Release patterns of long-term potentially toxic elements from coal and gangue in Ordos and adjacent regions.

Quantitative tracing of natural gas origins in shale systems using intramolecular isotopic compositions of propane: A case study from a Paleozoic gas reservoir, Ordos Basin, China

Origins of Paleozoic natural gas in the Ordos Basin, North China, and implications for carbon isotope reversal mechanism

Hydrocarbon generation, expulsion processes, and gas occurrence characteristics of humic source rocks: Insights from the eastern margin of the Ordos Basin

This study proposes a novel method for lithological identification to address the challenge of accurate classification in complex sedimentary successions. Leveraging principal component analysis, the methodology was developed using Ordovician data from the Western Ordos Basin, where the lithological assemblage predominantly comprises of marine carbonate rocks, claystones, and their associated transitional rocks. Five well-log curves highly sensitive to lithology were selected as original variables, from which two principal components (PCs) were extracted to capture key compositional discrepancies. The first PC effectively differentiates clay and carbonate minerals, thereby enabling identification of transitional rocks between carbonates and claystones. The second PC reflects variations in calcite and dolomite contents, allowing discrimination of transitional lithologies between limestone and dolostone. Following the geological interpretation of these two principal components, a lithological cross-plot was constructed, which reveals distinct clustering among various rock types and supports a well-defined classification framework for transitional lithologies. The reliability and accuracy of the proposed method were validated against X-ray fluorescence element logging data from an active borehole. By prioritizing geological interpretability through feature transformation rather than treating the classification as a "black box", this study establishes a transparent, observable, and highly practical framework for lithological identification.

Understanding the differential evolution characteristics of pore structures in coals of various ranks under heat treatment conditions is fundamental for developing coal thermal mining technologies. This study systematically collected coal samples from the Zhangjiamao Mine in the Ordos Basin, the Dahebian Mine in western Guizhou, and the Sihe Mine in the Qinshui Basin, representing a series of samples with different coal ranks. Through drying, saturation, and heating processes, combined with low-field nuclear magnetic resonance (NMR) testing and pore fractal geometry theory, the effects of heat treatment at 200 °C, 300 °C, and 400 °C on NMR T 2 spectra, pore size distribution, two-dimensional relaxation distribution, and pore fractal dimensions of coal samples were investigated. The results indicate that after heat treatment, the porosity of coal samples across all coal ranks increased, with low-rank coal exhibiting the greatest increase, suggesting that low-rank coal pore structures possess the strongest temperature sensitivity. When the temperature reached 300 °C, the porosity of high-rank coal began to increase significantly, suggesting that its pore structure requires a higher activation temperature for thermal evolution. The DHB coal sample exhibited caking properties, and significant thermal expansion of the coal matrix after heating led to a smaller porosity at 300 °C compared to ZJM and SH samples. Heat treatment promoted the transformation of micropores into mesopores and macropores. Low-rank coal was characterized by an increase in mesopore volume, the pore connectivity of medium-rank coal was optimized, and the macropore volume of high-rank coal increased markedly with rising temperature. After heat treatment, the fractal dimensions of mesopores and macropores generally decreased, indicating reduced pore surface irregularity and enhanced connectivity, which facilitates fluid seepage. This study reveals the thermophysical differential response patterns of pores in coals of different ranks. Heat treatment affects coal pores though expansion, increase in number, and improved connectivity, creating favorable conditions for the fluid product during thermal extraction. The research findings provide theoretical and parametric support for geological site selection and process control of thermal mining, thereby promoting the development of thermal extraction technologies for coal resources.

Microscopic characteristics and fractal Representation of limestone reservoirs in the Taiyuan Formation, Ordos Basin

Differences in hydrocarbon generation characteristics and mechanisms between organic-rich laminated shale and turbidite mudstone from Triassic Chang 73 Member in Ordos Basin and geological implications

Evaluation of gas content prediction models for deep coal seams: A case study from the Daning–Jixian Block, Eastern Ordos Basin

Fluid type identification of low-permeability reservoirs based on logging response and GWO–XGBoost model: A case from the 8th Member of Triassic Yanchang Formation in Hongde area, Ordos basin

The synergistic mechanism between cyclic liquid nitrogen (LN2) freeze-thaw and water saturation in tight sandstone remains unclear, hindering the optimization of waterless fracturing parameters. This study investigated pore evolution in Ordos Basin tight sandstone across five saturation levels (0%-100%) using metallographic microscopy and low-temperature nitrogen adsorption. Structural complexity was quantified via the fractal theory. The results show that LN2 treatment significantly enhanced pore networks, with surface porosity increasing by 120–180%. Specific surface area and mesopore volume followed a three-stage “decrease–increase–decrease” trend, driven by a three-stage pore-size transformation: initial micropore coalescence into mesopores, subsequent nascent micro-fracture initiation, and final mesopore-to-macropore expansion. High water saturation ( ≥ 75 % ) intensified structural damage through coupled cryogenic thermal stress and ice-wedging pressure, yielding greater connectivity than dry conditions. Fractal analysis confirmed increased heterogeneity, with larger pores showing the highest sensitivity to cryogenic shock. This study identifies a 75% water saturation threshold, determining that 75% saturation combined with 6 freeze-thaw cycles maximizes oil production. By elucidating the pore reconstruction mechanism driven by water-phase freeze-thaw interactions, this study establishes a theoretical and practical benchmark for optimizing field injection scheduling in large-scale low-temperature geological engineering, enabling maximum fracture complexity with reduced LN2 consumption.

