Ground-truth-calibrated onshore and offshore subsurface infrastructure image from deep-learning-based 3D inversion of magnetic data

Abstract In this study, we demonstrate the application of deep-learning-based 3D inversion of magnetic data to image subsurface infrastructure. We highlight results from two case studies: an onshore survey at Texas A&M University's Rellis Campus site and an offshore survey in the northern Gulf of Mexico (GOM), off the coast of Louisiana. The onshore case utilized drone-acquired magnetic data to map buried utilities before construction. The offshore case employed a boat-towed magnetometer approximately 3.5 m above the seafloor to locate oil well conductors disrupted by Hurricane Ivan in 2004 and presently buried under 35 to 45 m of sediment. The inversion results at the Rellis site were validated against excavation data, revealing strong agreement in target location and depth (within 17 cm). In the GOM survey, the artificial intelligence (AI)-driven inversion successfully extended conductor imaging beyond the limits of acoustic methods, providing critical information on conductor geometry near the well conductor bay. This work highlights the effectiveness of AI-driven inversion techniques in enhancing subsurface imaging, offering cost-effective and scalable solutions for applications in utility mapping, environmental monitoring, and hazard assessment. The results demonstrate that AI-based workflows can be adapted to various geophysical settings, providing new opportunities for high-resolution imaging of complex subsurface features in onshore and offshore environments.