arXiv:2605.30541v1 Announce Type: new Abstract: Full waveform inversion (FWI) is the gold standard for subsurface imaging, with applications from carbon sequestration to energy and mineral exploration to earthquake hazard assessment. Machine learning approaches to FWI need field-scale, geologically diverse, and physically realistic training data, but existing resources such as Marmousi, SEAM, and OpenFWI fall short on spatial extent, temporal extent, geological diversity, and physical realism. We address these limitations with SubsurfaceGen, a GPU-accelerated generator for 3D velocity models a
Advances in AI, particularly machine learning, are increasingly being applied to computationally intensive scientific fields, and the need for high-quality, realistic training data is a continuous bottleneck.
This development addresses a critical need for advanced AI applications in subsurface imaging, enabling more accurate and efficient exploration for resources and assessment of geological risks.
The availability of a GPU-accelerated generator for 3D velocity models significantly improves the scalability and realism of training data for Full Waveform Inversion (FWI), potentially accelerating FWI adoption across industries.
- · Energy exploration companies
- · Mineral exploration companies
- · Geological survey agencies
- · AI/ML research in geophysics
- · Traditional, less data-intensive geological modeling methods
- · Companies relying on limited, older geological datasets
Improved accuracy and efficiency in identifying subterranean resources and geological hazards using AI-driven FWI.
Reduced costs and environmental impact of exploration due to more precise targeting, potentially increasing resource availability.
Accelerated development of AI-driven 'digital twins' for complex subsurface environments, leading to predictive maintenance and optimized extraction strategies.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG