Pore structure plays a critical role in evaluating shale “sweet spots”. Compared to marine shale, lacustrine shale has more diverse lithofacies types and greater heterogeneity in pore structure due to frequently changing environmental conditions. Using methods such as mercury intrusion porosimetry (MIP), field emission scanning electron microscopy (FE-SEM), nuclear magnetic resonance (NMR), and X-ray diffraction (XRD), this study investigates the micropore structures and heterogeneity of different lithofacies in the Jurassic Dongyuemiao Member lacustrine shale. Image processing and multifractal theory were employed to identify the controlling factors of pore structure heterogeneity. The key findings are as follows. (1) Based on mineral content and laminae types, the lithofacies types of Dongyuemiao lacustrine shale are classified into four types: shell–laminae mixed shale (SLMS), silty–laminae clay shale (SLCS), clast–laminae clay shale (CLCS), and clay shale (CS). (2) Based on genesis, shale reservoirs’ pore and permeability space are categorized into inorganic pores, organic pores, and micro-fractures. Inorganic pores consist of inter-particle pores and intra-particle pores. Pore size distribution curves for all four lithofacies exhibit two main peaks, with pore sizes concentrated in the ranges of 2–10 nm and 50–80 nm. Mesopores and macropores dominate, accounting for over 80% of the total pore volume. Mesopores are most developed in CLCS, representing 56.3%. (3) Quartz content is positively correlated with the multifractal dimension, while clay content shows a negative correlation. Higher quartz content, coupled with lower clay content, weakens pore structure heterogeneity. A negative correlation exists between total organic carbon (TOC) and the multifractal dimension, indicating that higher organic matter content enhances organic pore development and increases microscopic heterogeneity. (4) Porosity heterogeneity in SLMS is effectively characterized by D₀-D_max, while in the other three lithofacies, it is characterized by D_min-D₀. Permeability across all lithofacies correlates with D₀-D_max. In CS, SLMS, and SLCS, permeability is positively correlated with D₀-D_max, with higher values indicating greater permeability heterogeneity. In CLCS, permeability is negatively correlated with D₀-D_max, such that lower values reflect stronger heterogeneity.

The pore structure plays a critical role in evaluating shale “sweet spots”. Compared to marine shale, lacustrine shale has more diverse lithofacies types and greater heterogeneity in pore struc-ture due to frequently changing environmental conditions. Using methods such as mercury intru-sion porosimetry (MIP), field emission scanning electron microscopy (FE-SEM), nuclear magnetic resonance (NMR), and X-ray diffraction (XRD), this study investigates the micropore structures and heterogeneity of different lithofacies in the Jurassic Dongyuemiao Member lacustrine shale. Image processing and multifractal theory were employed to identify the controlling factors of pore structure heterogeneity. The key findings are as follows: (1) Based on mineral content and laminae types, the lithofacies types of Dongyuemiao lacustrine shale are classified into four types: shell-laminae mixed shale (SLMS), silty-laminae clay shale (SLCS), clast-laminae clay shale (CLCS), and clay shale (CS). (2) Based on genesis, shale reservoirs pore and permeability space are cate-gorized into inorganic pores, organic pores, and micro-fractures. Inorganic pores consist of in-ter-particle pores and intra-particle pores. Pore size distribution curves for all four lithofacies ex-hibit two main peaks, with pore sizes concentrated in the ranges of 2-10nm and 50-80nm. Meso-pores and macropores dominate, accounting for over 80% of the total pore volume. Mesopores are most developed in CLCS, representing 56.3%. (3) Quartz content is positively correlated with the multifractal dimension, while clay content shows a negative correlation. Higher quartz content, coupled with lower clay content, weakens pore structure heterogeneity. A negative correlation exists between total organic carbon (TOC) and the multifractal dimension, indicating that higher organic matter content enhances organic pore development and increases microscopic heteroge-neity. (4) Porosity heterogeneity in SLMS is effectively characterized by D0-Dmax, while in the other three lithofacies, it is characterized by Dmin-D0. Permeability across all lithofacies correlates withD0-Dmax. In CS, SLMS, and SLCS, permeability is positively correlated with D0-Dmax, with higher values indicating greater permeability heterogeneity. In CLCS, permeability is negatively correlated with D0-Dmax, such that lower values reflect stronger heterogeneity.

High quantities of sporopollen fossil within geological formations areessential for the distinction and correlation of stratigraphic units. Asthe foundation of numerous geology-related tasks, the efficient classi-fication of sporopollen fossil directly relates to the subsequent works.Human-based methods are more time-consuming and dependent onexpertise for accuracy. Therefore, intelligent methods of identificationare needed. Although current automatic fossil identification is moreefficient compared to manual identification, this method is faced witha lack of versatility and applicability to the identification of multiplespecies within samples. To account for the complexity of the types ofsporopollen fossil and the presence of several distinct fossil particleswithin a sporopollen fossil image, this paper proposes a hierarchical intel-ligence localization and recognition model for sporopollen fossils basedon deep learning computer vision technology.This model initially used object detection technology to authenticate the position and main typeclassification of sporopollen fossil, and then a convoluted neural networkwas used to identify detailed characteristics in each type of fossil image,after which machine learning methods were applied to determine typeclassification and achieve hierarchical identification, which increases theaccuracy of recognition. In this paper, the obtained model was evalu-ated using 1,604 images of sporopollen fossil including 39 genera wherethe final accuracy of recognition averaged 95.42%, indicating the modelcan accurately localize and recognize particles of sporopollen fossil.

