite lower TOC. In contrast, the basin facies shale association exhibits constrained gas retention capacity owing to clay-dominated mineralogy and absence of argillaceous limestone interlayers. This study emphasizes the critical role of lithofacies heterogeneity and integrated "source-reservoir-seal" configurations in evaluating of shale gas accumulation under "slope-basin" depositional architectures, providing a theoretical basis for reservoir development in analogous geological settings.

How to cite: Chen, X.: Sedimentology Dominated Accumulation Mechanism of Marine Shale Gas, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2639, https://doi.org/10.5194/egusphere-egu26-2639, 2026.

With the continuous advancement of oil and gas exploration and development, unconventional resources have become a crucial component of national energy strategies. The extraction of resources such as shale oil and gas relies on technologies like horizontal drilling and multi-stage fracturing, where accurate geomechanical parameters are essential for engineering design. Conventional core-based experiments and well-log inversion methods, though reliable, are often costly, time-consuming, and limited in representativeness. Recent progress in digital cutting analysis—using scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS)—offers a fast, economical, and practical alternative. This study presents a workflow for predicting rock mechanical parameters from the mineral composition and pore structure of cuttings. Site-collected cuttings were characterized via SEM-EDS to analyze morphology, mineralogy, pore networks, and microfractures. Given the high heterogeneity and anisotropy of shale, a composite modeling approach integrating heterogeneous structure theory and equivalent medium models was applied. This included the Reuss and Voigt bounds, Voigt–Reuss–Hill average, Hashin–Shtrikman bounds, Kuster–Toksöz theory, and Gassmann fluid substitution to estimate equivalent static elastic parameters. These were then converted to dynamic parameters using linear regression to ensure consistency with logging data. Results show strong agreement between cutting-derived parameters and well-log inversions. Young’s modulus and Poisson’s ratio errors range from –12.62% to 4.03% and –10.18% to 10.47%, respectively, within acceptable limits. Although minor uncertainties arise from mineral identification and image segmentation, overall trends match well-log data closely. The introduction of a Weakness Index effectively highlights reservoir heterogeneity and correlates well with measured fracture pressures. A strong linear relationship between static and dynamic parameters, particularly Young’s modulus, supports the reliability of the regression-based conversion. This study confirms the feasibility and applicability of digital cuttings for building rock-physics models and predicting mechanical properties in unconventional reservoirs. The method not only aligns with well-log results but also better captures formation heterogeneity. More importantly, it enables real-time, cost-effective mechanical characterization from wellsite cuttings, offering a practical alternative to core- or log-dependent methods. This is particularly valuable in complex wells such as horizontals, where rapid formation evaluation and fracture design are critical. 

How to cite: Nie, X., Hou, Y., and Zhang, Z.: Rock Mechanical Parameters Prediction Based on Digital Drilling Cuttings Mineral Logging Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2877, https://doi.org/10.5194/egusphere-egu26-2877, 2026.

Marine shales of X-block are featured with poor organic matter, over maturity and complex mineralogical assemblages, which collectively result in weak organic-related logging responses. As a consequence, conventional total organic carbon (TOC) evaluation methods exhibit substantial uncertainties associated with baseline calibration and parameter generalization, thereby limiting prediction accuracy and robustness.

To overcome these limitations, this study develops a physics-constrained deep learning framework for TOC prediction that integrates multi-source logging data using a hybrid DNN–LSTM–GPC architecture. High-resolution nuclear magnetic resonance (NMR) and electrical imaging logs are incorporated as primary data sources to extract multi-scale, organic-matter-sensitive features. These features are further integrated with conventional well logs to construct a comprehensive feature space that captures organic matter distribution, pore structure characteristics, and lithological variability. In addition, an improved ΔLogR model and region-specific rock-physics constraints are embedded within the deep learning framework to ensure physical consistency and geological interpretability.

Application results demonstrate that the proposed method achieves superior prediction performance in low–organic-matter marine shales, yielding a root mean square error of 0.08% and a coefficient of determination (R²) of 0.95. The model consistently outperforms multivariate regression, uranium-based approaches, and porosity-difference methods, while maintaining stable predictive capability in intervals exhibiting pronounced TOC heterogeneity. These results indicate that physics-constrained deep learning integrated with multi-source logging data provides a reliable and effective approach for micro-scale TOC evaluation and favorable reservoir identification in low–organic-matter marine shale systems.

This research was supported by the Natural Science Foundation of Shandong Province of China (ZR2023YQ034) and Shida Jingwei Industry-Education Integration Research Institute Project (15572259-25-ZC0607-0011).

Fig. 1. Comparison of total organic carbon (TOC) prediction results for marine shale in Well JY6, X Block, obtained from different methods.

How to cite: Ge, X., Wang, Z., He, C., Ge, X., Fan, J., Wu, X., Yang, D., and Zhai, C.: A DNN–LSTM–GPC framework for TOC prediction in marine shales of the X Block using multi-source logging data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8697, https://doi.org/10.5194/egusphere-egu26-8697, 2026.

Due to the mutual flow between drilling mud and formation fluid, the pore pressure in the surrounding rock of the wellbore directly affects the rock’s mechanical properties. Since the properties and density of drilling mud are adjusted in real time during drilling operations, wellbore stability is inherently a dynamic process. This study aims to enhance wellbore integrity and reduce drilling accidents by investigating the dynamic characteristics of wellbore stability.

