UC selected for federal initiative to use AI to advance scientific discovery

Led by Reza Soltanian, professor of subsurface energy and hydrogeology in UC's Department of Geosciences, the multidisciplinary project will develop next-generation AI-enabled scientific workflows and digital twins that integrate geological information, advanced geophysical imaging, field observations, physics-based simulations and AI surrogate models to characterize and predict complex subsurface systems.

The U.S. Department of Energy announced Wednesday that Soltanian's project is among 278 awards selected from more than 5,000 proposals submitted in the project it calls Genesis Mission.

“America has no shortage of bold ideas or talented scientists, and the response to the Genesis Mission proves that,” U.S. Sec. of Energy Chris Wright said. “The 278 projects selected today represent the very best of our nation’s scientific enterprise.”

During Phase I, the team will demonstrate the framework at UC’s Theis Environmental Monitoring and Modeling Site, an extensively instrumented field research facility directed by Soltanian along the Great Miami River corridor. The team will use AI to infer multiscale subsurface architecture and integrate geological information, field observations, physics-based models and surrogate models into an interactive digital twin for predicting flow, transport and biogeochemical processes. The work has implications for contaminant remediation, nutrient cycling and groundwater sustainability.

In Phase II, the team plans to expand the framework beyond UC's groundwater observatory by incorporating additional datasets, field sites and scientific applications. The broader goal is to create a scalable AI-enabled framework supporting research in critical minerals, geothermal energy, carbon and hydrogen storage, and other subsurface energy and environmental challenges.

“This award demonstrates the power of Discovery in Action and the extraordinary work being done by our faculty and research partners,“ UC College of Arts and Sciences Dean James Mack said.

“By bringing together expertise in geosciences, artificial intelligence, advanced modeling and extended reality, Dr. Soltanian's team is developing innovative new approaches to understanding complex subsurface systems,“ he said. “The impact of this work extends far beyond the laboratory, opening new opportunities for scientific discovery while helping address issues that matter to communities around the world.“

“This project is about transforming how scientists characterize and predict complex subsurface systems,” Soltanian said. “Rather than analyzing geological information, geophysical measurements, field observations and numerical simulations separately, we are developing an integrated AI-enabled scientific workflow that brings these different sources of information together into a unified framework.

“One of the most exciting aspects of the project is that we are developing a general scientific framework that can be adapted to many different subsurface challenges rather than addressing only a single application,” he said.

Tang will lead development of the project’s extended-reality interface, allowing researchers to interactively explore the spatial datasets and model predictions generated by the AI-enabled workflow.

“My primary role will be to develop an AI-enabled digital twin that allows scientists to interactively explore spatial data generated from surrogate predictions,” Tang said.

“We believe that extended reality and interactive environments are essential for keeping the human in the loop in the age of AI,” Tang said.

National news outlets such as Politico highlighted the new initiative.

Featured image at top: UC Professor Reza Soltanian's project was selected for a new Department of Energy initiative to use artificial intelligence to speed scientific breakthroughs. Photo/Andrew Higley/UC