Research on expansion of hydraulic fractures based on reservoir architecture: implications for well spacing in the northern Ordos Basin tight gas reservoir

Diversity of calcareous cementation in uranium reservoir: a case study on the Shenshangou outcrop in the northeastern Ordos Basin

Autoregressive model with discrete feature representation for well log interpretation

ABSTRACT In recent years, automated identification of coal macerals based on deep learning has attracted widespread attention. However, automatic maceral identification remains challenging due to complex textures, pronounced scale variations, uneven illumination, and blurred boundaries commonly observed in coal petrographic images. To address these problems, this study proposed an attention-enhanced semantic segmentation framework based on DeepLab V3 for automated coal maceral identification. The proposed method adopted ResNet-101 as the backbone network and employed atrous convolution together with Atrous Spatial Pyramid Pooling (ASPP) to effectively capture multi-scale contextual information while preserving spatial resolution. To mitigate noise interference caused by scale variations and grayscale differences in coal petrographic images, attention mechanisms were introduced after the ASPP module to recalibrate feature responses. Three representative attention modules, including Squeeze-and-Excitation (SE), Efficient Channel Attention (ECA), and Convolutional Block Attention Module (CBAM), were comparatively evaluated to investigate their effectiveness in coal maceral segmentation.A pixel-level coal maceral dataset was constructed using coal petrographic images collected from the Ordos Basin, China, and extensive experiments were conducted under identical training and evaluation settings. Quantitative results demonstrated that the proposed DeepLab V3 framework outperformed several classical segmentation models with an average pixel accuracy of 0.92 and an mIoU of 0.82, including ResNet101-based, DenseNet121, U-Net, and VGG16 architectures. Among the evaluated attention mechanisms, CBAM exhibited the best performance, achieving a PA of 0.93 and a mIoU of 0.85, respectively, with class-wise IoU values of 0.89 for the vitrinite group, 0.81 for the inertinite group, and 0.85 for the background. The proposed model showed robust segmentation performance in regions with complex textures, uneven illumination, and grayscale overlap between macerals.Finally, the application performance of the model on the test dataset was discussed, and existing limitations as well as potential directions for improvement are identified. This study demonstrated that coupling the CBAM with ASPP in the DeepLab V3 semantic segmentation framework can effectively improve the accuracy of maceral identification in petrographic images.

CO2 capture and storage (CCS) is a pivotal technology for mitigating global climate change, but CO2 leakage from geological storage sites risks disrupting soil rhizosphere bacterial communities critical to soil functionality. To clarify their response mechanisms, this study focused on CO2-sensitive soybean (Shanning 17) in the Ordos Basin, establishing a simulated CO2 leakage platform with four treatments: CK (control, ambient CO2), 10%, 30%, and 50%. Illumina NovaSeq 16S rRNA sequencing combined with PICRUSt functional prediction was used to analyze rhizosphere bacteria. Results showed CO2 leakage altered the bacterial community: Chao1 index decreased by up to 4.27%, Pielou's evenness index increased, and Beta diversity changed notably. Bacteroidetes relative abundance increased significantly, while Proteobacteria, Acidobacteria, and Nitrospirae decreased; the rare phylum Deferribacteres was only detected under 30% and 50% CO2; Five key genera (RB41, MND1, Nitrospira, Solirubrobacter, Gaiella) dominated community shifts, with abundance trends matching their phyla. Functionally, The dominant Level 1 metabolism pathway (average relative abundance 81.02%), as well as the Level 2 pathways of amino acid metabolism, carbohydrate metabolism, and energy metabolism, were significantly affected, and these functional changes were closely linked to shifts in bacterial community structure. This study supports environmental impact and risk assessment for CCS projects in the Ordos Basin and similar geological storage areas.

The Daning Jixian block is situated on the eastern periphery of the Ordos Basin. it faces multiple challenges including strong reservoir heterogeneity and unclear identification of fracturing sweet spot interval interval intervals, which restrict the enhance productivity and operational efficiency of deep coal rock gas. This study adopts an integrated geology-engineering approach and employs ensemble optimized machine learning algorithms to predict fracturing sweet spot interval intervals. The recent exploration and development experiences along with extensive geological and production data in the block are summarized, 10 key controlling factors for reservoir quality and engineering quality are selected and organized into a dataset. The PSO-ELM algorithm was programmed on the MATLAB platform, into which this dataset was imported for iterative prediction. This process established a graded evaluation model for fracturing sweet spot interval interval intervals based on the PSO-ELM algorithm. The model was applied to predict the fracturing sweet spot intervals in the No. 5 and No. 8 coal seams of Well X-11, the overall prediction accuracy exceeds 85%, with Class I sweet spot intervals being predominant. Spatial distribution analysis across the block revealed abundant Class I and II sweet spots, confirming substantial exploitable high-quality resources. These predictions were validated by subsequent perforation and fracturing operation data, demonstrating that this methodology provides reliable logging technology support for perforation interval selection and fracturing stimulation optimization in deep coal reservoir development.

