Application of machine learning methods to forecast petrophysical properties in basalts of the Serra Geral Group: Implications for carbon storage

Carbon capture and storage (CCS) is an important technology to combat climate change, since it removes carbon dioxide emissions from major industrial sources, helping to reduce greenhouse gas concentrations while the world transitions to cleaner energy systems. In the implementation of CCS projects, petrophysical properties are crucial in determining the most suitable sites to store the CO₂. Therefore, this study applies machine learning techniques to forecast petrophysical properties (density, porosity, and permeability) in the basalts of the Serra Geral Group of the Paraná Basin, located in the State of Santa Catarina, Brazil. By applying machine learning algorithms (XGBoost, Gradient Boosting, and Random Forest) to 28 well log datasets, this research aims to overcome the limitations of traditional empirical methods that often fail to capture the complex variabilities in basalt formations and thereby determine the most suitable locations and sections for carbon storage. The applied machine learning models demonstrated considerable improvements in predictive performance compared to empirical methods from the literature, demonstrating the potential of machine learning to enhance the feasibility and reliability of CCS in basaltic formations. Furthermore, the interpolation of the well data indicated that the northern region of the Serra Geral Group in the State of Santa Catarina exhibits optimal conditions for geological storage. From 600 to 900 m, the basalts present suitable intervals ranging from 10 to 22 m thick, with density lows of almost 2.2 g/cm³, high peaks of 16.9% porosity, and permeability of 45.7 μD. Overall, this study advances CCS technology by applying machine learning to predict petrophysical properties in basalts with higher precision than traditional empirical methods. This approach provides an effective framework for delineating suitable storage and sealing sections, improving site selection, and cost-effective evaluation for CCS projects in the Paraná Basin.