The intense series of over 25,000 earthquakes that rattled the Greek islands of Santorini, Amorgos, and Anafi in early 2025 were caused by a massive surge of molten rock moving horizontally through deep underground channels, scientists have determined. The swarm, which included tremors exceeding magnitude 5.0, sparked fears of a volcanic eruption or a larger earthquake – reminiscent of the devastating 1956 quake in the same region.
How the Swarm Was Unlocked: AI and Physics Combined
Researchers from University College London (UCL) and elsewhere used an innovative approach blending physics with artificial intelligence to pinpoint the cause. Instead of relying on traditional seismic monitoring, they treated each tremor as a virtual sensor, then used AI to analyze patterns across the entire swarm.
This method allowed them to create a detailed 3D model of the Earth’s crust around Santorini, revealing that magma from beneath both Santorini and the Kolumbo underwater volcano had moved roughly 20 kilometers horizontally through a 30-kilometer-long channel over three months. The volume of magma involved was estimated to be enough to fill 200,000 Olympic-sized swimming pools.
Why This Matters: Forecasting Volcanic Activity
The study, published in the journal Science, demonstrates a new capability in understanding volcanic unrest. Traditionally, monitoring volcanic activity relies on detecting magma rising toward the surface. However, this swarm occurred due to horizontal magma movement at a depth exceeding 8 kilometers. The researchers believe this approach could revolutionize volcanic forecasting:
“The tremors act as if we had instruments deep in the Earth, and they’re telling us something,” explained Anthony Lomax, a research geophysicist involved in the study. “When we analyze the pattern those earthquakes make in our 3D model of the Earth, it matches very, very well what we expect for magma moving horizontally.”
The Current Status: Activity Has Subsided, But Risks Remain
While the immediate threat appears to have passed – with the magma likely cooling and solidifying deep underground – volcanoes can remain restless for years. Recent volcanic activity in Iceland highlights this unpredictability. The scientists emphasize that combining AI analysis with fundamental physics could transform how we monitor and predict volcanic behavior, potentially saving lives in seismically active regions.
The key takeaway is that seismic swarms don’t always signal an immediate eruption, but instead can indicate complex underground processes. This research underscores the need for advanced monitoring techniques to interpret these signals accurately.