Rock mechanics experiments were conducted to simulate the mechanical properties of rocks under varying pore pressure conditions, yielding the quantitative relationship between pore pressure and rock mechanical behaviors. In conjunction with the radial seepage characteristics around the wellbore at different drilling stages, dynamic fluid-solid coupling simulations of the wellbore were performed based on geomechanical principles, enabling the development of a dynamic wellbore stability prediction method. This method further facilitates the prediction of key parameters such as wellbore collapse period and radial collapse depth.

This dynamic wellbore stability prediction technology was applied to three wells drilled in sandstone formations. It successfully predicted the collapse depth and period under different drilling mud densities and properties, significantly improving drilling efficiency while ensuring wellbore integrity. Compared with static wellbore stability prediction techniques, this technology provides drilling engineers with a richer set of drilling parameters and defines a clear wellbore collapse period, thereby effectively preventing stuck pipe accidents. Comparative drilling tests in the same block and formation layer showed that drilling efficiency increased by 12%, and the stuck pipe accident rate decreased by 27.3%.

This exploratory research demonstrates that wellbore stability in permeable formations during drilling is indeed a dynamic equilibrium process. As drilling mud properties, time, and stress conditions change, wellbore stability evolves accordingly. Predicting dynamic parameters can provide valuable references for drilling design and optimization, thereby enhancing wellbore integrity and minimizing drilling accidents.

How to cite: Liu, X. and Xiong, J.: Research on Dynamic Wellbore Stability Based on Wellbore Seepage, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4382, https://doi.org/10.5194/egusphere-egu26-4382, 2026.

Due to its favorable pore-permeability characteristics, sandstone serves a crucial role in applications such as oil and gas production, freshwater extraction, and underground CO₂ storage. As the primary reservoir space and migration pathway for hydrocarbons, accurate porosity data are essential for seismic exploration and reservoir development. In unconventional reservoirs, establishing an appropriate tight sandstone rock physics model is key to understanding how petrophysical parameters influence porosity and permeability. However, conventional models often fail to dequately represent both the three-dimensional irregularity of pore geometries and the spatial distribution of clay minerals.

To address these limitations, this study develops a novel rock physics model that incorporates superspherical pores and clay distribution to characterize argillaceous sandstone. Clay is categorized into structural clay (within the matrix) and dispersed clay (within pores). Following a sequential approach that reflects natural sandstone grain stacking, the proposed model is constructed by employing Voigt-Reuss-Hill averaging methods to incorporate structural clay, coupling the superspherical pores model, and applying solid substitution equations to determine the saturated rock modulus, which contains dispersed clay. This framework allows quantitative analysis of how pore geometry and clay occurrence states affect rock elastic properties.

However, in practical settings, the high cost of well logging often prevents direct measurement of these parameters. Therefore, a simulated annealing algorithm is employed to inversely determine pore characteristics and clay content in different occurrence states at each sampling point along the wellbore.

The validity and practical applicability of the proposed model are demonstrated through comprehensive sensitivity analyses and real-data applications.

How to cite: Wang, B., Yin, X., Ma, Z., and Li, K.: Development and Application of a Rock Physics Model to Infer Pore-Geometry and Clay-Distribution in Argillaceous Sandstone, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16220, https://doi.org/10.5194/egusphere-egu26-16220, 2026.

The effective development of unconventional shale oil reservoirs depends on the accurate prediction of seepage pathways. In the lacustrine shales of the upper submember of the fourth member of the Shahejie Formation (Es4U) in the Dongying Sag, bedding fractures serve as both primary storage spaces and critical fluid conduits. However, intense reservoir heterogeneity makes it challenging to reliably identify fracture-prone intervals using well-log responses alone.

Based on systematic core descriptions and fracture statistics from research wells, this study integrates thin-section petrography, X-ray diffraction (XRD), total organic carbon (TOC) analysis, and logging data to perform a multi-scale correlation of bedding fracture occurrences.

The results demonstrate that: (1) bedding fractures are diagenetic products triggered by clay mineral dehydration/transformation and hydrocarbon-induced overpressure; (2) the development of these fractures is synergistically governed by carbonate and clay mineral contents, organic matter abundance, and lamina density, with laminated clay-rich lithofacies identified as the dominant zones for fracture occurrence.Building upon these geological insights, a vertical development model for bedding fractures was established to provide quantitative geological constraints for logging evaluation. By utilizing conventional logging suites (e.g., integrating AC, GR, and DEN curves), intervals characterized by carbonate content of 20%–60%, clay content of 20%–35%, TOC > 2%, and well-developed laminated structures are identified as potential fracture-bearing zones.

This model effectively bridges geological genesis with logging attributes, providing a direct geological foundation for the prediction of favorable seepage pathways in shale reservoirs using borehole geophysical data.

How to cite: Liu, C., Wang, G., and Yin, Z.: Geological Patterns and Conventional Well-log Prediction of Bedding Fractures in Lacustrine Shales: A Case Study from the Upper Submember of the 4th Member of the Shahejie Formation, Dongying Sag, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10363, https://doi.org/10.5194/egusphere-egu26-10363, 2026.

     In deep coal seams, the primary storage space for adsorbed methane resides in the micropores and associated adsorption surfaces of the coal matrix, whereas the fracture network provides the dominant pathway for gas–water transport. Fractures promote the migration and local enrichment of free gas, with their connectivity exerting a fundamental control on methane desorption efficiency and overall seepage capacity. Therefore, elucidating the pore‑scale mechanisms of gas‑driven water displacement is critical for understanding gas–water occurrence and flow dynamics, as well as for assessing the reservoir‑controlling role of different microstructures.In light of the pronounced heterogeneity of coal, this study constructs two pore‑scale digital core models—a matrix‑pore model and a pore–fracture model—based on micro‑CT imaging of low‑rank coal. Utilizing the Volume of Fluid (VOF) method within an immiscible two‑phase displacement framework, we simulate methane‑driven water displacement under reservoir‑formation conditions. The simulation results are used to systematically analyze how the contact angle (θ) and capillary number (Ca) govern displacement morphology, phase connectivity, and residual phase distribution. Furthermore, a capillary‑number–contact‑angle (Ca–θ) phase diagram characterizing the displacement process is established.

       Key findings are summarized as follows: (1) Under gas–water viscosity ratios representative of the reservoir‑formation stage, pore‑scale gas‑driven water displacement exhibits three distinct regimes: capillary fingering, viscous fingering, and a transitional capillary–viscous fingering regime. In the matrix‑pore model, displacement is dominated by capillary fingering due to pore‑throat constrictions and microstructural bottlenecks, resulting in a dispersed and fragmented displacement front. In contrast, displacement in the pore–fracture model is governed by an interconnected fracture network, where capillary forces are substantially weakened, leading to displacement patterns characteristic of viscous fingering. (2) At low Ca, displacement follows the capillary‑fingering regime, and the gas phase predominantly forms a connected flow network after displacement. At high Ca, viscous fingering dominates, generating numerous isolated gas bubbles and yielding poor gas‑phase connectivity. At intermediate Ca, a transitional regime emerges, combining features of both capillary and viscous fingering. (3) The influences of wettability, capillary number, and pore‑structure type on final gas saturation and gas‑phase connectivity differ markedly between the two models. Under identical displacement conditions, the pore–fracture model attains significantly higher gas saturation and superior gas‑phase connectivity compared to the matrix‑pore model. These insights advance the understanding of methane occurrence and migration in deep coal seams and provide a basis for optimizing coal reservoir development strategies.

Keywords：Deep coal;Gas-water two-phase flow;Displacement pattern;Wettability, Capillary number

How to cite: Jin, X. and Chen, Z.: Pore‑Scale Simulation of Methane‑Water Two‑Phase Flow in Deep Coal Seams Using the Volume of Fluid Method, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2800, https://doi.org/10.5194/egusphere-egu26-2800, 2026.

Mean grain size (Mz) is a key indicator of depositional processes and reservoir quality, yet its continuous characterization is commonly limited by the availability of core or cuttings data. This study presents a new method for predicting Mz from conventional well logs by integrating Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and Principal Component Analysis (PCA). A density–neutron separation parameter is first constructed to capture lithological and grain-framework variations. Multi-scale components sensitive to grain-size changes are then extracted from this parameter using CEEMDAN, and the dominant controlling features are identified through PCA. The resulting model enables continuous Mz prediction along the wellbore. Comparisons with measured data demonstrate that the proposed approach reliably captures vertical grain-size variations, providing a practical and robust solution for quantitative grain-size characterization and supporting detailed reservoir analysis and geological modeling.

How to cite: Shen, B., Wang, C., Ma, X., and Sun, K.: Quantitative Inversion of Mean Grain Size from Conventional Well Logs Using Complete Ensemble Empirical Mode Decomposition （CEEMDAN）and Principal Component Analysis（PCA）, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3232, https://doi.org/10.5194/egusphere-egu26-3232, 2026.

The accurate quantification of karst volume is essential for evaluating storage capacity and fluid flow behavior in carbonate reservoirs, which are often highly heterogeneous due to complex diagenetic and karstification processes. However, conventional approaches to karst characterization frequently rely on qualitative descriptions or isolated datasets, lacking an integrated framework that effectively combines petrophysical properties with lithological controls. To address this gap, this study proposes a novel and practical workflow that systematically integrates routine well logs and core data to quantitatively estimate karst-dominated porosity in carbonate sequences. The methodology is designed to be both log-based and scalable, making it suitable for field-wide application even in data-constrained environments.

The proposed method is structured into four sequential and interpretative steps: (1) Data Integration and Quality Control: Multi-source data—including conventional well logs, core measurements, geological models, and lithology logs—are carefully depth-matched and subjected to rigorous quality checks to ensure consistency and reliability. Special attention is paid to correcting for borehole environmental effects and log normalization across multiple wells. (2) Definition of Karst Porosity Threshold: Based on systematic analysis of core-derived porosity-permeability cross-plots, a porosity value greater than 15% is established as a robust indicator for significant karst development. This threshold effectively distinguishes karst-related pore space from matrix porosity and is validated through thin-section and CT-scan observations. (3) Identification of Host Lithology: Lithofacies modeling and petrographic analysis are employed to confirm that karst features are predominantly hosted within dolomite intervals, highlighting the lithological control on karst distribution. Log-based facies classification is calibrated with core data to ensure accurate lithology discrimination in non-cored sections. (4) Calculation of Karst Volume Percentage: The proportion of karst volume is quantitatively computed as the ratio of high-porosity (>15%) dolomite volume to the total pore volume, which includes matrix pores, vugs, and fractures. This volumetric approach enables a more realistic representation of karst contribution to total porosity and reservoir performance.

