Skip to main content Skip to secondary navigation
Main content start

AI model reveals hidden earthquake swarms and faults in Italy’s Campi Flegrei

A new AI model detects thousands of previously unseen earthquakes in near real time, helping scientists understand changes in an Italian volcanic area where earthquakes have been intensifying since 2018.

Overhead view of a shallow volcanic crater near an urban area by the coast
View of the Solfatara crater, part of the densely populated Campi Flegrei volcanic area. (Image credit: Alessandro Fedele)

Scientists are using artificial intelligence to understand escalating unrest in Italy’s Campi Flegrei, a volcanic area that is home to hundreds of thousands of people.

Like adjusting a camera lens so a blurry image becomes clear, the new approach makes it possible for researchers to identify earthquakes that previous tools could not pick out from massive sets of seismic monitoring data. 

The research, a collaboration between Stanford University, Italy’s National Institute of Geophysics and Volcanology (INGV) - Osservatorio Vesuviano, and the University of Naples Federico II, reveals four times as many earthquakes than earlier tools had detected and pinpoints previously unknown faults. 

People swimming near Baia beach, Pozzuoli in Campi Flegrei, Italy. The Solfatara crater and fumarolic field are in the background. (Image credit: Xing Tan)

Knowing the location and length of a fault – the space between two blocks of rock that move and cause earthquakes – can help researchers determine the range of magnitudes of future quakes. This information is critical for informing residents and city planners of potential risks and mitigation options. 

The study, published in Science on Sept. 4, uses an AI model created at Stanford to provide precise earthquake location and magnitude information in near real time. According to the researchers, the promising results in Campi Flegrei suggest the system could be adapted to improve understanding of other areas with seismic monitoring systems, such as Santorini, Greece, which experienced a prolonged earthquake swarm at the beginning of the year. Rapidly pinpointing an earthquake’s source during sudden, intense seismic activity is critical for an effective emergency response.

“Seismicity could change at any time, and that may be the most important thing about this study: this capability of getting a clear view is now operational,” said study co-author Greg Beroza, a geophysics professor in the Stanford Doerr School of Sustainability. “INGV is now running the tool by themselves as needed, so it should be helpful for scientific response and ultimately public response if something changes.”

A history of unrest

Campi Flegrei is an active volcano located within the densely populated Neapolitan area. This volcanic region, which is home to more than 500,000 people, has experienced episodes of unrest dating back to the late 1950s. The last period of unrest started in 2005, with a significant increase of the seismicity in 2018, including five earthquakes above a magnitude 4 in the first eight months of 2025. The new research expands the seismicity recorded by monitoring stations from 2022 to 2025 from about 12,000 to more than 54,000 earthquakes. 

These long faults suggest that an earthquake in the magnitude 5 range is not out of the question

Bill Ellsworth Co-director of the Stanford Center for Induced and Triggered Seismicity and Emeritus Professor of Geophysics

This image shows earthquakes that occurred in the Pozzuoli-Campi Flegrei region about 7 miles west of central Naples from 2022 to 2025 overlaid on Google Earth satellite imagery. (Image credit: Xing Tan)

The data revealed two faults converging under the town of Pozzuoli west of Naples, which has been continuously monitored since the early 1980s, when unrest caused the land to rise more than 6 feet and more than 16,000 earthquakes prompted evacuation of 40,000 residents. 

“These long faults suggest that an earthquake in the magnitude 5 range is not out of the question,” said study co-author Bill Ellsworth, who co-directs the Stanford Center for Induced and Triggered Seismicity. “We’ve known that this is a risky place for a long time, since the ’80s when part of the city was evacuated, and now we’re seeing for the first time the geologic structures that are responsible.”

The stakes are high for understanding the complex natural system of Campi Flegrei, which in the past 40,000 years has produced two of the largest eruptions in Europe. However, because of the potential to endanger people and damage buildings and infrastructure, “one of the biggest concerns in the short term in Campi Flegrei is not an eruption, but a moderate earthquake at shallow depth,” Beroza said. 

Under pressure

Campi Flegrei is an 8-mile-wide caldera, a massive depression formed by major volcanic eruptions about 39,000 and 15,000 years ago. In addition to eruptions, the caldera experiences uplift and subsidence – rising and sinking of the land called bradyseism. 

“Previously, the structure of seismicity in the caldera was indistinct, and now we’ve seen a very thin and well-marked ring fault that is consistent with the surface features, especially offshore, and also with the area that’s being uplifted,” Beroza said. 

A distinct structure of seismicity can be seen based on earthquakes that occurred in the whole Campi Flegrei caldera about 7 miles west of central Naples from 2022 to 2025. The data is overlaid on Google Earth satellite imagery. (Image credit: Xing Tan)

“Our Italian colleagues were surprised to see the ring so clearly,” added lead study author Xing Tan, a geophysics PhD student in Beroza’s lab. “They expected to see something in the south where previous data had revealed scattered seismicity, but in the north, they’d never seen it so clearly.”

The research suggests overall inflation of the caldera is driving earthquake activity through pressure. The study authors did not observe any evidence for the upward migration of magma, which reduces concern over the short term that the area will experience a magmatic eruption, according to the study.

Media Contacts

Gregory Beroza

Stanford Doerr School of Sustainability

William Ellsworth

Stanford Doerr School of Sustainability

Xing Tan

Stanford Doerr School of Sustainability

Anna Tramelli

Italy’s National Institute of Geophysics and Volcanology (INGV) - Osservatorio Vesuviano

Sergio Gammaldi

Italy’s National Institute of Geophysics and Volcanology (INGV) - Osservatorio Vesuviano

Warner Marzocchi

University of Naples Federico II

Danielle T. Tucker

Stanford Doerr School of Sustainability
650-497-9541

Explore More

  • Tiny movements in Earth’s outermost layer may provide a Rosetta Stone for deciphering the physics and warning signs of big quakes. New algorithms that work a little like human vision are now detecting these long-hidden microquakes in the growing mountain of seismic data.