Digital Twins and Generative AI in Subsurface Geophysics
Seismic Soundoff23 Okt 2025

Digital Twins and Generative AI in Subsurface Geophysics

"Generative modeling is a game-changer. We can now capture high-dimensional statistics that we could never have captured in the past." Felix Herrmann explains how digital twins and generative AI are reshaping subsurface geophysics. He highlights the importance of open-source tools, multimodal data, and uncertainty-aware models for better decision-making in energy and storage projects. By combining physics with AI, his work shows how geophysics can move beyond silos and create more reliable and efficient solutions. KEY TAKEAWAYS > Digital twins informed by multimodal data can reduce uncertainty and improve reservoir management. > Open-source tools and agreed benchmarks are essential for accelerating innovation in geophysics. > Combining physics-based models with generative AI creates robust, practical solutions for complex subsurface challenges. Read Felix's article in The Leading Edge, "President's Page: Digital twins in the era of generative AI," at https://doi.org/10.1190/tle42110730.1. GUEST BIO Felix J. Herrmann earned his Ph.D. in engineering physics from Delft University of Technology in 1997, followed by research appointments at Stanford and MIT. He later joined the University of British Columbia faculty in 2002 and moved to the Georgia Institute of Technology in 2017, where he is the Georgia Research Alliance Scholar Chair in Energy with cross-appointments across multiple schools. Dr. Herrmann leads a cross-disciplinary research program in computational imaging, spanning seismic and medical applications, and is recognized for innovations in machine learning, optimization, and high-performance computing that have reduced costs in seismic data acquisition and imaging. A past SEG Distinguished Lecturer and recipient of the SEG Reginald Fessenden Award, he directs the Seismic Laboratory for Imaging and Modeling and co-founded the Center for Machine Learning for Seismic (ML4Seismic) to advance AI-assisted seismic imaging and reservoir monitoring through industry collaboration.

Episoder(301)

OTC 2026 Emerging Leaders on the Future of Offshore Energy Collaboration

OTC 2026 Emerging Leaders on the Future of Offshore Energy Collaboration

“Nothing can happen in a vacuum anymore. We need to have the developers talking to the geoscientists, talking to the environmental professionals.” Two OTC 2026 Emerging Leaders share why the future o...

23 Apr 28min

Why Seismic Acquisition Is Making a Quiet Comeback

Why Seismic Acquisition Is Making a Quiet Comeback

"What has happened in the last few years is exploration overall has taken a little bit of a backseat. So they are starting to relook at seismic acquisition to explore new areas and solve more complex ...

16 Apr 27min

From Hype to Reality: What Machine Learning Can Actually Do

From Hype to Reality: What Machine Learning Can Actually Do

“You can't be stationary in this field. I've never seen anything like it." Machine learning is changing geophysics faster than most people expected, but knowing what actually works versus what is hyp...

9 Apr 27min

DAS and Seismic Innovation: What Geophysicists Need to Know

DAS and Seismic Innovation: What Geophysicists Need to Know

"For early career geophysicists, I think it's really important to understand that DAS is going to have a unique role in reservoir management, be it onshore or offshore." Distributed acoustic sensing ...

19 Mar 27min

From Oil and Gas to Offshore Wind: Why Everyone Meets at OTC

From Oil and Gas to Offshore Wind: Why Everyone Meets at OTC

“And what the world really needs is this flexibility on energy and the agility to ensure access and affordability. And that's where steering the offshore energy business is really critical.” Offshore...

12 Mar 29min

Applying Near-Surface Geophysics to Agricultural Challenges

Applying Near-Surface Geophysics to Agricultural Challenges

“It’s not only about discovering resources, but about safeguarding them and safeguarding our future. When you understand the subsurface, you understand the foundation of food security, water security,...

5 Mar 26min

Inside the Workflow - Unsupervised Machine Learning for Seismic Interpretation

Inside the Workflow - Unsupervised Machine Learning for Seismic Interpretation

“The major pitfall of machine learning of any kind is to be overly confident in the results. We run the risk of garbage in gospel out.” This discussion offers a rare chance to go a little deeper into...

19 Feb 35min

Why High-Performance Computing Is No Longer Optional in Geophysics

Why High-Performance Computing Is No Longer Optional in Geophysics

“I think that for geophysicists out there, people need to realize that it's an integrated career path. You can't separate the geophysics from the HPC anymore, if we ever did to begin with.” High-perf...

12 Feb 21min

Populært innen Vitenskap

fastlegen
tingenes-tilstand
jss
forskningno
liberal-halvtime
rekommandert
sinnsyn
tomprat-med-gunnar-tjomlid
rss-paradigmepodden
pod-britannia
villmarksliv
rss-inn-til-kjernen-med-sunniva-rose
fjellsportpodden
tidlose-historier
nevropodden
rss-overskuddsliv
nordnorsk-historie
dekodet-2
diagnose
rss-rekommandert