AI Model Boosts Accuracy of Small Earthquake Measurements in Yellowstone

Written by Jeremy Werner

Jeremy is an experienced journalist, skilled communicator, and constant learner with a passion for storytelling and a track record of crafting compelling narratives. He has a diverse background in broadcast journalism, AI, public relations, data science, and social media management.
Posted on 08/11/2025
In News

Scientists are turning to artificial intelligence (AI) to tackle one of Yellowstone’s trickiest seismic challenges—accurately calculating the magnitudes of tiny earthquakes that often occur in rapid bursts.

 

Most of Yellowstone’s earthquakes are small and go unnoticed by humans, but they can provide important insights into the park’s geologic processes and earthquake hazards. Traditionally, the University of Utah Seismograph Stations (UUSS) measures magnitudes by averaging readings from multiple seismic stations. However, when earthquakes happen close together—especially during swarms—signals can overlap, making it difficult to determine individual magnitudes. Roughly 2% of Yellowstone earthquakes end up without a calculated magnitude, recorded as “-9.99” in the catalog.

 

A new machine learning approach aims to change that. Instead of relying solely on raw seismic data, researchers train models to recognize key features of earthquake signals—such as amplitude—and pair them with event locations to estimate magnitudes. By building a separate model for each seismic station in the Yellowstone network, scientists can draw on more stations and more measurements, increasing available data for calculations up to fourfold.

 

The AI-assisted method can better handle earthquakes that occur close together and make use of data from stations not used in traditional calculations. However, it’s not without limits. The models perform best on earthquakes similar to those in the training set, and results can be less reliable in areas with few historical examples.

 

Researchers say the new method will complement—not replace—traditional approaches, which work well in most cases. The goal is to fill in gaps where conventional methods struggle, especially for swarms of small events.

 

Yellowstone’s dense seismic activity makes it an ideal testing ground for this cutting-edge technique, which could improve earthquake monitoring in other complex regions worldwide.

 

Need Help?

 

If you have questions or concerns about how to navigate the global AI regulatory landscape, don’t hesitate to reach out to BABL AI. Their Audit Experts can offer valuable insight, and ensure you’re informed and compliant.

 

Subscribe to our Newsletter

Keep up with the latest on BABL AI, AI Auditing and
AI Governance News by subscribing to our news letter