Fitzroy Minerals Inc. and Fleet Space Technologies have completed an integrated seismic and machine-learning targeting study at the Buen Retiro Copper Project in Copiapo, Chile, generating a ranked portfolio of nine previously untested exploration targets. The announcement matters because it represents one of the first applications of advanced 3D subsurface imaging combined with machine learning to identify blind copper targets beneath extensive alluvial gravel cover, a setting that has historically frustrated conventional exploration methods. If successful, the approach could establish a cost-effective discovery tool in one of Chile’s most prolific copper-producing regions.
The ML targeting combined 3D Ambient Noise Tomography (ANT) data with prior geophysical surveys, geological models, and approximately 24,400 copper assays from 140 drill holes. Fleet’s regional and infill ANT surveys used natural seismic noise to build a 3D picture of the rock beneath the surface, mapping the project’s deeper geology well beyond the depths of current drilling. The model identified nine target areas where sub-surface characteristics closely resemble the signature of rocks hosting Buen Retiro’s known copper mineralization, all within approximately 400 metres of surface. A summary of the four datasets and copper assay results used in the new ML targeting is available Click Here to see Table 1 - Summary of the 4 datasets and copper assay results used in the new ML targeting.
The implications of this announcement extend beyond the immediate drill program. Merlin Marr-Johnson, President and CEO of Fitzroy, noted that Buen Retiro shares many geological characteristics of major iron oxide copper-gold (IOCG) deposits at Candelaria and Mantoverde, yet the sulphide potential of the wider property has barely been tested. To date, the focus has been on oxide copper mineralization and its potential to support a low-capital-intensity, near-term production operation. The new ML tool allows Fitzroy to test the exploration optionality of a large IOCG system, prioritizing blind targets beneath gravel cover that were previously difficult to identify. If the ML targeting proves successful, it could create significant value for shareholders by enabling a cost-effective discovery path.
Fitzroy has already started a 5,000-metre drilling program using diamond and reverse circulation rigs. As drill logs and assay results come in, the company and Fleet will use the new data to update the ML model, so each new drill hole serves three purposes: testing a specific target, improving or validating the ML model, and helping to inform the next drilling decision. The targets were refined by excluding previously mined ground, mapped gravel cover, and areas within 200 metres of existing drill holes, focusing on ground with genuine exploration value.
The technical context is important: the new exploration domains are conceptual targets that have not yet been tested by drilling. They do not confirm the presence, grade, size or continuity of mineralization. The reliability of the model depends on the quality and coverage of the underlying data and tends to decrease with distance from existing drill holes and with depth. For this reason, the targets are intended to be used alongside conventional geological interpretation, not in place of it. The news was distributed through NEWMEDIAWIRE.
