Energy Geoscience最新文献

{"title":"Experimental techniques for qualitative and quantitative characterization of shale oil reservoir space, minerals, and fluids","authors":"Siyu Chen ,&nbsp;Zhenkai Huang ,&nbsp;Guangyou Zhu ,&nbsp;Xi Li ,&nbsp;Zhiyong Chen ,&nbsp;Ruilin Wang ,&nbsp;Yue Huang ,&nbsp;Shengnan Liu ,&nbsp;Bijun Huang","doi":"10.1016/j.engeos.2025.100516","DOIUrl":"10.1016/j.engeos.2025.100516","url":null,"abstract":"<div><div>Characterizing shale oil reservoirs encompassing pore space, mineralogy, and fluids is fundamental to effective exploration and development. Recent advances in experimental techniques have significantly improved both the qualitative and quantitative analysis of these components; however, a comprehensive systematic review is lacking. High-resolution imaging technologies, such as Scanning Electron Microscopy (SEM), Field Emission Scanning Electron Microscopy (FE-SEM), and Focused Ion Beam Scanning Electron Microscopy (FIB-SEM), enable detailed visualization of pore structures. Gas adsorption and high-pressure mercury intrusion methods provide accurate pore-scale quantification. Moreover, techniques like X-ray Diffraction (XRD), X-ray Fluorescence Spectroscopy (XRF), and Electron Probe Microanalysis (EPMA) allow precise mineral identification and compositional analysis. Confocal Scanning Laser Microscopy (CSLM), Raman Spectroscopy, Nuclear Magnetic Resonance (NMR), and Rock Pyrolysis provide insights into fluid occurrence and content within shale reservoirs. Based on a comprehensive review of existing research, this study identifies several key future directions: (1) addressing the challenges of nanopore observation in reservoir space characterization while minimizing the impact of sample preparation on pore structure; (2) improving the accuracy of quantitative mineral analysis and developing advanced new technologies for the precise measurement of complex mineral compositions; (3) enhancing the fluid quantitative evaluation of fluids by more effectively restoring subsurface geological conditions. This paper presents a current synthesis and forward-looking perspective on experimental techniques supporting shale oil exploration, aiming to guide future research and technological innovation in this field.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100516"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145814456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Granite thermal conductivity based on mineral and chemical compositions: An evaluation based on hybrid models and implications on geothermal resource","authors":"Meijuan Liu ,&nbsp;Wei Zhang ,&nbsp;Lixia Xu ,&nbsp;Yuzhong Liao","doi":"10.1016/j.engeos.2025.100465","DOIUrl":"10.1016/j.engeos.2025.100465","url":null,"abstract":"<div><div>Rock thermal conductivity (TC) is a critical parameter in geothermal energy exploration and geological engineering, and is strongly influenced by the mineralogical and geochemical composition of rocks. In this study, granite from the Huangshadong geothermal field in Huizhou, Guangdong Province, was selected as the primary research subject. Combined with supplementary data from other regions, this work systematically investigates the quantitative relationships between granite thermal conductivity and both its mineral and chemical compositions. Moreover, the performance and applicability of four commonly used hybrid models for predicting rock thermal conductivity were comprehensively evaluated. The results reveal a strong positive correlation between granite thermal conductivity and quartz content, and a significant negative correlation with biotite content. Chemically, thermal conductivity exhibits a notablely positive correlation with SiO₂ content, and negative correlations with Al₂O₃, CaO, MgO, and TiO₂ contents. These findings suggest that evaluating rock thermal conductivity from a geochemical perspective offers a more nuanced understanding than mineralogical analysis alone. Among the evaluated models, the Harmonic Mean model demonstrated the best predictive performance for granite thermal conductivity, with an average mean error (<em>AME</em>) of 9.06 and a root mean square error (<em>RMSE</em>) of 0.34. However, the coefficient of determination (<em>R</em>²) for all four models remained in the range of 0.43–0.44, indicating moderate predictive accuracy. This limitation underscores the significant influence of rock structural characteristics, which are not fully captured by composition-based models. Future research should focus on the enhancement of predictive accuracy of granite thermal conductivity models by integrating both compositional and structural parameters. This study provides valuable insights into the microscale mechanisms governing granite thermal conductivity, contributing to a deeper scientific foundation for geothermal resource assessment and development.