Study of the Influence of Non-Deposit Locations in Data-Driven Mineral Prospectivity Mapping: A Case Study on the Iskut Project in Northwestern British Columbia, Canada

The accuracy of data-driven predictive mineral prospectivity models relies heavily on the training datasets used. These are usually trained using data for “known” deposit locations as well “non-deposit” that based randomly generated point patterns. In this study, related to Seabridge Gold Inc Iskut project, an epithermal Au in northwestern British Columbia (BC), Canada, used test utility modeling. input spatial dataset is comprised mostly publicly available data. Data 18 vein and known occurrences (KMO) obtained from BC Geological Survey’s MINFILE repository selected locations. A total eleven sets non-deposit (NDL) were also created, including one set non-prospective KMO deposits ten random Given scale most recorded property type. Hence, they could not be a selection criterion. Data-driven potential forest (RF) algorithm multiple sets. comparison RF demonstrated generates more accurate predictions than pattern. produced maps delineated areas with higher discovery potential, which matched viable targets Au-Cu epithermal-porphyry system identified through previous Inc. (Toronto, ON, Canada) field reconnaissance drilling programs.