A consensus has been reached through previous studies that organic matter (OM) pores are crucial to porosity in many shale gas reservoirs; however, their origins and types remain controversial. Here, we report the OM pore types hosted in algae, bitumen, graptolite and other fossil fragments in the Wufeng-Longmaxi Formations of the Sichuan Basin, Southwest China. Algae types mainly include multicellular algae, unicellular algae, etc. The OM pores in multicellular algae usually exhibit irregular, bubble-like, spherical and/or elliptical profiles, and their diameters vary between 300 and 800 nm. The shapes of the OM pores in unicellular algae are either irregular or oval, and the pores are hundreds of nanometres in size. The pores associated with solid bitumen are sporadic, isolated and variable in size, ranging from 500 nm to 3 μm. The pores in the graptolite, sponge spicule, radiolarian and other fossil fragments are much smaller and fewer. The pores may only have developed in the surface of the graptolite and bitumen by filling in the biological cavity of the sponge spicule. These new findings provide stronger evidence that multicellular algae are the main hydrocarbon generating organisms of OM pores development.

The total organic carbon (TOC) content of organic-rich shale is a key parameter in screening for potential source rocks and sweet spots of shale oil/gas. Traditional methods of determining the TOC content, such as the geochemical experiments and the empirical mathematical regression method, are either high cost and low-efficiency, or universally non-applicable and low-accuracy. In this study, we propose three machine learning models of random forest (RF), support vector regression (SVR), and XGBoost to predict the TOC content using well logs, and the performance of each model are compared with the traditional empirical methods. First, the decision tree algorithm is used to identify the optimal set of well logs from a total of 15. Then, 816 data points of well logs and the TOC content data collected from five different shale formations are used to train and test these three models. Finally, the accuracy of three models is validated by predicting the unknown TOC content data from a shale oil well. The results show that the RF model provides the best prediction for the TOC content, with R² = 0.915, MSE = 0.108, and MAE = 0.252, followed by the XGBoost, while the SVR gives the lowest predictive accuracy. Nevertheless, all three machine learning models outperform the traditional empirical methods such as Schmoker gamma-ray log method, multiple linear regression method and ΔlgR method. Overall, the proposed machine learning models are powerful tools for predicting the TOC content of shale and improving the oil/gas exploration efficiency in a different formation or a different basin.