The subsurface resources covered by the SEL deal with from traditional resources concepts such as petroleum resources to underground space itself such as compressed air storage and carbon dioxide storage. For safe and efficient use of the subsurface resources, researches to investigate unseen underground spaces reliably and to reduce their uncertainty through data integration are essential.

Our lab tries to build reliable subsurface model to understand and predict behavior of geofluids such as shale gas and injected CO2. For efficient and reasonable decision making, machine learning algorithms couples with conventional methods in petroleum industry such as geostatistics, reservoir simulation, and data assimilation.