Organic-matter content prediction based on the random forest algorithm: application to a Lower Silurian shale-gas reservoir

摘要

Organic-matter content (OMC) is an important property of a shale-gas reservoir. Quantitative seismic interpretation of OMC can be difficult because OMC can be determined by various properties of shales. In this paper, we propose an OMC seismic interpretation method based on the random forest algorithm (RFA). Real log data is used to evaluate the viability of this method. One major advantage of this method is that the prediction result is stable and has high accuracy even if the types and quantities of predictor are small. We also applied the new method to a real 3D seismic dataset of a Lower Silurian shale-gas reservoir formation in the south of Sichuan Basin. A comparison between seismic interpretation result and well logs demonstrates the feasibility of this method.