Subsurface Energy Production and Storage for a Sustainable Energy Future

Subsurface energy production and storage play a central role in the energy transition. However, subsurface systems are inherently complex. Predicting and optimizing these systems remains challenging due to coupled multiphysics processes, multiscale behavior, and substantial uncertainty. In this talk, I will present three lines of research spanning reservoir stimulation, underground hydrogen storage (UHS), and geological CO₂ storage (GCS). For reservoir stimulation, I will introduce a physics-based hydraulic fracturing model that simulates Lagrangian proppant transport in propagating three-dimensional fractures at the field scale, and show how it can inform proppant design for applications such as enhanced geothermal systems. For UHS, I will demonstrate how reservoir simulation provides mechanistic insight into salt precipitation during hydrogen cycling, and how deep-learning surrogates can accelerate prediction to support storage design and operational decisions. For GCS, I will present an end-to-end workflow for stochastic modeling of CO₂ migration across and along faults, quantifying and reducing uncertainty in fault behavior relevant to secure sequestration. Together, these topics illustrate how advances in numerical modeling and machine learning can enable more effective subsurface energy production and storage toward a resilient and sustainable energy future.

Speaker: Shaowen Mao, Massachussets Institute of Technology