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100465"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147612152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-source hydrocarbon accumulation in the faulted and deeply depressed Tangdong area, Qikou Sag, Bohai Bay Basin","authors":"Xiugang Pu ,&nbsp;Guochao Xia ,&nbsp;Jiangchang Dong ,&nbsp;Zhannan Shi ,&nbsp;Wenzhong Han ,&nbsp;Qianru Shi ,&nbsp;Wei Zhang ,&nbsp;Yueqi Dong ,&nbsp;Xiongying Dong","doi":"10.1016/j.engeos.2026.100522","DOIUrl":"10.1016/j.engeos.2026.100522","url":null,"abstract":"<div><div>Based on organic geochemical analysis data, this study examined the characteristics of multi-source hydrocarbon accumulation in the Tangdong area of Qikou Depression, Bohai Bay Basin. The results indicate that the Tangdong area contains four source rock sequences: the third member of the Dongying Formation (E<em>d</em>³), the middle and lower sub-members of the first and third member of Shahejie Formation (E<em>s</em>^1M, E<em>s</em>^1L, E<em>s</em>³). Both the salinity of sedimentary water bodies and the contribution of bacteria in the four source rock sequences gradual increase in the order of E<em>s</em>³, E<em>s</em>^1M, E<em>s</em>^1L and E<em>d</em>³, as indicated by the Ga/C₃₀H, <em>ETR</em>, sterane/C₃₀ hopane, C₂₄TeT/C₂₆TT, C₂₃TT/C₃₀H, and the carbon isotope data of group components. Additionally, certain quantities of 4-methylsterane have been detected in source rocks in the E<em>s</em>³, E<em>s</em>^1M, and E<em>d</em>³, suggesting the special contribution of dinoflagellates to these source rocks. Based on biomarker parameters and carbon isotope data, the crude oil can be categorized into classes A to D, which show complex multi-source accumulation characterized by different sources in the same well and even in the same layer. Class A crude oil occurring in the higher part of the E<em>d</em>³ originates from source rocks in the E<em>s</em>^1L and E<em>s</em>³. Class B oil existing in the lower part of the E<em>d</em>³ is derived from source rocks in the Ed³. Class C oil found in the E<em>s</em>^1U is sourced from source rocks in the E<em>s</em>^1M. Class D oil present in the E<em>s</em>^1L is contributed indigenously by the source rocks in the E<em>s</em>^1L, establishing this unit as a lithologic hydrocarbon reservoir with source rocks and reservoirs in the same layer. Under the guidance of the understanding of multi-source hydrocarbon accumulation, currently, three wells have been drilled, all yielding satisfactory outcomes.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100522"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146090385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the impact of AI technology innovation on energy intensity: The role of ICT development and renewable energy integration","authors":"Emmanuel Baffour Gyau ,&nbsp;Yaya Li ,&nbsp;Michael Appiah ,&nbsp;Bright Akwasi Gyamfi ,&nbsp;Sakiru Adebola Solarin","doi":"10.1016/j.engeos.2026.100518","DOIUrl":"10.1016/j.engeos.2026.100518","url":null,"abstract":"<div><div>As global energy demand continues to intensify, artificial intelligence technology innovation (AITI) offers emerging possibilities for improving energy efficiency and supporting sustainable energy transitions. At the same time, the integration of renewable energy is reshaping energy systems and influencing energy consumption patterns. This study examines the relationship between AITI and energy intensity using panel data for 42 countries over the period of 2000–2022. Analysis with the Driscoll and Kraay standard errors fixed effects regression method reveals that AITI significantly reduces energy intensity; however, the evidence of a U-shaped relationship indicates that the reducing effect of AITI may reverse beyond a threshold. Economic growth, trade openness, financial development, and government regulations reduce energy intensity while foreign investment and skilled labor increase it. Mechanism tests reveal that AITI reduces energy intensity by promoting information and communication technology development, and renewable energy enhances the reducing effect of AITI. Heterogeneity analyses reveal that the reducing impact of AITI on energy intensity is significant in developed economies but insignificant in developing nations. These findings provide empirical insight for policymakers, emphasizing the need to align AITI with renewable energy transitions through tailored strategies that reflect economic diversity and regional conditions.