My research spans two complementary directions in computational Earth science. For nearly 20 years I have developed higher-order finite element methods — mixed-hybrid and discontinuous Galerkin formulations — for multiphase compositional flow in fractured porous media. Applications include geological CO₂ sequestration, shale gas desorption and transport, and natural hydrogen reservoirs. My group holds four patent families with ten fully issued patents on subsurface hydrogen exploration and production, and has been funded by DOE, ARPA-E, the American Chemical Society, and industry partners.                                                                                                                                                               

More recently, I have built a parallel research program in AI-powered Earth observation, beginning during a 2020–2021 sabbatical and now institutionalized through the BuckAI Observatory, which I founded at OSU in August 2025. We apply deep learning and physics-informed neural networks to multi-modal satellite imagery — mapping rivers at sub-meter resolution, estimating coastal bathymetry from ICESat-2 and Sentinel-2, prospecting for natural hydrogen reservoirs via global semantic segmentation of surface features, and tracking deforestation and agroforestry dynamics in Sub-Saharan Africa. This work is funded by NSF, NASA, CGIAR, OSU, and industry, and has been covered in New Scientist and other international media.