Data used for the three case studies to validate 'A Data-Driven Approach to Geological Modeling Using Geophysical Proxies' (Hernández, Alberdi & Goti, 2026)

These data are the ones used for the three case studies presented to validate the proposed data-driven, multivariate method of the paper entitled ‘A Data-Driven Approach to Geological Modeling Using Geophysical Proxies’, authored by Heber Hernández, Elisabete Alberdi and Aitor Goti. The first case, entirely synthetic and in 2D, focuses on explaining the novel aspects of the DDIM method in detail, highlighting its ability to integrate secondary information and generate more accurate geological models across six different spatial sampling scenarios. The second case, also synthetic and in 2D, demonstrates how incorporating geophysical information as a proxy significantly improves the modeling of mineralized and barren rock identified in boreholes, overcoming the limitations of traditional and reducing entropy. Finally, the third case, developed in 3D, addresses the modeling of oxidized and sulfidized minerals in a copper deposit located in northern Chile, using geophysical information to guide geological interpretation and reduce uncertainty in areas with limited sampling. These application cases, developed in Python, demonstrate the advantages of a data-driven, multivariate method in contrast to traditional implicit methodologies available in most mining software packages, such as Leapfrog by Seequent, Vulcan by Maptek, and Studio Geo by Datamine.