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100518"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146090473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating calcite as a catalyst for pyrite-hydrogen reactions and hydrogen sulfide generation","authors":"Muhammad Hammad Malik ,&nbsp;Ahmed Al-Yaseri ,&nbsp;Khalid Al-Ramadan","doi":"10.1016/j.engeos.2026.100525","DOIUrl":"10.1016/j.engeos.2026.100525","url":null,"abstract":"<div><div>In nature, the presence of accessory minerals found in the subsurface have a potential to react with stored hydrogen in process of underground hydrogen storage. Pyrite (FeS₂), the most abundant sulfide mineral on earth and occurring in almost all geological rock formations specially the sedimentary rocks, is an ideal candidate for such reactions. Pyrite is also found in seal rocks in higher quantities and considered a critical mineral while working with subsurface hydrocarbon reservoirs. Although relatively stable at surface conditions, pyrite is still prone to oxidation in the presence of moisture and high temperatures found in subsurface reservoirs, leading to the release of sulfur. This sulfur can readily react with other elements, and in context of hydrogen storage in sedimentary rocks, it can react with hydrogen to form hydrogen sulfide (H₂S) gas. H₂S being extremely toxic, pose significant handling dangers along with potential degradation of the stored hydrogen fuel. Therefore, the presence of reactive and toxic material like H₂S require careful consideration when evaluating subsurface hydrocarbon reservoirs for hydrogen storage. In this study, we examined the reaction of pyrite combined with varying concentrations of calcite (1 %, 5 %, and 50 %), brine water (3.5 wt% NaCl), and hydrogen gas (H₂) under different reservoir conditions (120 °C and 250 °C, and 500 psi). The findings revealed that approximately 12 % H₂S was released, irrespective of the calcite concentrations, and the pyrite did not transform into its secondary mineral, pyrrhotite. These findings highlight the need of paying careful attention to the geochemical interactions within potential hydrogen storage sites to ensure safety and efficacy.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100525"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Method for determining the reasonable wellhead pressure of gas storage production wells: A case study of the Liuzhuang gas storage reservoir, China","authors":"Yuangang Wang ,&nbsp;Yidan Wang ,&nbsp;Lilin Fan ,&nbsp;Chuanzhi Cui ,&nbsp;Jingming Zhao ,&nbsp;Bin Huang ,&nbsp;Zhongwei Wu","doi":"10.1016/j.engeos.2026.100535","DOIUrl":"10.1016/j.engeos.2026.100535","url":null,"abstract":"<div><div>For the Liuzhuang gas storage reservoir with edge and bottom water, excessively high bottomhole pressure of production wells can cause edge and bottom water breakthrough, which reduces the effective storage capacity; whereas excessively low bottomhole pressure makes it difficult to achieve the gas production targets. Therefore, determining the reasonable wellhead pressure of production wells is of great significance for the efficient operation of gas storage reservoir. In this study, a comparison of seven critical liquid-carrying flow rate models, including the droplet inversion model and the liquid film inversion model, was made. The Belfroid critical liquid-carrying model was identified as the most accurate for evaluating liquid loading in the Liuzhuang gas storage reservoir. On this basis and in combination with nodal analysis, the relationship curves between gas production rate and critical liquid-carrying flow rate under different reservoir permeabilities and skin factors were obtained, leading to the development of wellhead pressure charts indicating the risk of liquid loading in gas wells of the Liuzhuang gas storage reservoir. This study provides technical support for field operations in diagnosing liquid loading in production wells.