Coal texture is important for predicting coal seam permeability and selecting favorable blocks for coalbed methane (CBM) exploration. Drilled cores and mining seam observations are the most direct and effective methods of identifying coal texture; however, they are expensive and cannot be used in unexplored coal seams. Geophysical logging has become a common method of coal texture identification, particularly during the CBM mining stage. However, quantitative methods for identifying coal texture based on geophysical logging data require further study. The support vector machine (SVM), a machine-learning method, has received great interest due to its remarkable generalization performance, and it has been used to quantitatively identify hard and soft coal using geophysical logging data. In this study, four well-logging curves, the acoustic time difference (AC), caliper log (CAL), density (DEN), and natural gamma (GR), were used for coal texture analysis. Hard coal (undeformed and cataclastic coal) exhibited higher DEN, GR, lower CAL, and lower AC than soft coal. The accuracy rate of coal texture identification was highest (97%) when the linear kernel function was applied, and the maximum training accuracy rate was achieved when the penalty parameter value of the linear kernel increased to 1. The results of verification with a newly cored CBM exploration well indicated that the SVM-based identification method was effective for coal texture analysis. With the increasing availability of data, this method can be used to distinguish hard and soft coal in a coal-bearing basin under numerous sample learning conditions.

Using 2D/3D seismic data and a large number of drilling and logging data and applying sequence stratigraphy, seismic sedimentology, and petroleum geology concepts, the characteristics of the sedimentary sequence of the forced regression have been analysed, the migration trajectory of the coastline have been reconstructed, the evolution model of the forced regression have been presented, and the significance for petroleum geology of the forced regressive sandbodies have been discussed. The falling stage systems tract (FSST) of the Zhujiang Formation present offlap high-angle oblique foreset reflection structure in the seismic profiles of the depositional trends and turbidite fan deposits with strong amplitude mound reflection structure are developed in the downdip direction of its front. The trajectory of migration of the shoreline shows a terraced downtrend in the direction of basin. The FSST is characterized by the shelf-edge delta without topset beds. The FSST was formed in the fall of relative sea-level. Five sets of foreset beds controlled by high-frequency relative eustatic were developed, therefore ordinal regressive overlap can be observed for the five sets of shelf-edge deltas in the depositional trends. The favourable reservoirs which were located close to the upper boundary of the falling stage systems tract and the basal surface of forced regression are sandbodies of the shelf-edge delta front and wave-dominated shoreface sands and the sandbodies of the turbidite fan. Those sandbodies favour the formation of lithologic oil–gas reservoirs by means of good trap sealing conditions, excellent oil–gas reserving performance, and effective oil source communication of fracture system.

In situ desorption gas measurement can be used to evaluate shale gas potential, sweet spot prediction, and production strategy optimization. However, gas contents and carbon isotope compositions of in situ desorption gas and the relationship to reservoir properties and shale compositions are not systematically studied from the actual production situation. In this study, 63 core shales of Wufeng-Longmaxi formation from the YY1 well in the eastern Sichuan Basin were subjected to TOC (total organic carbon), solid bitumen reflectance (Rb), maceral fractions of kerogen analysis, and X-ray diffraction (XRD) analysis to obtain shale compositions, and 10 selected samples were conducted on low-pressure N₂/CO₂ (N₂/CO₂GA), mercury injection capillary pressure (MICP), and field emission scanning electron microscopy (FE-SEM) tests to acquire reservoir properties. Meanwhile, 60 samples were also subjected to in situ desorption tests to obtain shale gas content, and 5 selected samples were used to investigate variations in shale gas compositions and carbon isotopes during the desorption process. Results indicated that transient rates of shale gas during desorption process are significantly controlled by desorption time and temperature. In terms of in situ desorption process, total gas is divided into desorbed gas and lost gas. Desorbed gas is mainly comprised of CH₄, N₂, CO_2, and C₂H_6, with desorption priorities of N₂ > CH₄ > CO₂ ≈ C₂H₆, which are consistent with their adsorption capacities. The δ¹³CH₄ values tend to become heavier during desorption process, varying from -37.7‰ to -16.5‰, with a maximum increase amplitude of 18.8‰, whereas the change of δ¹³C₂H₆ value, from -38.2‰ to -34.8‰, is minor. Desorbed gas shows carbon isotope reversals, due to that preferential desorption of ¹²C-CH₄ during desorption process results in δ¹³C value less negative in CH₄. The tested desorbed gas, lost gas, and total gas ranged 0.088 to 1.63 cm³/g, 0.15 to 3.64 cm³/g, and 0.23 to 5.20 cm³/g, respectively. Shale gas content, i.e., desorbed gas and lost gas, is controlled primarily by TOC content and organic matter (OM)-hosted nanometer-size pores. Clay mineral content is negatively correlated with shale gas content, due to that, clay mineral pores are more easily compacted during burial and occupied by water molecules. Compared with shale gas reservoirs in North America, the studied shale reservoir has high brittle mineral content and permeability, which is considered to have great potential of shale gas resource and to be the next commercial development zone in south China.

《地球科学》(ISSN 1000-2383;CN42-1874/P) (原刊名《地球科学——中国地质大学学报》ISSN 1000-2383;CN42-1233/P)，创刊于1957年，是由教育部主管、中国地质大学主办的综合性地学学术理论期刊，月刊，2015年改名为《地球科学》,她以反映中国地质大学教学与科研最新的、高水平的基础地质、应用地质、资源与环境地质及地学工程技术研究成果为主要任务。

Organic-rich shales were found in the Ordovician to Silurian Wufeng-Longmaxi Formation in northwestern Guizhou province, China, which has high shale gas content revealed by field measurement. Shale gas occurrence, free gas/sorbed gas ratio, and their influencing factors are crucial for shale gas exploitation strategy. Results indicated that the Wufeng-Longmaxi shales are dominated by type I kerogen, with total organic carbon (TOC) and equivalent vitrinite reflectance (eqvRo) of 0.77%–6.98% and 2.37%–2.53%, respectively. The total porosity and permeability are in the range of 1.23%–8.43% and – mD, respectively. FE-SEM observation and correlation analysis show shale porosity is dominated by organic matter (OM) pores, followed by interparticle (interP) pores related to brittle minerals. CH₄, derived from oil cracking, is the main component of shale gas, but its proportion is lower than that in Fuling and Weiyuan areas, probably due to the weak preservation condition. Desorption gas and lost gas determined by in situ desorption test are 0.42–1.54 cm³/g and 1.9–7.14 cm³/g, respectively, and Langmuir volume () from isothermal adsorption experiment is 1.63–4.78 cm³/g. Shale gas content is positively correlated with micropore volume, mesopore volume and TOC content but negatively correlated with macropore volume and clay mineral content, indicating that methane is preferentially stored in micropores (<2 nm) and mesopores (2–50 nm) related to OMs. By comparing actual total gas content with theoretical gas content, shale gas is considered to exist primarily in sorbed state, and the free gas proportion can increase with increased TOC content, due to that OM pores with larger sizes are also main space for free gas. Combined with the two methods, it can result in accurate calculations of shale gas reserves and free/sorbed gas ratio. Based on this understanding, a model of shale gas occurrence was proposed, which can provide a reference for shale gas exploitation in normal pressure areas.