From Data to Discovery: Applying Artificial Intelligence to Mineral Resources

Natural Resources Canada (NRCan) is uniquely positioned in the impact assessment ecosystem. The department plays a seminal role in the assessment of major projects by providing scientific expertise and leading the delivery of the Open Science and Data Platform (OSDP). The OSDP is a single-window digital tool that enables open access to federal, provincial and territorial government science, data and regulatory information.

From its unique vantage point, NRCan is exploring how artificial intelligence (AI) can transform the way scientific and geospatial information is discovered, interpreted and applied to streamline and improve assessment processes. AI technologies offer opportunities to unlock previously inaccessible, fragmented or difficult-to-use information and to enhance environmental monitoring and modelling. By connecting open data, scientific expertise and emerging AI capabilities, NRCan is turning data into actionable knowledge, supporting more efficient impact assessment and accelerating innovation in mineral exploration. This work also underscores the importance of data quality, interoperability, model interpretability and responsible AI as foundations for trusted, sustainable digital transformation.

Ambient Noise Tomography (ANT) has emerged in recent years as an increasingly relevant geophysical tool for mineral exploration. Initially perceived as an emerging or experimental approach, ANT has undergone rapid methodological and operational evolution, enabling it to meet the growing demands of modern exploration today.

This presentation highlights the dynamic and evolving nature of ANT and its growing role in building coherent geological models at the property scale. Thanks to major advances in sensor array design, processing algorithms, and inversion schemes, ANT now enables three-dimensional imaging of large subsurface volumes, with improved resolution, greater depth of investigation, and turnaround times compatible with exploration decision-making cycles.

Through real case studies, the presentation illustrates how ANT contributes to a better understanding of geological architecture and the structural controls on mineralization, in both greenfield and brownfield contexts. Integrating ANT with other geophysical datasets, such as gravity and magnetics, through joint or constrained inversions, helps reduce interpretation ambiguities and produce more robust, geologically coherent models.

Finally, the presentation discusses the future directions of ANT toward integrated geological solutions, combining multi-physics, geological modeling, and AI-assisted interpretation.