Predicting gas-bearing properties in tight sandstone reservoirs presents a global challenge. Traditional methods based on well log interpretation rely heavily on individual experience, which can introduce significant unknown errors. Prediction methods using seismic data and logging labels often fail to capture complex interactions between geological features, resulting in low accuracy. Furthermore, these methods typically determine only gas presence without providing quantitative results. To address these limitations, this study proposes a novel interpretable machine learning (ML) framework. Its novelty lies in: (1) directly linking well testing conclusions to logging data to provide high-resolution, semi-quantitative gas-bearing labels, eliminating intermediate interpretation errors; (2) a systematic comparison of 19 ML algorithms across different paradigms (traditional ML, deep learning, and ensemble learning) using five tailored evaluation metrics, identifying LightGBM as the optimal model for this task (Accuracy = 99.76%); and (3) integrating interpretability directly into the prediction workflow based on cooperative game theory to provide global and local explanations that align with petroleum geological knowledge, significantly enhancing the model’s transparency and credibility. Applied to the Xujiahe Formation in the Sichuan Basin, this framework achieves decimeter-level accuracy and demonstrates strong generalization capability. This work proposes a novel framework that enables semi-quantitative gas-bearing property predictions with the potential for basin-scale application, directly identifying sweet spots and offering a more streamlined and interpretable high-accuracy artificial intelligence method for oil and gas resource exploration and development.

Low-temperature nitrogen adsorption can form a “hysteresis loop”, whose geometry and area can effectively reflect the pore structure of porous media and its retention effect on adsorbed gas. However, the role of the “hysteresis loop” in quantitative characterization of shale pore structure has often been ignored. The study aims to clarify whether shale can form “hysteresis loop” in low-temperature nitrogen adsorption-desorption experiment and the determinants on hysteresis loop’s area with experiments on the 7^th member of the Upper Triassic Yanchang Formation shale (Chang 7 shale) in the Ordos Basin. Various measures are applied in the study, including qualitative observation of pore structure under field-emission scanning electron microscopy (FE-SEM) and quantitative characterization of pore structure by low-temperature nitrogen adsorption test, hysteresis-loop quantitative analysis, total organic carbon (TOC) content analysis, pyrolysis and X-ray diffraction (XRD) experiments. The following results are obtained. First, whether shale can form a hysteresis loop in the low-temperature nitrogen adsorption-desorption experiment has an apparently positive correlation with the specific surface area, specific pore volume, clay mineral content, and pore structure fractal dimension, and an evidently negative correlation with the TOC content, while no apparent correlation with the average pore size, pore surface fractal dimension, highest pyrolysis peak temperature, and content of brittle minerals. Second, the hysteresis loop area depends on the development degree of the cylindrical pores with both ends open, ink-bottle pores, or parallel plate pores, the proportion of which to the total pores can be quantitatively evaluated by the hysteresis loop area. Third, the open-ended cylindrical pores, ink-bottle pores, or parallel plate pores in the samples from Chang 7 shale are mainly of the intergranular pores in clay minerals. Therefore, there is an apparently positive correlation between the hysteresis loop area and the content of clay minerals.

As the global petroleum exploration domain gradually shifts from conventional to unconventional hydrocarbon resources, the classical petroleum system theory faces new challenges in terms of guiding the deepening exploration practices in the petroleum industry. After years of research, Chengzao Jia proposed the whole petroleum system concept and established an orderly distribution model for the coexistence of conventional and unconventional petroleum, which provides a new theoretical framework for the joint assessment and integrated exploration of conventional and unconventional petroleum resources. In this context, the 1^st International Symposium on Whole Petroleum System Theory and New Directions for Petroleum Geology Development was held in Beijing in October 2-3, 2023. The theme was “Whole petroleum system theory and new frontiers in petroleum exploration”. Experts engaged in in-depth discussions on the progress of whole petroleum system theory and development directions of petroleum geology; they systematically reviewed the new theory developments and advances in sequence stratigraphy, tight oil and gas, shale oil and gas reservoir characteristics, genetic mechanisms, and development mechanisms. The conference also proposed unified genetic models for conventional and unconventional petroleum resources, and novel methods and technologies for joint assessment. Furthermore, it also included case studies on the whole petroleum system in clastic and carbonate formations in oil and gas basins, challenges, opportunities, and new directions in the development of petroleum geology. This symposium provided a valuable opportunity for the petroleum geology community to gain a deep understanding of the “whole petroleum system theory” and to summarize and refine the development directions of petroleum geology. Undoubtedly, this event contributes to the advancement of the whole petroleum system theory, guiding the development of petroleum geology theory and further promoting the joint assessment and integrated future development and utilization of conventional and unconventional petroleum resources.