This workflow was applied to a carbonate reservoir in the Pre-Caspian Basin, where it yielded a karst volume proportion of 11.1%. This result aligns closely with independent core statistics and regional geological understanding, validating the method’s accuracy and applicability. Sensitivity analyses were conducted on the porosity threshold, confirming that the selected 15% cutoff optimally balances discrimination capability and geological plausibility. The integrated approach not only enhances the reliability of karst assessment but also offers a scalable and reproducible tool for reservoir characterization. It supports improved geological modeling, reservoir performance prediction, and development planning in complex carbonate settings, ultimately contributing to more efficient hydrocarbon recovery. Future work will focus on extending the workflow to incorporate seismic attributes and dynamic production data for enhanced 3D karst modeling.

Keywords: Karst Volume; Carbonate; Well Logging; Porosity; Integrated Workflow

How to cite: Hou, J. and Yang, Y.: An Integrated Method for Quantifying Karst Volume in Carbonate Reservoirs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3739, https://doi.org/10.5194/egusphere-egu26-3739, 2026.

Pyrite is widely developed in shale formations with diverse occurrence morphologies and relatively low contents. However, it exerts a significant impact on the electrical properties of reservoirs and severely restricts the accuracy of reservoir evaluation. Based on computed tomography (CT) scanning technology, combined with mathematical morphology methods and finite element simulation techniques, this study focuses on the coupling relationship between pyrite occurrence states and reservoir resistivity, along with the quantitative calculation method of pyrite content. The results indicate that: (1) The influence of pyrite occurrence morphologies on the electrical conductivity of shale varies remarkably. Dispersed pyrite particles exert a weak interference on reservoir resistivity, and the rock-electrical relationship conforms to Archie's law. In contrast, massive and banded pyrite have a more prominent impact on resistivity, their conductive contribution exceeds that of pore water. This renders the traditional Archie's law inapplicable. Furthermore, under the condition of the same content, banded pyrite exerts the most significant influence on reservoir resistivity. (2) Three parameters, namely the Pyrite Resistivity Ratio (PRR), the Conductive Path Influence Characterization (VPYRZ), and the Pyrite Occurrence Index (PYI), are constructed to quantitatively describe the nonlinear influence of pyrite with different occurrence morphologies on reservoir resistivity. Combined with scanning electron microscopy (SEM) data, the PYI thresholds for different occurrence morphologies are determined as follows: dispersed pyrite (< 7.8623), massive pyrite (7.8632–16.986), and banded pyrite (>16.986). On this basis, a quantitative calculation model for pyrite content is established. (3) The verification results of actual well logs show that the calculated results of the proposed model are in high agreement with the formation element logging data, and the accuracy meets the practical requirements of reservoir evaluation.

How to cite: Liu, Y.: A Study on the Coupling Relationship Between Pyrite Occurrence Morphologies in Shale and Resistivity, and the Quantitative Calculation Method of Its Content Based on Numerical Simulation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2946, https://doi.org/10.5194/egusphere-egu26-2946, 2026.

Abstract: Buried hill reservoirs in the eastern South China Sea exhibit highly heterogeneous mechanical properties and complex fracture networks that exert significant control on reservoir productivity, fluid pathways, and wellbore stability. Their tectonic evolution, multistage deformation, and lithological diversity make traditional fracture prediction methods insufficient for supporting safe and efficient offshore drilling operations. To address these challenges, this study develops a new integrated framework that combines geomechanical analysis, finite-element numerical simulation, and multi-source geological and geophysical data to characterize fracture attributes and predict fracture behavior under present-day stress conditions. The workflow incorporates acoustic and imaging logging data, high-resolution seismic attributes, lithology-based mechanical property modeling, and 3D Mohr circle stress analysis. These datasets are used to construct a heterogeneous geomechanical model that captures the spatial variability of elastic and strength parameters across the buried hill. Numerical simulations are performed to evaluate the distribution of stress perturbations associated with structural relief and lithological layering, and to assess the likelihood of fracture initiation, propagation, and reactivation under different stress regimes. The results demonstrate pronounced variations in fracture intensity and failure potential both laterally and vertically. Zones near faulted structural highs exhibit locally elevated differential stress and shear strain concentration, leading to enhanced fracture connectivity and increased reactivation probability. In contrast, massive crystalline units and mechanically strong lithologies show limited deformation and lower fracture susceptibility. By integrating the simulated stress field with observed fracture indicators, the study identifies high-risk intervals with elevated risks of borehole collapse, drilling fluid loss, or induced fracturing, as well as fracture-favorable sweet spots that may enhance reservoir penetration and productivity. Furthermore, the framework provides quantitative guidance for optimizing well trajectories, selecting safe drilling windows, and improving well placement strategies in offshore buried hill settings. The results highlight the importance of incorporating geomechanical constraints into fracture characterization workflows to reduce drilling uncertainty and improve the reliability of fracture prediction models. This integrated analytical–numerical approach offers a robust and transferable methodology for evaluating fracture development in complex tectonic reservoirs. It provides practical insights into wellbore stability management, reservoir development optimization, and risk mitigation in the challenging geological environment of the eastern South China Sea. The proposed workflow can also be adapted to similar buried hill or basement reservoirs worldwide.

Keywords: fracture characterization; geomechanical modeling; finite-element simulation; buried hill reservoirs; in-situ stress analysis; south China Sea

How to cite: Deng, C., Tan, J., and Yin, L.: An integrated geomechanical–numerical simulation framework for fracture characterization and prediction in buried hills of the eastern south China Sea, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2650, https://doi.org/10.5194/egusphere-egu26-2650, 2026.

The fluid inclusion homogenization temperature method is the most widely used approach for reconstructing hydrocarbon accumulation history in deep to ultra-deep carbonate rocks. However, there are three major limitations: (1) The difficulty in restoring the eroded thickness of target strata, leading to uncertainty in the reconstruction of tectonic and burial history. (2) It is common to use the brine inclusion homogenization temperature as a substitute for hydrocarbon inclusion trapping temperature. However, it is often challenging to find coexisting hydrocarbon and brine inclusions within the same host mineral. (3) It is difficult to determine the trapping time based on the trapping temperature. As in multi-cycle superimposed basins, the same trapping temperature may correspond to multiple ages on the tectonic-burial history curve, resulting in non-unique solutions.

The application of laser U-Pb isotopic dating of carbonate minerals and clumped isotope thermometry has addressed these limitations and led to the development of a new method for reconstructing hydrocarbon accumulation history in deep to ultra-deep carbonate reservoirs. (1) Restoration of the eroded thickness of strata is achieved through multi-phase diagenetic mineral age-temperature constraints, effectively resolving the uncertainty in the tectonic-burial history reconstruction. (2) Direct measurement of hydrocarbon inclusion trapping temperature, overcoming the challenge of determining trapping temperature when no coexisting brine inclusions are present in the host mineral. (3) Direct measurement of hydrocarbon inclusion trapping time, addressing the issue of non-uniqueness in accumulation ages for multi-cycle superimposed basins.

This method has been applied to reconstruct the accumulation history of natural gas reservoirs in the Dengying Formation of the Sichuan Basin. It has significantly higher accuracy and success rates compared to the fluid inclusion homogenization temperature method.

How to cite: Anping, H. and Anjiang, S.: A New Method for Reconstructing Hydrocarbon Accumulation History of Deep to Ultra-Deep Carbonate Reservoirs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8804, https://doi.org/10.5194/egusphere-egu26-8804, 2026.

Unconventional reservoirs are confronted with numerous challenges due to their complex pore structures and diverse fluid occurrence states; therefore, the accurate acquisition of pore structure characteristics and fluid property parameters has become the key factor restricting development efficiency. 2D nuclear magnetic resonance (NMR) logging obtains more observation information by introducing longitudinal relaxation time (T₁), and it exhibits good application effects in wettability evaluation, reservoir parameter calculation and fluid property identification .The factors affecting NMR relaxation signals are multivariate, mainly including pore fluid properties (viscosity, density, etc.), pore structure characteristics (pore size and distribution, pore connectivity, etc.), magnetic susceptibility differences between rocks and fluids, and the content and distribution of paramagnetic minerals. The coupling of these factors complicates the mapping relationship between relaxation signals and reservoir properties, which urgently requires in-depth analysis.In the field of shale oil exploration and development, there are various types of 2D NMR instruments, and significant differences exist in their observation methods, measurement sequences and acquisition parameters, which further pose challenges to the cross-scale characterization and comprehensive interpretation of reservoir parameters. Therefore, this study analyzes the influencing factors of acquisition parameters on NMR from both experimental and numerical simulation scales.The influence of instrument frequency on 2D NMR is investigated through experiments, and the effects of echo spacing, waiting time and number of echoes on NMR measurement and inversion results are studied via numerical simulation. This research further clarifies the influence of acquisition parameters on fluid distribution characterized by 2D NMR, and verifies the reliability of NMR logging interpretation data in shale reservoirs.

How to cite: Zhao, J., Xie, B., Wang, Y., and Bai, L.: Research on the Influence of NMR Acquisition Parameters on Fluid Distribution in Shale Reservoirs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15411, https://doi.org/10.5194/egusphere-egu26-15411, 2026.

Tight sandstone reservoirs are generally characterized by low porosity, low permeability, and complex pore structures, which pose significant challenges to reservoir evaluation. Nuclear Magnetic Resonance (NMR) T₂ spectra can characterize pore structures and fluid states in porous media. Therefore, gas-displacing-water petrophysical experiments were carried out to investigate the variations of T₂ spectra under different saturation states.The study finds that the T₂ spectrum in the original state presents a multi-peak distribution dominated by long relaxation components. With the increase of gas saturation, the macropore peak value decreases and the porosity tends to decrease. Meanwhile, compared with the gas-bearing state, the macropore peak widens when the core is water-saturated. Analysis shows that this phenomenon is caused by the low hydrogen index of gas.Thus, a spectrum correction model for gas-bearing sandstone reservoirs was established based on the Gaussian distribution, which corrects the T₂ spectrum morphology and porosity components to their original states. After gas saturation correction, the T₂ spectrum mainly presents a bimodal distribution dominated by short relaxation components. The irreducible water saturation calculated from the corrected T₂ spectrum is more consistent with the core measurement results.Combined with NMR experiments, this study clarifies the NMR response mechanism of reservoirs under different saturation states, establishes a Gaussian distribution model for T₂ spectrum correction of gas-bearing sandstone, and ultimately achieves accurate characterization of pore structures.