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100535"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thermoelastic full-waveform inversion for joint estimation of subsurface temperature and seismic velocities in geothermal exploration","authors":"Sen Yang ,&nbsp;Benbing Ren ,&nbsp;Xinmin Wu ,&nbsp;Hao Xu ,&nbsp;Tao Yang ,&nbsp;Yingkai Qi ,&nbsp;Zheng Qin ,&nbsp;Kelin Chen","doi":"10.1016/j.engeos.2026.100524","DOIUrl":"10.1016/j.engeos.2026.100524","url":null,"abstract":"<div><div>Seismic exploration is the most precise geophysical method for investigating and forecasting subsurface resources, yet its conventional fully elastic assumption neglects temperature effects, limiting its utility in geothermal studies. We present a novel full-waveform inversion (FWI) approach based on thermoelastic theory, enabling simultaneous inversion for subsurface temperature and elastic velocities by incorporating thermal influences into seismic wave propagation. The inversion framework minimizes the least-squares misfit between observed and calculated multi-component seismic data, with gradients for P-wave velocity, S-wave velocity, and temperature derived via the adjoint-state method. To address the nonlinearity and mitigate cycle-skipping, we progressively incorporate the multiscale inversion strategy into higher frequencies. Synthetic experiments on a modified Marmousi 2 model with coupled velocity and temperature anomalies demonstrate that thermoelastic FWI can accurately recover high-resolution velocity and temperature fields, closely matching the true models. These results confirm the feasibility of joint velocity-temperature inversion, establishing a theoretical foundation for direct seismic characterization of high-temperature geothermal reservoirs.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100524"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147612150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characterization of gravity-flow deposits in the Chang 7–9 oil groups in the Yanchang Formation, Ganquan County, Ordos Basin","authors":"Jiale Zhang ,&nbsp;Jingzhou Zhao ,&nbsp;Zhendong Gao ,&nbsp;Weimin Shen ,&nbsp;Fanrong Wei ,&nbsp;Xuangang Meng ,&nbsp;Xinzhi Yan ,&nbsp;Xin Zhang ,&nbsp;Weitao Wu ,&nbsp;Zilong Zhao ,&nbsp;Lei Cao ,&nbsp;Yubin Bai","doi":"10.1016/j.engeos.2026.100520","DOIUrl":"10.1016/j.engeos.2026.100520","url":null,"abstract":"<div><div>This study aims to address the challenges associated with the identification of the sedimentary facies of gravity-flow deposits in the 7th to 9th oil groups of the Yanchang Formation (also referred to as the Chang 7–9 oil groups) in Ganquan County, Ordos Basin. The sedimentary environments and developmental patterns of gravity-flow sand bodies in these oil groups are analyzed to provide a theoretical basis for petroleum exploration in deep-water deposits. Specifically, the type, spatial distribution, and differences in oil-bearing properties of gravity-flow deposits in the study area are determined through centimeter-scale observations of cores from 12 cored wells, grain size data analyses of 97 samples from 91 wells, and microscopic examination of 93 thin sections from 16 wells, as well as single-well log facies analyses, well-correlation section analyses, and the analyses of reservoir physical properties and oil-bearing properties. The results indicate that deep-water gravity-flow deposits are widely distributed across the Chang 7–9 oil groups in Ganquan County. These deposits can be categorized into slide, slump, sandy-debris-flow, and turbidity-current types, among which the latter two are the most prevalent. Sandy-debris-flow deposits dominate in the Chang 7 and 9 oil groups, while both sandy-debris-flow and turbidity-current deposits occur in high proportions in the Chang 8 oil group. Both sandy-debris-flow and turbidity-current deposits exhibit two types of grain-size cumulative probability curves: one-segment curves and short-tailed two-segment curves. Nevertheless, the sandy-debris-flow deposits are slightly coarser, with the coarse ends (i.e., portions corresponding to coarse grain sizes) of grain-size curves typically beginning below 2.5 φ and a mean grain size of 3.10 φ. In contrast, the turbidity-current deposits are relatively finer, with the coarse ends of grain-size curves mostly starting above 2.5 φ and a mean grain size of 3.84 φ. Compared to the turbidity-current deposits, the sandy-debris-flow deposits exhibit more favorable reservoir physical properties and oil-bearing properties. Therefore, sandy-debris-flow deposits should be prioritized as the primary target for tight oil exploration in the lower hydrocarbon play of the Yanchang Formation in the study area, followed by turbidity-current deposits.