Learning the Earth's Hidden Structure with Sparse Observations and Subsurface World Models

Understanding the structure of the Earth’s subsurface is fundamentally an inverse problem with sparse direct observations and abundant indirect measurements. Drillholes provide precise but extremely limited information, while geophysical signals such as gravity, magnetics, and seismic data provide broader but indirect constraints on the underlying geology.

In this talk, I introduce the concept of subsurface world models, generative AI systems trained to learn the statistical and physical structure of the Earth from large ensembles of synthetic geological models and geophysical simulations. These models combine sparse hard constraints such as drilling and geochemistry with dense indirect observations from geophysics to infer plausible three-dimensional geological structures while quantifying uncertainty.

Such models offer a new approach to mineral exploration and resource discovery. More broadly, subsurface world models provide a framework for reconstructing hidden planetary interiors from limited observations, with potential applications in geohazard assessment, geothermal exploration, carbon storage, and planetary exploration.

Speaker: Gerrit Olivier, Fleet Space Technology