How to cite: Xie, B., zhang, B., tang, Q., zhu, X., zhao, J., wang, Y., bai, L., and lai, Q.: A Method for Characterizing Pore Structure of Gas-Bearing Sandstone Based on Nuclear Magnetic Resonance, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15708, https://doi.org/10.5194/egusphere-egu26-15708, 2026.

Acoustic Remote Detection Logging can extend the investigation range of conventional borehole logging from the immediate vicinity of the borehole to several meters and even tens of meters, providing complementary information for formation evaluation in reservoirs with heterogeneity such as fracture–cavity systems and thin interbeds. Such complex anomalous bodies may reduce the effectiveness of conventional rock-physics log interpretation. Overall, the current work still faces two main issues: the lack of geology-constrained 3D models and the lack of more efficient and transferable interpretation tools, which makes it difficult to establish a stable relationship between complicated reflected wavefields and subsurface heterogeneity.

This study focuses on Acoustic Remote Detection Logging and develops geology-constrained 3D heterogeneous modeling and response analysis. First, a 3D stochastic medium is generated using the spectral synthesis method to produce correlated random fields, which are then mapped to elastic-parameter perturbation fields. Then, a bisection-based thresholding scheme is applied to satisfy a target volume fraction (or porosity), and local-maximum detection is used to determine seed points and spatial distributions for cavities, enabling 3D construction of fracture–cavity heterogeneity. Cavity geometry and scale parameters are constrained by statistical characteristics derived from existing geological and logging data. Finally, CUDA-based parallelization is introduced to improve the efficiency of full 3D model generation and support rapid construction of high-resolution reservoir models.

Based on forward simulations of Acoustic Remote Detection Logging responses excited by dipole shear-wave sources, we design parameter combinations covering formation porosity and fracture–cavity geometric parameters, and summarize the results into response charts/templates for interpretation. These charts provide quantitative relationships between fracture parameters and key waveform features, supporting comparative identification and interpretation of fracture development, orientation, and spatial position under different geological scenarios. In addition, we explore using a model for feature discrimination under chart-based constraints, thereby providing auxiliary support for comparative evaluation and interpretation of fracture parameters.

How to cite: Wang, H., Guo, Y., Zhang, Z., Ye, Z., and Sun, H.: Acoustic Remote Detection Logging for Heterogeneous Reservoirs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4177, https://doi.org/10.5194/egusphere-egu26-4177, 2026.

Carbonate reservoirs possess substantial hydrocarbon resource potential and have become a major focus of exploration in recent years. However, the complex geological settings, pronounced lithological heterogeneity, and strong vertical variability of carbonate reservoirs pose significant challenges to conventional lithofacies identification methods. Such approaches are often time-consuming, highly subjective, and limited in their ability to accurately discriminate complex lithofacies assemblages, thereby constraining the efficient development of carbonate reservoirs. Geophysical well logging offers advantages such as low acquisition cost, continuous coverage, and high vertical resolution, making it a fundamental dataset for continuous lithofacies characterization. In this study, integrated geological information, including whole-rock analysis, petrographic thin sections, and scanning electron microscopy, is employed to systematically investigate the mineral compositions and pore structure characteristics of different lithofacies, and the corresponding logging response mechanisms are quantitatively investigated. Five conventional logging curves—gamma ray, acoustic, neutron porosity, bulk density, and resistivity—are selected to construct a multidimensional feature parameter set. Considering the differences in numerical ranges among logging curves and the imbalance in lithofacies sample distributions, data normalization and class imbalance correction are performed prior to model training. Subsequently, a high-precision lithofacies identification model based on conventional well logs is developed by integrating Bayesian optimization with the Extreme Gradient Boosting (XGBoost) algorithm. The results indicate that the predicted lithofacies are highly consistent with geological interpretations, core observations, and thin-section identifications, demonstrating that the Bayesian-optimized XGBoost model exhibits robust classification performance and effectively captures the complex nonlinear relationships between lithofacies and logging responses. Compared with traditional machine learning methods, the proposed approach achieves significantly improved identification accuracy, providing robust technical support and a reliable theoretical foundation for carbonate reservoir evaluation and hydrocarbon exploration.

How to cite: Yang, J., Wang, L., Gu, M., Ai, Y., and Li, G.: Bayesian-Optimized XGBoost for High-Precision Lithofacies Identification in Carbonate Reservoirs Using Geophysical Well Logs, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11545, https://doi.org/10.5194/egusphere-egu26-11545, 2026.

