A Cornell-led study published June 2 in Geochemistry, Geophysics, Geosystems has found that two of Mount Etna’s most powerful historic eruptions — one in 122 B.C. and another roughly 4,000 years ago — were set in motion by fundamentally different forces. Lead author Maxim Gavrilenko and principal investigator Esteban Gazel determined that carbon dioxide drove one event while water drove the other, suggesting that even within a single volcano, the mechanics of catastrophic eruption can vary dramatically.
Study finds two Etna eruptions had different volatile drivers
The Cornell-led research compared two of Etna’s most violent prehistoric events. The 122 B.C. eruption — classified as Plinian, the most explosive category — was primarily driven by water vapor released at shallow depths of 2 to 5 kilometers. The earlier event, known as the Fall Stratified eruption and dating back roughly 4,000 years, was propelled instead by high concentrations of carbon dioxide originating from depths of 24 to 30 kilometers.
Gavrilenko, a former postdoctoral researcher in Gazel’s lab, served as lead author. Gazel — the Charles N. Mellowes Professor in Cornell’s Department of Earth and Atmospheric Sciences — has long focused on what makes eruptions explosive and what mechanisms control them. His analogy is simple: a shaken soda bottle releases its bubbles rapidly and violently. Volcanoes, he says, work exactly the same way.
Raman spectroscopy technique enables precise reconstruction of volcanic plumbing
To reach these conclusions, the team used Raman spectroscopy to examine micron-sized bubbles trapped inside crystals that formed within magma. These bubbles are roughly 1 to 10 percent the thickness of a human hair — vanishingly small, yet they contain measurable information about conditions deep underground.
“That technique gives us the density of CO2, and using a state equation we can transform that density into pressure, and pressure can be transformed into depth,” Gavrilenko said. The result is an unusually precise map of a volcano’s internal plumbing system — a level of detail that was not achievable before.
The method was first pioneered by Gazel’s group in 2023, when they demonstrated that carbon dioxide — not just water, as had long been assumed — can trigger explosive eruptions. Field sampling at Mount Etna for the current study was conducted by co-authors Terry Plank of Columbia University and Bruce Houghton of the University of Hawaii, Manoa.
Different volatile thresholds produce distinct eruption styles and speeds
The contrast between the two eruptions is considerable. In the 122 B.C. event, magma rose slowly from roughly 22 kilometers below the surface, then paused for several weeks at a shallow level of 2 to 5 kilometers, gradually releasing water vapor before finally erupting. Unhurried, almost incremental — until it was not.
The Fall Stratified eruption followed a very different path. Driven by elevated CO2 concentrations, magma rose rapidly from depths of 24 to 30 kilometers and reached the surface within hours.
Gazel notes that Mount Etna is unusual in having both volatiles as competing drivers. Most volcanoes are dominated by one or the other: oceanic island volcanoes tend to be CO2-heavy, while those in subduction zones are typically water-dominated. “This shows that at a certain threshold of CO2, the eruption will come from very deep and really fast,” Gazel said, “but when you have a higher threshold of water, then the process is controlled at shallow levels.”
Findings aim to improve volcanic risk assessment worldwide
The implications extend well beyond Etna. Gazel’s team is now applying the same Raman spectroscopy method to volcanoes in Chile, Hawaii, and other locations around the world, with the goal of building detailed physical models of eruption behavior for as many volcanoes as possible.
“Ideally this should be done in every volcano on the planet,” Gazel said. “This is data we need for physical models of eruptions that are the base of risk assessment.” For communities living near active volcanoes, those models could eventually translate into better early warning systems and more informed evacuation planning.
The research was supported by the National Science Foundation. Additional co-authors include postdoctoral researchers Kyle Dayton and Ellyn Huggins, and Anna Barth of the University of California, Berkeley.
What the study ultimately establishes is that a single volcano can switch between fundamentally different eruption mechanisms depending on which volatile species dominates at a given time. The 122 B.C. eruption was a slow, water-driven process playing out near the surface; the Fall Stratified event was a deep, fast, CO2-driven explosion. The Raman spectroscopy technique that revealed this distinction is now being expanded globally — with the long-term aim of giving scientists, and the populations they serve, a clearer picture of what lies beneath active volcanoes.
Carlos is an engineer with strong expertise in technical and industrial topics. He previously worked at international companies such as Siemens and speaks Spanish, German, English, and Italian.