Currently, the Longmaxi shale in the Sichuan Basin is the most successful stratum of shale gas production in China. However, because Longmaxi shale mostly has high over-maturity, a low-maturity sample cannot be obtained for gas generation thermal simulations, and as a result, a gas generation model has not yet been established for it. Therefore, models of other shales are usually used to calculate the amount of gas generated from Longmaxi shale, but they may produce inaccurate results. In this study, a Longmaxi shale sample with an equivalent vitrinite reflectance calculated from Raman spectroscopy (EqVR_o) of 1.26% was obtained from Well Yucan 1 in the Chengkou area, northeast Sichuan Province. This Longmaxi shale may have the lowest maturity in nature. Pyrolysis simulations based on gold tubes were performed on this sample, and the gas generation line was obtained. The amount of gas generated during the low-maturity stage was compensated by referring to gas generation data obtained from Lower Silurian black shale in western Lithuania. Thus, a gas generation model of the Longmaxi shale was built. The model showed that the gas generation process of Longmaxi shale could be divided into three stages: (1) First, there is the quick generation stage (EqVR_o 0.5–3.0%), where hydrocarbon gases were generated quickly and constantly, and the generation rate was steady. A maximum of 458 mL/g TOC was reached at a maturity of 3.0% EqVR_o. (2) Second, there is the stable stage (EqVR_o 3.0–3.25%), where the amount of generated gas reached a plateau of 453–458 mL/g TOC. (3) Third, there is the rapid descent stage (EqVR_o > 3.25%), where the amount of generated gas started to decrease, and it was 393 mL/g TOC at an EqVR_o of 3.34%. This model allows us to more accurately calculate the amount of gas generated from the Longmaxi shale in the Sichuan Basin.

The Jurassic lacustrine oil shale in southwest China has become a primary production layer due to its high yield and substantial reserves. However, influenced by the lacustrine environment, the vertical profile of the lacustrine shale reservoir shows alternating deposits of shale and carbonate rock. This complex lithological combination results in significant heterogeneity in reservoir types, reservoir distribution, and internal structure. Currently, research on micro-pore structure and hydrocarbon storage mechanisms in lacustrine shales is insufficient, necessitating the elucidation of their micro-characteristics to support future exploration and development. This research focuses on the Da’anzhai Member of Jurassic Ziliujing Formation. Various techniques—including organic geochemical analysis, X-ray diffraction, physical property testing, gradient centrifugation, and gradient drying NMR monitoring—were employed to investigate the micro-pore structure and fluid storage mechanisms of the lacustrine shale reservoir. The following insights were gained from this research. The organic matter pores (OMP) and inorganic pores (IP) developed within the Da’anzhai lacustrine shale reservoir together create the storage space for shale oil, while micro-fractures further enhance the reservoir’s storage capacity and flow performance. Lacustrine shale oil exists in three storage states: mobile oil, bound oil, and adsorbed oil. Mobile oil is primarily located within the micro-fractures and large pores (greater than 350 nm) of the shale reservoir and is the main target for industrial extraction. Bound oil is mainly found in the meso-pores, micropores, and narrow pore structures between rock grains (30 nm to 350 nm), and, theoretically, could potentially be developed through engineering methods such as hydraulic fracturing. Adsorbed oil, due to its close binding with organic matter and clay mineral surfaces, is difficult to release effectively using conventional techniques. The OM abundance, the mineral composition of lacustrine shale, and the pore structure all influence the storage states of shale oil. While a high TOC value increases the amount of mobile oil, the strong adsorption properties of kerogen and organic matter lead to the accumulation of adsorbed oil, which inhibits oil flow. Clay minerals further restrict oil flow by enhancing adsorption, while brittle minerals facilitate the movement of mobile oil by expanding pore space. Based on fractal geometry theory and multi-scale testing results, the large pores in the Da’anzhai lacustrine shale have a high fractal dimension and exhibit complex shapes. However, as pore complexity increases, the amount of adsorbed oil rises significantly, which in turn reduces the proportion of movable oil.

Closely spaced lithologic contacts create scale-dependent variations in coal-measure well logs, while incomplete logging suites complicate cross-well comparison. A geology-aware multiscale stability-regularized fractal coupling graph (GeoMSR-FCG) links local analysis support to stratigraphic boundaries and represents cross-log coupling at explicit physical scales. A multilog gradient field regulates Higuchi fractal dimension (HFD) support; detrended cross-correlation analysis (DCCA) defines graph edges at 3, 6, and 12 m. Separate roof and floor contrasts capture stratigraphic asymmetry, and projection onto measured target edges enables transfer between logging suites. Evaluation combines 960 held-out synthetic realizations with a five-well proof of concept comprising 35 coal intervals in the eastern Ordos Basin. For a 25-sample embedded layer, boundary regulation reduces HFD root-mean-square error from 0.1931 to 0.1868; overall error is 0.1828, compared with 0.1823 for regularization without the boundary term. In the field wells, median effective weighted support decreases from 5.42–5.77 m to 5.28–5.49 m. At 3 m, well-balanced coal–host coupling-magnitude contrasts are −0.652 for gamma ray–deep resistivity and −0.160 for acoustic slowness–deep resistivity, while density–neutron contrasts are positive in the three wells with both logs. Transfer to the reduced Jin2 and Gao2 suites yields ROC-AUC values of 0.762 and 0.781. GeoMSR-FCG characterizes coal-measure heterogeneity through boundary-adaptive fractal support and scale-specific coupling among the measured logs.