Journal of Petroleum Geology最新文献

{"title":"A Comparative Study of Machine Learning Methods for Gas Hydrate Identification in Qilian Mountain Permafrost, China","authors":"Kun Xiao,&nbsp;Rong He,&nbsp;Changchun Zou,&nbsp;Zhenquan Lu,&nbsp;Hongxing Li,&nbsp;Xudong Hu,&nbsp;Pengbo Yang,&nbsp;Mengshi Chen","doi":"10.1111/jpg.70035","DOIUrl":"https://doi.org/10.1111/jpg.70035","url":null,"abstract":"<div>\u0000 \u0000 <p>Determining gas hydrate reservoirs using well-logging data is necessary for gas hydrate resource assessment. Conventional well-logging data interpretation can be tedious and time-consuming. Machine learning models can automate the well-logging interpretation process, saving time and reducing the need for the expertise of trained engineers. In this research, gas hydrate-bearing sediments in Qilian mountain permafrost of the Qinghai–Tibet Plateau were investigated. An extreme gradient boosting (XGBoost) ensemble learning model was constructed to identify gas hydrate-bearing sediments, with its performance benchmarked against typical machine learning algorithms including random forest (RF), K-nearest neighbors (KNNs), and gradient boosting decision tree (GBDT). The XGBoost model achieved the highest precision (99.0%) in identifying the gas hydrate of minority class in permafrost regions, demonstrating significant improvements over RF, KNN, and GBDT models. Through dimensionality reduction and sensitivity analysis of well-logging parameters for gas hydrate-bearing layers, the optimal parameter combination for identification was determined as caliper, resistivity, and bulk density. The optimized identification model based on ensemble learning algorithms provides theoretical foundations and technical support for detecting gas hydrate reservoirs in permafrost regions.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"502-515"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancements in Machine Learning–Driven Proxy Modelling Techniques for Reservoir Engineering: A Systematic Review","authors":"Eghbal Motaei,&nbsp;Tarek Ganat,&nbsp;Mehdi Tabatabai,&nbsp;Mian Umer,&nbsp;Shanker Krishna","doi":"10.1111/jpg.70037","DOIUrl":"https://doi.org/10.1111/jpg.70037","url":null,"abstract":"<div>\u0000 \u0000 <p>Application of machine learning (ML) in proxy models has grown in recent years and is evident across all disciplines in petroleum engineering for various tasks such as asset optimisation, production forecasting and production optimisation. This study reviews the literature and examines advancements in artificial intelligence/ML (AI/ML) usage in proxy models. A systematic literature review approach is applied to study the published relevant papers, and 83 papers were screened out of 1503 in the related area. Our findings indicate that there is a lack of validation processes in the workflow of proxy model creation in a majority of published works, which reduces the reliability of models for real-time applications. This makes the proxy models static designed for a specific problem only and results in single-use models. Various research tools are used to analyse the papers and understand the input parameters of the proxy models, their targeted parameters, and the approach utilised for proxy model creation. The study shows that neural networks are the most utilised and promising ML approaches for complex reservoir models with high performance metrics.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"516-540"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147614856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated Fractured Reservoir Modeling and Secondary Oil Recovery Techniques: A Historical Case Study for Productivity Improvement in the Chinguetti Offshore Oil Reservoir in the Coastal Basin of Mauritania","authors":"Bah Dahmane,&nbsp;Leila Aliouane,&nbsp;Bassem S. Nabawy,&nbsp;Mourad Guareche","doi":"10.1111/jpg.70027","DOIUrl":"https://doi.org/10.1111/jpg.70027","url":null,"abstract":"<div>\u0000 \u0000 <p>Naturally fractured reservoirs are inherently heterogeneous and anisotropic due to faulting and tectonic complexity, which makes reservoir characterization highly challenging. Additionally, these reservoirs deplete quickly, causing exacerbated productivity decline over time. Fracture modeling is therefore a key development issue for naturally fractured reservoirs and requires an integrated approach from geology to reservoir simulation and from micro- to field-scale modeling and field planning. Moreover, as fractures are below the limit of seismic resolution, the static fractures models are mainly constrained by borehole data (i.e., core images or image logs) using conventional structural geology techniques. Due to the importance of the Mauritania, Senegal, Gambia, Guinea-Bissau, and Guinea-Conakry basin (MSGBC basin) in the offshore of West Africa MSGBC basin, which was developed as a tectonically developed basin, including the naturally fractured Chinguetti oil field, and to address these challenges, a comprehensive approach has been proposed, including detailed statistical analyses for the