Tight sandstone reservoirs are generally characterized by low porosity and permeability, complex pore structures, and ambiguous electrical responses, which significantly limit the applicability of conventional water saturation evaluation models. To address these challenges, this study proposes a physically constrained reinforcement learning–based symbolic regression framework that integrates nuclear magnetic resonance (NMR)–derived pore size distribution information to automatically derive a dynamic Archie water saturation model with explicit physical interpretability.In the proposed approach, pore size distribution features are embedded into the formation factor formulation, enabling a dynamic correction of the classical Archie equation. A policy neural network combined with reinforcement learning is employed to jointly optimize the model structure and parameters, while explicitly enforcing physical constraints such as monotonicity and nonlinear electrical response behavior.Experimental results demonstrate that, compared with the conventional Archie model, the proposed dynamic model achieves a significant improvement in water saturation prediction accuracy for tight sandstone reservoirs, reducing the mean absolute error by approximately 11%. Moreover, the model more effectively captures the influence of pore structure heterogeneity on the relationship between electrical resistivity and water saturation.This study provides a novel water saturation evaluation methodology that combines physical interpretability with data-driven adaptability for tight sandstone reservoirs and offers valuable insights into the intelligent construction of logging interpretation models for complex reservoirs.

How to cite: Liao, W. and Zhao, B.: A Physically Constrained Dynamically Corrected Archie Saturation Model Based on Pore Size Distribution and Its Application to Tight Sandstone Reservoir Evaluation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4245, https://doi.org/10.5194/egusphere-egu26-4245, 2026.

Continental shale oil reservoirs in China are abundant in Solid Organic Matter (SOM), which represents a significant potential hydrocarbon resource. Nuclear Magnetic Resonance (NMR) is an effective tool for evaluating shale oil reservoirs. However, restricted by instrument dead time, it is difficult for the conventional CPMG acquisition mode to capture signals from ultra-short relaxation components such as SOM.To address this, this work introduces the Free Induction Decay (FID) acquisition mode, proposing two novel quantification workflows based on FID signal integration and a direct T₁-T₂^* spectrum, respectively. These approaches are designed to effectively capture the ultra-short components that are typically missed by standard CPMG sequences.The proposed methodology was validated against comprehensive geochemical benchmarks. The results demonstrate that the FID-driven approach yields a superior correlation with geochemical data compared to conventional methods. It exhibits high statistical consistency with Step-by-Step (SBS) Rock-Eval pyrolysis data, proving its capability to accurately quantify total solid hydrogen content, including solid petroleum hydrocarbons, bitumen, and kerogen. This work provides a non-destructive and high-precision means for evaluating the resource potential of shale oil reservoirs.

How to cite: Xu, C., Xie, R., Guo, J., Wang, X., and Zhang, G.: A novel method for quantitative characterization of organic matter based on NMR FID acquisition mode, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21370, https://doi.org/10.5194/egusphere-egu26-21370, 2026.

Micro-resistivity imaging logging is one of the primary techniques for subsurface fracture identification. However, conventional manual interpretation is time-consuming and highly subjective, while existing deep learning-based methods generally require large-scale, well-annotated datasets, resulting in substantial labeling costs and limited applicability in data-scarce scenarios. To address these challenges, this study proposes a fracture identification method for electrical image logs under limited-sample conditions by incorporating prior knowledge derived from outcrop fractures. Leveraging the morphological and statistical similarities between surface outcrop fractures and subsurface electrical image responses, a two-stage training strategy based on a Multi-scale Attention Network (MANet) backbone is developed. In the first stage, optical images of outcrop fractures are preprocessed through grayscale transformation and noise injection to approximate the feature distribution of electrical image logs, enabling the network to learn generalizable edge and texture features. In the second stage, outcrop–electrical image pairs with similar fracture morphology and texture characteristics are generated through similarity matching, and the model is fine-tuned using a composite loss function incorporating Correlation Alignment (CORAL), thereby accelerating domain adaptation to subsurface logging environments. Experimental results from basement reservoirs in the Dongping area of the Qaidam Basin demonstrate that the proposed method significantly improves fracture identification performance under limited-sample conditions. Compared with baseline models, the proposed approach achieves improvements of 13.05% in accuracy and 12.88% in Intersection over Union (IoU), reaching 81.13% and 75.74%, respectively. These results indicate that the proposed method effectively alleviates data scarcity issues in electrical image log interpretation and provides robust technical support for fracture characterization and hydrocarbon resource evaluation.

How to cite: Ai, Y., Wang, L., Gu, M., Gang, L., Shi, P., and Zhao, Z.: Fracture Identification in Electrical Image Logs with Limited Samples by Incorporating Outcrop Priors, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8628, https://doi.org/10.5194/egusphere-egu26-8628, 2026.