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100520"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Big data analysis of waterflood performance in mature conventional oilfields in Eastern China","authors":"Tianrui Ye,&nbsp;Zhiqiang Chen,&nbsp;Cheng Dai","doi":"10.1016/j.engeos.2025.100514","DOIUrl":"10.1016/j.engeos.2025.100514","url":null,"abstract":"<div><div>Most conventional oilfields in Eastern China with waterflood operations have reached ultra-high water cut in recent decade. The high water injection demand and produced water treatment cost pose significant environmental threats. Therefore, optimizing waterflood performance is key to improving production efficiency. This study performs data analysis on waterflood operations of all the oilfields operated by Sinopec across Eastern China. The production mechanisms and most effective operations for different reservoir types at diverse production stages are identified using data-driven methods. Random Forest models (RFMs) are constructed and integrated with Shapley Additive exPlanations (SHAP) analysis to quantify the weights and patterns of key geological and engineering features. A comparison of the estimated ultimate recovery factors for different blocks shows that geological factors play dominant roles in medium-to-high permeability reservoirs while development parameters are more critical for low-permeability reservoirs. The analysis of temporal data regarding field development and production history is conducted to select oil production-increasing operations in blocks. The results show that the most influential field operations vary for the diverse production stages, and well patterns should be carefully designed to improve production efficiency and reduce ineffective water circulation.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100514"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145940058","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Toward basin-agnostic well log imputation and anomaly detection via a pre-trained time-series foundation model","authors":"Ardiansyah Koeshidayatullah ,&nbsp;Abdulrahman Al-Fakih ,&nbsp;SanLinn Ismael Kaka","doi":"10.1016/j.engeos.2026.100536","DOIUrl":"10.1016/j.engeos.2026.100536","url":null,"abstract":"<div><div>Interpretation of subsurface storage characterization heavily depends on the quality and continuity of well log data. However, missing measurements and anomalous responses are common due to geological heterogeneity, tool limitations, and borehole conditions. Conventional machine learning and deep learning methods, such as Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNNs), have improved data recovery but remain constrained by the need for extensive retraining, labeled data, and basin-specific tuning. In this study, we present a novel application of a pre-trained time-series foundation model, TimeGPT, for well log imputation and anomaly detection that enables zero-shot inference across different basins and lithologies. This work represents one of the first applications of a generative pre-trained transformer GPT-based model to geoscientific time-series data, bridging recent advances in AI foundation modeling with subsurface analytics. We applied and fine-tuned TimeGPT using multi-log datasets (Gamma ray, Deep resistivity, Bulk density, Neutron porosity, and Sonic transit time) from the Groningen gas field in the Netherlands, and validated its performance against conventional machine learning and deep learning benchmarks. The proposed approach achieved a mean absolute error (<em>MAE</em>) between 0.02% and 0.32%, demonstrating a &gt;10% improvement over conventional models and comparable accuracy to advanced architectures such as bidirectional LSTM and Transformers. Moreover, the model attained 93% anomaly detection accuracy using conformal prediction intervals, effectively distinguishing among geological heterogeneity, tool noise, and borehole-related anomalies. The time-series foundation models can generalize well across different geological settings without retraining, enabling basin-agnostic and data-efficient well log analysis. The integration of self-attention mechanisms and conformal uncertainty quantification provides robust, interpretable predictions for real-world reservoir characterization. This work highlights the transformative potential of generative AI in geosciences, advancing well log interpretation toward scalable, low-risk, and foundation-model-driven analytics for the next generation of subsurface intelligence.</div></div>","PeriodicalId":100469,"journal":{"name":"Energy Geoscience","volume":"7 2","pages":"Article 100536"},"PeriodicalIF":3.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}