fractures, identifying the fractures’ trends, and correlating them with the geological and production data. Fracture models were developed to predict and characterize the various fracture sets, their 3D spatial distribution, and their impact on the reservoir behavior. The Chinguetti field's production history illustrates the technical and operational challenges of managing a compartmentalized turbidite reservoir in a deepwater environment. It stands as a valuable case study in adaptive reservoir management, waterflood optimization, and probabilistic development planning in structurally complex offshore settings. The applied secondary recovery strategies include water and gas injection from five injection wells along with five producing wells, which are highly recommended to maintain the reservoir pressure. These scenarios may be extended to further offshore fields in West Africa and elsewhere.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"357-376"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147614870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction to “a New Saturated and Aromatic Hydrocarbon-based Scheme to Decipher the Origin of Extracted Organic Matter in the Outer Carpathians: From the Jurassic to the Miocene”","authors":"","doi":"10.1111/jpg.70038","DOIUrl":"https://doi.org/10.1111/jpg.70038","url":null,"abstract":"<p>Słowakiewicz, M. (2025), A New Saturated and Aromatic Hydrocarbon-Based Scheme to Decipher the Origin of Extracted Organic Matter in the Outer Carpathians: From the Jurassic to the Miocene. Journal of Petroleum Geology, 48: 188–204. https://doi.org/10.1111/jpg.70001</p><p>In Section 3.2, “Sample and Extraction Preparation”, information on the extraction protocol was missing and should have been added as follows:</p><p>Rock and drill core samples were first trimmed to remove weathered or stained surfaces, then washed with pure acetone and dried at room temperature. Samples of approximately 10 g each were subsequently ground to fractions &lt;0.2 mm before extraction.</p><p>Extraction was carried out in an ultrasonic bath using a dichloromethane–methanol (9:1, v/v) mixture as the solvent for 15 min, with the extraction repeated twice. The asphaltene fraction was isolated from the extracted bitumens using <i>n</i>-hexane. The resulting maltenes (a mixture of saturated hydrocarbons, aromatic hydrocarbons and resins) were separated into individual fractions using a chromatographic column (0.8 × 150 mm) packed with activated (125°C) silica gel, with <i>n</i>-hexane, <i>n</i>-hexane–dichloromethane (3:1, v/v) and toluene–methanol (1:1, v/v) used as eluents, respectively. Only compounds eluting in the saturated and aromatic hydrocarbon fractions are reported herein.</p><p>We apologize for this error.</p>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":""},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jpg.70038","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147569907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated Sequence Stratigraphic and Petrophysical Evaluation of Middle Miocene Reservoirs, Offshore East Esh El Mallaha, Gulf of Suez, Egypt","authors":"Ahmed K. El-Behery,&nbsp;Ahmed Abdeldaim,&nbsp;Khaled Al-Kahtany,&nbsp;Emad A. Abdallah,&nbsp;Moataz Kh. Barakat,&nbsp;Sherif Farouk","doi":"10.1111/jpg.70028","DOIUrl":"https://doi.org/10.1111/jpg.70028","url":null,"abstract":"<div>\u0000 \u0000 <p>The reservoirs of the Middle Miocene Kareem and Belayim formations in the offshore East Esh El Mallaha area of the southern Gulf of Suez are significant as they represent hydrocarbon zones resulting from an active rift phase in the basin. The difficulty in predicting and correlating the reservoirs arises from their laterally and vertically different facies, in addition to the complicated syn-depositional faulting. As a result, the objective of this research is to characterize the reservoirs in question by using a combination of methods that focus on formation evaluation, sequence stratigraphy, and high-resolution cyclostratigraphy to establish the stratigraphic architecture and depositional cyclicity. Well-log interpretation, further validated by cyclostratigraphy frequency analysis and sedimentation rate modeling, leads to the identification of five third-order depositional sequences within the Miocene Kareem and Belayim formations. These sequences represent the considerable changes in accommodation space, ultimately leading to a significant sea-level regression approximately 15.3 million years ago (Ma), as the Rhami were nonexistent at that time. The stratigraphic succession, consisting of basal evaporites capped by clastic Hammam Faraun sandstones, displays different sediment accumulation rates (SARs) of up to 20 cm/kyr in wells NG-3X and EEMM-39. In contrast, in EEMM-25 and EEMM-37, the rates drop to 6 and 2.5 cm/kyr, respectively, indicating lesser sedimentation over the structural highs. Seismic interpretation has identified three seismic facies, and a structural analysis has demonstrated that NW–SE trending syn-depositional faults played a primary role in controlling sediment distribution and facies organization. The most productive reservoirs are located in the highstand systems tract (HST) of the Hammam Faraun Member and the lowstand systems tract (LST) of the Shagar Member, respectively. The identified intervals comprise sandy, feldspathic facies, which exhibit excellent reservoir characteristics, with porosity reaching as high as 23% and hydrocarbon saturation levels of up to 68%. The present study reported six lithofacies that illustrated the interactions of tectonic activity, sea-level changes, and sediment supply.