Analysis of Structural Position of Carlin-Type Gold Deposits with Lineament Analysis of Remote Sensing Data Using pyLEFA Software

Previous research on Carlin-type gold deposits in North America and China has revealed peculiarities in their genesis, distribution, and prospectivity. However, pinpointing these deposits within known ore districts and prospective areas is a complex and resource-demanding task. Studying the structural and geological characteristics of Carlin-type deposits in areas with a longer history of exploration using machine learning techniques is crucial, especially considering the potential for discovering Carlin-type deposits in Russia. Crustal fracturing fields detected in space imagery and digital relief models can serve as a foundation for prospectivity mapping of Carlin-type deposits, even without evidence of magmatic sources of ore matter. The detection of disjunctive features, observed as linear elements (lineaments) in remote sensing images of the Earth, allows for a quantitative description of the Earth’s crust permeability to ore-bearing magmas and fluids. This can be accomplished using open source pyLEFA software. Optical detection methods facilitate this process, while the assessment of heterogeneity in the distribution of fracture field parameters is achieved using unsupervised learning and classification. Machine learning based on datasets produced with pyLEFA enables the assessment of the contribution of predictor variables to the result. The knowledge acquired can be applied to areas with the potential for discovering Carlin-type deposits.