The construction of three-dimensional multi-mineral digital rock cores is essential for the fine characterization of reservoir pore–throat structures and for the quantitative evaluation of reservoir electrical, acoustic, and mechanical properties.Existing digital rock core modeling approaches can be broadly classified into physical experimental methods and numerical reconstruction methods.Physical experimental methods can produce relatively realistic three-dimensional digital rock cores; however, they are costly and struggle to achieve fine mineral discrimination.Numerical reconstruction methods offer advantages such as low cost and high efficiency; however, most high-fidelity approaches remain limited to single-mineral digital rock cores, whereas multi-mineral modeling methods often rely on idealized assumptions.To address these limitations, this study proposes a 2D–3DGAN-based deep learning algorithm capable of generating three-dimensional multi-mineral digital rock cores from a single AMICS image.Using the “AMICS + 2D–3DGAN” modeling framework, three-dimensional multi-mineral digital rock cores are constructed with high accuracy, efficiency, and realistic multi-mineral representation.The accuracy of the generated results is systematically evaluated by analyzing diagenetic characteristics and by comparing the generated cores with training images in terms of mineral content, pore size distribution, and two-point correlation functions.The results demonstrate that the proposed method significantly enhances reconstruction accuracy and generation efficiency while maintaining economic feasibility, thereby providing a solid foundation for subsequent simulations of multi-mineral diagenetic evolution and reservoir property analysis.Previous studies on diagenesis have largely relied on qualitative approaches, such as identifying diagenetic evolution sequences and constructing two-dimensional schematic representations.Based on the generated three-dimensional multi-mineral digital rock cores, this study proposes a “nucleation–growth” algorithm for numerically simulating diagenetic processes, enabling quantitative modeling of diverse cementation morphologies that closely reflect geological conditions.Meanwhile, the four-parameter structure method is improved to quantitatively simulate dissolution and replacement processes with varying morphologies and to control the precipitation location of dissolution by-products, either inside or outside dissolution pores.Ultimately, a three-dimensional digital rock core diagenetic evolution model based on diagenetic sequences is established, enabling analysis of the evolution of reservoir properties—such as porosity, permeability, tortuosity, and coordination number—through diagenetic processes.

How to cite: Lu, Z., Xi, K., and Wu, Y.: Deep Learning–Based 3D Multi-Mineral Digital Rock Modeling and Diagenetic Simulation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7042, https://doi.org/10.5194/egusphere-egu26-7042, 2026.

Accurate evaluation of pore systems in shale reservoirs is critical for understanding fluid flow, gas storage capacity, and overall reservoir performance. Extensive research has been conducted on the micro- and nano-pore structure of shale reservoirs using various experimental methods and shale samples. However, discrepancies in sample preparation standards and experimental parameters—such as observation techniques and data interpretation—raise concerns about the reliability and comparability of these results across different shale reservoirs. This review systematically evaluates the current methods for characterizing shale pore systems, with particular emphasis on commonly used techniques such as scanning electron microscopy (SEM), gas adsorption, and CT scanning. Key challenges related to sample preparation (e.g., sample size) and experimental conditions (e.g., voltage and current) are discussed, as these factors can introduce significant inaccuracies into pore structure characterization, even for well-established methods. Additionally, we examine the difficulties in integrating these methods to achieve a comprehensive understanding of shale pore parameters, including the quantification of organic and inorganic porosity, full-scale pore size distribution, and pore connectivity. Addressing these challenges requires the establishment of standardized processing workflows to enhance the comparability of results and minimize experimental errors. We also highlight often-overlooked issues, such as the potential discrepancy between pore structures observed under laboratory conditions and those at in-situ depths. The review concludes with recommendations for future research, including the development of advanced experimental techniques and more efficient data processing strategies to improve pore characterization in shale reservoirs. This review provides a new perspective for future research and addresses a critical gap in understanding the impact of experimental conditions on pore structure characterization outcomes.

How to cite: Gou, Q., Xu, S., Xiong, Q., and Li, Z.: A multiple review on identification and fine evaluation of shale pores: Prospects and challenges, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2531, https://doi.org/10.5194/egusphere-egu26-2531, 2026.

The accurate petrophysical characterization of unconventional reservoirs, particularly deeply buried fractured–vuggy carbonate systems, remains challenging due to extreme heterogeneity, ultra-low permeability, and strong scale dependency of reservoir properties. Conventional formation evaluation workflows often fail to reconcile geological observations at the core scale with log- and seismic-scale petrophysical interpretations in a physically consistent manner.In this study, we propose an integrated geological–geophysical petrophysical characterization workflow that links core and thin-section observations, electrofacies classification, and seismic data through an electrofacies-constrained seismic waveform-guided inversion framework. Rock types identified from core and petrographic analyses are translated into electrofacies at the log scale to represent pore-structure variability. Electrofacies are incorporated as conditional constraints in a Bayesian seismic waveform-guided porosity inversion framework. Electrofacies-dependent porosity priors derived from well-log statistics restrict the admissible pore-structure space, while seismic waveform similarity controls the lateral propagation of high-frequency porosity features, implicitly embedding seismic facies within the inversion.The resulting porosity volume exhibits enhanced vertical resolution and improved lateral continuity, allowing thin-layer and non-layered heterogeneities to be resolved beyond the limitations of conventional impedance-based inversion. Recognizing that permeability is not directly observable at the seismic scale, permeability is subsequently derived from the inverted porosity volume under electrofacies control, ensuring that pore connectivity and flow characteristics are consistently represented at the appropriate scale.This workflow establishes a causally consistent and scalable solution for advanced petrophysical characterization and formation evaluation of heterogeneous unconventional reservoirs. By integrating geological constraints with seismic waveform-driven inversion, the proposed method effectively bridges core-, log-, and seismic-scale information and demonstrates strong potential for application in complex carbonate systems as well as other unconventional plays.

How to cite: Zhao, A., Tian, F., and Cao, W.: Multi-Scale Petrophysical Modeling and Characterization of Heterogeneous Carbonate Reservoirs Based on Facies-Constrained Seismic Waveform Integration, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6054, https://doi.org/10.5194/egusphere-egu26-6054, 2026.