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"377-396"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147615423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Paleogeographic and Diagenetic Controls on Oolitic Reservoirs: East–West Differences Along the Southern K–L Trough, NE Sichuan Basin (Triassic Feixianguan Formation)","authors":"Xiong Ding,&nbsp;Yi Song,&nbsp;Yuefeng Sun,&nbsp;Xinyang He,&nbsp;Qiqi Che,&nbsp;Xiao Chen,&nbsp;Chang Xu,&nbsp;Jing Tang,&nbsp;Xiran Yang,&nbsp;Zhongfan Xu,&nbsp;Meimei Han","doi":"10.1111/jpg.70033","DOIUrl":"https://doi.org/10.1111/jpg.70033","url":null,"abstract":"<div>\u0000 \u0000 <p>The Triassic Feixianguan Formation's platform-margin oolitic shoal is a major gas reservoir in the northeastern (NE) Sichuan Basin, along the Kaijiang–Liangping (K–L) Trough. Currently, the platform-margin oolitic shoal reservoirs in the southern sector of the K–L Trough have been studied relatively little, particularly regarding their characteristics and development. Our detailed studies of petrology and isotope geochemistry have revealed differences in the characteristics of oolitic shoal reservoirs on the eastern and western sides within the southern sector of the K–L Trough. The oolitic shoal reservoirs on the eastern side are mainly composed of intragranular dissolved pores and moldic pores of oolitic dolomite, with high δ¹⁸O, high Fe and Mn contents, and low ⁸⁷Sr/⁸⁶Sr ratios, with a weak positive correlation between porosity and permeability; pore throats are relatively undeveloped. In contrast, the oolitic reservoirs on the western side are dominated by residual intergranular pores and intergranular dissolved pores of oolitic limestone, with low δ¹⁸O, low Fe and Mn, and high ⁸⁷Sr/⁸⁶Sr ratios, and their porosity and permeability show a good positive correlation. The throats are thick and the pore throats have a good distribution. The differences of sedimentary paleogeomorphology and diagenetic evolution are considered to be the main causes of the differences of oolitic reservoir characteristics on the eastern and western sides within the southern sector of the K–L Trough. The oolitic shoals on the eastern side formed in a higher and steeper sedimentary topography (slope angles &gt;15°), which promoted vertical stacking because of the steep gradient, and there was greater exposure to meteoric diagenesis. Meteoric diagenesis led to considerable cementation and selective dissolution of the grain fabrics, causing substantial reduction in the original intergranular pores and the development of intragranular dissolved pores and moldic pores. Conversely, the oolitic shoals on the western side formed in a lower and gentler sedimentary topography (slope angles &lt;5°), which migrated laterally toward the trough over large-scale distances of 9–15 km, and limited exposure to atmospheric diagenesis, which was conducive to the preservation of residual intergranular pores and the further formation of intergranular dissolved pores in the burial process. These results enhance our understanding of the characteristics and development of oolitic reservoirs in the southern sector of the K–L Trough and provide insights for further hydrocarbon exploration of deep oolitic reservoirs.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"409-432"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Depositional Facies, Sequence Stratigraphy, and Diagenesis of Lower Cretaceous Carbonate Reservoir, Southern Iraq: Implications for Petroleum Exploration","authors":"A. K. A. Mohammed,&nbsp;F. Velledits,&nbsp;I. Q. Mohammed,&nbsp;M. Alsuwaidi,&nbsp;H. Mansurbeg,&nbsp;F. Kristály,&nbsp;D. K. Ivanova,&nbsp;Eloi González-Esvertit,&nbsp;A. Awdal","doi":"10.1111/jpg.70025","DOIUrl":"https://doi.org/10.1111/jpg.70025","url":null,"abstract":"<p>The Lower Cretaceous Yamama Formation of southern Iraq represents a key carbonate reservoir within Iraq and the Middle East, yet its complex depositional facies architecture and diagenetic alterations present challenges for predicting reservoir quality. This study integrates well-log interpretation, detailed description of about 400 m of core intervals, petrographic analysis, fluid-inclusion microthermometry, and stable-isotope (δ¹³C, δ¹⁸O) geochemistry from several oilfields to determine the controls on reservoir heterogeneity. Fourteen depositional facies are recognized across a W–E-trending, slightly steepening homoclinal ramp. These vary into four reservoir (YRA–YRD) and four nonreservoir (B0–B3) units. Reservoir intervals consist