Beneath Maine’s salt marshes, an invisible transformation is constantly underway. Water moves through soil and rock. Chemicals react. Tiny microbes living underground help determine whether minerals dissolve or form, how fluids move and whether pollutants break down or spread.
Existing computer models used to guide decisions about energy infrastructure and contamination often treat microbes as fixed and unchanging, even though they constantly adapt to their surroundings. For decades, incorporating molecular biology information into those models has remained one of the field’s biggest unsolved challenges.
Now, a research team led by University of Maine professors Jiaze Wang and Amanda Albright Olsen is aiming to close that gap after receiving funding through the U.S. Department of Energy’s inaugural Genesis Mission .
Using artificial intelligence, the UMaine researchers are working to incorporate molecular biology information into the physical models scientists rely on to predict what’s happening underground. The challenge is one of scale. Microbial genomic data is extraordinarily detailed but has historically been too complex to integrate into large-scale underground models.
The project will analyze microbial genetic information using AI to identify the biological processes that most strongly influence underground chemical reactions. Those insights will then be integrated into existing subsurface models, allowing scientists to better simulate how microbes, minerals and groundwater interact in changing environments.
By bringing molecular biology into subsurface models, the researchers hope to improve predictions that support groundwater management, environmental remediation, responsible critical mineral development and energy infrastructure planning. Improved models could help scientists and decision-makers protect groundwater, clean up contaminated sites and plan energy projects with greater confidence.
The Genesis Mission award
The Genesis Mission is a historic new national initiative led by the U.S. Department of Energy, which is building the world’s most powerful integrated science discovery platform. By uniting government, industry, academia and philanthropy, it is accelerating breakthroughs in energy, scientific discovery and national security through a new platform that combines AI, supercomputing, quantum systems and advanced scientific instruments.
UMaine was selected as part of the inaugural cohort of Genesis Mission-funded research teams, making it one of the first universities participating in the innovative initiative. The awards were announced Wednesday at a U.S. DOE event in Washington, D.C.
The UMaine-led project was one of 278 selected by the DOE from more than 5,000 proposals. It was also the only project selected from Maine.
UMaine is leading the project in partnership with the University of Delaware and Argonne National Laboratory. The collaboration brings together expertise in bioinformatics, computational biology, AI, subsurface modeling and geochemistry.
What it means for UMaine
The project builds on UMaine’s nationally recognized strengths in environmental research and computational modeling while expanding collaborations across disciplines and institutions.
The grant funds a graduate student and postdoctoral researcher at UMaine, both of whom will train directly on the project. That hands-on experience reflects UMaine’s mission as the state’s R1, learner-centered university, giving students opportunities to contribute to nationally significant research while working alongside leading scientists and preparing to lead and innovate in the state’s workforce.
“We are proud to be part of the inaugural Genesis Mission and to see our researchers contributing to this new era of scientific discovery,” President Joan Ferrini-Mundy said. “This project exemplifies what makes UMaine such a special place to learn and innovate. Our students have the opportunity to work alongside nationally recognized experts on projects that address some of the world’s most pressing challenges while developing solutions that can make a difference here in Maine and well beyond.”
Giovanna Guidoboni, interim vice president for research, said the award reinforces UMaine’s leadership in interdisciplinary research.
“At UMaine, leading research that addresses complex, real-world challenges is at the heart of our mission,” said Guidoboni, also the dean of the Maine College of Engineering and Computing. “This project demonstrates the impact of bringing together expertise across disciplines while creating meaningful, research-based learning opportunities for our students. What we learn matters, but how we apply that knowledge to improve lives and communities is what defines us.”
AI and environmental science converge
Wang, the principal investigator, said the project brings together disciplines that have historically worked in isolation.
“This project is trying to bring together concepts from very different disciplines and unite them,” Wang said. “We know the subsurface is a combination of geology and biology, and our goal is to bring that expertise together.”
The challenge, she said, extends to the microbes themselves.
“The microbes are like magicians,” Wang said. “They can transform the subsurface in ways we don’t fully understand yet, and we know so little about them.”
For Olsen, the co-principal investigator, the use of AI and the Genesis Mission award represents a turning point.
“This is a problem I didn’t think we would solve,” she said. “Having this project, where I can actually see a path to solving something people have worked on for my entire career, that feels like a paradigm shift in how we do this kind of work.”
Starting with the coast
In this first phase, the team is focusing on coastal systems, including salt marshes like the ones that define much of Maine’s shoreline. The team will test its work using real-world data from coastal wetlands from Lake Erie to the Chesapeake Bay.
Coastal environments change quickly, making it easier to observe how microbes, minerals and groundwater interact than in slower-changing systems. They also face challenges such as saltwater intrusion, groundwater contamination and declining water quality, making them an ideal testing ground for the team’s approach.
Salt marshes are systems people in Maine already understand and value, Olsen said, making them a natural starting point for tools that could eventually benefit land-locked communities.
Applications far beyond the coast
Although the work begins on coastlines, the researchers say the tools they’re developing could be applied to subsurface systems around the world. The same modeling approach could help determine where it’s safe and healthy to build energy infrastructure, whether contamination found at a site will break down or spread and how critical minerals can be mined more responsibly.
The work could also inform decisions about using the subsurface in other rapidly changing environments, including polar regions.
By the end of the project, the team plans to deliver:
A library of microbial genetic information for subsurface physical models
An upgraded version of a widely used subsurface simulation model
A faster AI model emulator capable of running large-scale subsurface predictions
“Any system where you’re trying to understand whether a soil is healthy, or whether water is contaminated, requires understanding how these reactions couple together,” Olsen said. “That’s true almost anywhere on Earth’s surface.”