Machine-Assisted Granite Mapping in the Archaean Yilgarn Craton

The Challenge

Much of the Yilgarn’s remaining mineral potential lies beneath cover, where exploration is costly and high-risk. Traditional mapping methods are subjective and limited by surface data, creating a need for consistent, scalable approaches to accurately classify granites and target prospective mineral systems beneath cover. 

Key Findings 

Machine learning can successfully combine large geophysical datasets to consistently map granites across the entire Yilgarn Craton. This process identified new geological patterns, refined existing boundaries, and revealed hidden greenstone belts beneath cover. These results improve understanding of the region and highlight new areas with potential for gold, lithium and rare earth mineral exploration. 

Benefits to WA

By making the mapping of granite geology across the Yilgarn more objective and consistent, the outputs of this work will help geologists target mineral exploration more efficiently and with less risk. This increased efficiency will support new discoveries under cover, reduce unnecessary environmental disturbance, and strengthen Western Australia’s position as a global leader in exploration technology. 

Report DOI

DOI: 10.71342/560531747294