mainly of shoal grain-supported peloids and ooids and Lithocodium–Bacinella reefal facies, whereas mud-supported lagoonal and middle-ramp facies form permeability barriers. Diagenetic processes exerted strong control on reservoir quality evolution. Early marine rim and scattered calcite cements preserved interparticle pores in the grain-supported limestone facies, whereas burial equant calcite sourced by stylolitization of the host limestones reduced porosity. The precipitation of pore-filling kaolin and saponite in the mud-supported limestone facies limited porosity. Saponite is suggested to have been formed from alkaline brines developed under restricted, evaporative conditions. The formation of kaolin is attributed to the migration of Al³⁺-bearing brines charged with organic acids along stylolites, during tectonic compression of the basin. The presence of framboidal and euhedral pyrite indicates a diagenetic shift from microbial to thermochemical sulfate reduction. TSR suggests that brines were thermally evolved and migrated upward from the underlying, hotter evaporitic formation, such as the Upper Jurassic Gotnia. Fluid-inclusion microthermometry in blocky calcite in moldic/vuggy pores indicates homogenization temperatures of 85°C–140°C (exceeding the maximum burial temperature, ∼116°C) from brines composed of NaCl–KCl–H₂O, with total salinity of 10 wt.% NaCl eq. and δ¹⁸O_VSMOW values of +4‰ to +8‰. These values, along with depleted δ¹⁸O_VPDB (−6.9‰ to −5.6‰), suggest precipitation from evolved basinal fluids related to stylolitization. The occurrence of saddle dolomite in close association with euhedral pyrite suggests precipitation from brines that were affected by TSR. Four third-order depositional sequences bounded by Type 2 sequence boundaries are recognized; regressive shoal and reefal facies formed the reservoir units, whereas transgressive, mud-dominated facies are impervious. The integration of depositional facies, diagenesis, and sequence stratigraphy indicates that depositional facies architecture and fluid migration allow better understanding of the reservoir heterogeneity in the Yamama Formation.</p>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"312-345"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jpg.70025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147567042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geochemical and Seismic Constraints on the Role of Mud Diapirs in Petroleum Migration and Accumulation in the Recôncavo Basin, NE Brazil","authors":"Cora Mattos,&nbsp;Egberto Pereira,&nbsp;Lucas Pinto Heckert Bastos,&nbsp;René Rodrigues","doi":"10.1111/jpg.70030","DOIUrl":"https://doi.org/10.1111/jpg.70030","url":null,"abstract":"<p>In the Recôncavo Basin, Northeast of Brazil, mud diapirs have been reported for decades, but their role within the basin's petroleum system remains poorly constrained. Here, we combine 2D seismic interpretation, field observations, and organic geochemical analyses of rocks and oil seeps to investigate the relationship between diapirism and hydrocarbon migration in the southern Recôncavo Basin. Oil samples were collected from natural seeps along the unconformity between the Cinzento Diapir (Valanginian–Hauterivian) and the overlying Marizal Formation (Aptian), whereas potential source rocks were sampled 10 km south of the diapir (outcrop D-01). For the rock samples, bulk geochemical parameters indicate total organic carbon (TOC) contents of up to 4.0 wt% and hydrogen index values exceeding 700 mg HC/g TOC in the upper stratigraphic interval, pointing to substantial preservation of organic matter. However, both <i>T</i>_max and biomarker data consistently indicate that the organic matter remains in an immature to early-mature thermal stage to samples collected in the outcrop. Stable carbon isotope compositions (δ¹³C_org = −26‰ to −31‰) and biomarker assemblages—dominated by long-chain <i>n</i>-alkanes (<i>n</i>-C₂₅–<i>n</i>-C₃₁), abundant C₂₇ regular steranes, methylsteranes, tetracyclic polyprenoids, and gammacerane—indicate a lacustrine depositional setting with variable salinity and microbial reworking of algal organic matter. The petroleum sample exhibited a hydrocarbon biomarker configuration consistent with thermally evolved material (petroleum), including higher <i>T</i>_s/<i>T</i>_m ratios, increased tricyclic terpane abundances, and a predominance of thermally stable hopanes and steranes. Despite this maturity difference, biomarker correlations—such as the predominance of C₂₃ tricyclic terpane, elevated tetracyclic polyprenoids, comparable gammacerane/hopane ratios, and similarities in regular steranes distribution—suggest that the collected oil was sourced from organic facies similar to those observed at outcrop D-01, potentially deposited in deeper basin depocenters. Seismic interpretation of the Cinzento Diapir reveals listric fault geometries and chaotic internal facies consistent with diapiric ascent and fault-controlled deformation. These features define potential vertical migration conduits linking mature source intervals to shallow reservoirs and outcrop seeps. The combined geochemical and seismic evidence supports a model in which mud diapirs in the Recôncavo Basin could have acted as important pathways for hydrocarbon migration and accumulation that occurred along structural discontinuities, thereby exerting a first-order control on the spatial distribution of petroleum occurrences in this early Cretaceous rift system.