The Lower Jurassic Da’anzhai Member in the SichuanBasinpreserves a rare continental archive of the Toarcian Oceanic AnoxicEvent (T-OAE). However, the mechanisms governing differential organicmatter (OM) enrichment in its lacustrine shale successions remaindebated. This study presents an integrated organic geochemical, biomarker,and paleoenvironmental investigation of the Da13 and Da1 submembers from the central Sichuan Basin, aimingto elucidate their contrasting OM accumulation processes and linkthem to global climatic forcing. The results demonstrate that theDa13 submember, deposited during the T-OAE climaticoptimum, is dominated by aquatic (algal) OM, reflecting high primaryproductivity under warm-humid conditions and stable thermal stratificationin an open lacustrine system. In contrast, the Da1 submemberexhibits greater terrestrial input and elevated salinity under a hot-aridclimate, where OM preservation was largely confined to deep, stratifiedsettings due to frequent fluvial disruption in a closed basin. Integratingthese findings with regional T-OAE records from the marine Qiangtangand lacustrine Ordos basins, we establish a unified open-lake (warm-humid)versus closed-lake (hot–arid) enrichment model that links globalclimatic perturbation, basin evolution, and sediment supply to differentiallacustrine OM accumulation. Importantly, the interpreted lake evolutionfrom a warm-humid, open system to a hot-arid, closed system findsa strong parallel in lacustrine records of the Early Cretaceous OAE1afrom the Jiaolai Basin, eastern China. This consistency suggests thatour proposed framework captures a recurring pattern of lacustrineresponse to extreme climate perturbations, thereby reinforcing itsreliability and potential transferability to other basins and geologicalperiods characterized by similar climatic forcing. This study advancesthe understanding of continental paleoenvironmental responses to globalanoxic events and provides a predictive framework for shale oil explorationin analogous lacustrine settings worldwide.

Deep geological sequestration is an emerging strategy for managing high-salinity coal mine water. However, the migration behavior of fluoride (F-) during sequestration poses significant risks to groundwater safety and system stability. Elucidating the mechanisms of F- migration and transformation is therefore critical for ensuring the long-term safety of sequestration projects. In this study, static batch experiments were conducted to investigate F- migration and retention in sequestration formations with different lithologies. Three representative rock types were collected from the Liujiagou Formation (fine sandstone-muddy sandstone and fine sandstone-mudstone) and the Shiqianfeng Formation (medium sandstone-mudstone) in the Ordos Basin. The experiments were carried out under simulated formation temperature (55C) using synthetic high-salinity water with varying initial F- concentrations (20-150 mg/L) and pH conditions (2.0-10.0). Adsorption kinetics and isotherms were analyzed using pseudo-first-order, pseudo-second-order, and Freundlich models, while water-rock interaction mechanisms were evaluated through ionic ratio analysis, chloro-alkaline indices, and sodium adsorption ratio. Results showed that F- removal suggests dominance of chemical adsorption, following a rapid-slow-equilibrium pattern, and was well described by the pseudo-second-order kinetic model and Freundlich isotherm (R2 > 0.999). F- migration was likely governed by coupled processes of cation exchange, competitive adsorption, and mineral precipitation-dissolution. Fine sandstone-mudstone exhibited the strongest F- fixation capacity (462.7 mg/kg) due to synergistic adsorption and precipitation. Sensitivity analysis identified pH as the most influential factor. These findings provide a experimental basis for predicting F- pollution and support the safe implementation of deep geological sequestration in high-salinity coal mine water management.

ABSTRACT Under the “dual carbon” goals, coalbed methane, as a clean energy source, is of significant importance to energy security and carbon emission reduction. The gas content of coal is a key parameter for evaluating coal reservoirs. Traditional experimental methods are limited by core integrity and data constraints, making widespread application difficult. Logging data has the advantages of continuity and low cost, and combined with machine learning, it can effectively characterize the nonlinear relationship between logging responses and gas content. This study focuses on the No. 5 coal seam of the Shanxi Formation in the Yanchang Block on the southeastern edge of the Ordos Basin. Five logging parameters were selected: sonic transit time, natural gamma, deep lateral resistivity, shallow lateral resistivity, and density. Three prediction models were constructed: BP neural network, random forest, and convolutional neural network (CNN). Comparisons show that the CNN model performs best, with a coefficient of determination (R 2 ) of 0.93079 between predicted and measured gas content, and a mean absolute error of 0.36 m 3 /t. Using the CNN model to predict gas content distribution, the results show that the No. 5 coal seam gas content ranges from 2.16 to 25 m 3 /t, with an average of 21.34 m 3 /t, generally exhibiting higher values in the northwest and lower values in the central area. This study provides machine learning support for efficient prediction of coalbed methane gas content and has reference value for coalbed methane exploration and development.