</p>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"433-447"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/jpg.70030","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147565922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Laminated Shale nm–µm-Scale Pore System Structure, Fractal Dimensions and Its Controlling Factors of the Paleocene Shahejie Formation: Evidences From CO2 Adsorption and N2 Adsorption Experiments","authors":"Biao Sun,&nbsp;Xiaoping Liu,&nbsp;Xiaoguang Li,&nbsp;Yongcheng Chen,&nbsp;Chang Chen,&nbsp;Tian Liu,&nbsp;Zilong Zhang,&nbsp;Wenhui Xie,&nbsp;Murray Gingras","doi":"10.1111/jpg.70039","DOIUrl":"https://doi.org/10.1111/jpg.70039","url":null,"abstract":"<div>\u0000 \u0000 <p>Characterizing the pore system of shale oil reservoirs is critical to elucidating petroleum accumulation mechanisms, enhancing oil recovery efficiency, and evaluating unconventional petroleum resources. However, research on lacustrine laminated shale pore systems remains limited and poorly understood, primarily constrained by the strong heterogeneity induced by differential lamina development. To address this gap, we collected 20 lacustrine shale samples from the fourth member of the Shahejie Formation in the Bohai Bay Basin and systematically investigated their nm–µm scale pore structure using CO₂ and N₂ adsorption experiments. The results show that the laminated shales are composed of light-colored mineral-rich layers (quartz, feldspar, calcite, and dolomite) alternating with dark clay–organic matter layers. Their pore system is dominated by matrix intercrystalline, intergranular, dissolution pores, and microfractures, rather than organic-matter-associated pores, with pore diameters primarily ranging from 2 to 40 nm. Notably, mesopores account for over half of the total specific surface area and pore volume in all tested samples, and the abundance of clay and quartz minerals is identified as the key factor controlling micro–mesopore composition and complexity. These findings provide critical insights for understanding shale oil resource potential and delineating optimal exploitation targets in lacustrine basins.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"484-501"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147570278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Comparison of Supervised Machine Learning Algorithms to Predict Porosity in Carbonate Reservoirs From Well Logs: A Case Study of Quissamã Formation Carbonate in the Campos Basin, Brazil","authors":"Gisela M. S. Almeida,&nbsp;Carlos H. S. Barbosa,&nbsp;Maira C. O. L. Santo,&nbsp;Luiz Landau","doi":"10.1111/jpg.70036","DOIUrl":"https://doi.org/10.1111/jpg.70036","url":null,"abstract":"<div>\u0000 \u0000 <p>Porosity estimation is a critical step in characterizing carbonate reservoirs, particularly given their complex pore systems and heterogeneity. This study proposes a data-driven approach to predict porosity in the Quissamã Formation, a carbonate unit in Brazil's Campos Basin, using supervised machine learning techniques. Routine core analysis data served as the target variable, while conventional well logs were used as input features. Eight machine learning algorithms were evaluated: Ridge Regression (RR), Support Vector Regression (SVR), Decision Tree (DT), Random Forest (RF), Extremely Randomized Trees (ET), K-Nearest Neighbors (KNN), and two Multilayer Perceptrons (MLP-1 and MLP-2). The models were optimized through hyperparameter tuning and validated using cross-validation. A total of 328 samples were used, of which 210 (from 3 wells) were allocated for training and validation and 118 (from 5 wells) reserved for blind-well testing, ensuring an independent evaluation of model generalization. Their performance was compared against four traditional porosity estimation methods based on individual and combined well logs. Overall, the machine learning models achieved higher accuracy, with RR demonstrating the most consistent results across wells. This work highlights the potential influence of data representativeness and tuning strategy on prediction performance. Although focused on the Quissamã Formation, the methodology is scalable and adaptable to other reservoirs. Moreover, machine learning models offer practical advantages, requiring no prior petrophysical knowledge and enabling real-time application during well logging, thereby facilitating timely and informed reservoir evaluations.</p>\u0000 </div>","PeriodicalId":16748,"journal":{"name":"Journal of Petroleum Geology","volume":"49 2","pages":"448-465"},"PeriodicalIF":1.7,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147568876","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}