Same volcano, two entirely different stories written in the magma
Beneath the surface of Mount Etna, two ancient forces — carbon dioxide and water — have long competed to determine how, and how violently, the earth will speak. Cornell University researchers have now decoded that rivalry with unprecedented precision, using Raman spectroscopy to read gas bubbles trapped in magma crystals like messages sealed in stone. Their findings reveal that the same volcano can erupt through fundamentally different mechanisms depending on which volatile prevails, a discovery that reframes how humanity might anticipate the behavior of volcanoes worldwide.
- For decades, water was assumed to be the primary trigger of volcanic eruptions — but carbon dioxide has now been shown to drive its own rapid, deep explosive pathway, upending long-held assumptions in volcanology.
- Two eruptions at Mount Etna, separated by nearly four thousand years, tell entirely different stories: one slow and water-controlled from shallow depths, the other a fast, CO2-driven surge from over 24 kilometers below the surface.
- A new Raman spectroscopy technique can now measure micron-scale gas bubbles inside magma crystals, allowing researchers to reconstruct the pressure, depth, and timeline of eruptions with a precision previously out of reach.
- Etna's rare position as a battleground between both volatiles makes it an ideal laboratory — and understanding the threshold at which CO2 overtakes water could prove critical for predicting eruption type before it happens.
- The method is already being extended to volcanoes in Chile, Hawaii, and beyond, with the long-term ambition of feeding richer physical models into global volcanic risk assessment systems.
Inside the magma chambers of Mount Etna, carbon dioxide and water have been waging a slow, invisible competition — and a Cornell University-led research team has now mapped that rivalry with a clarity that was impossible just years ago.
The key insight is deceptively simple: what makes a volcano explode is less about the magma itself than about the gases trapped within it, and how quickly those gases escape. Volcanologist Esteban Gazel compares it to shaking a soda bottle — the carbonation separates violently, while an undisturbed bottle opens calmly. For decades, water was considered the primary volatile trigger. But Gazel's group had already suggested in 2023 that carbon dioxide might be equally important. The challenge was proving it.
The answer came through Raman spectroscopy, a technique that allows researchers to peer into the microscopic bubbles — some just a fraction of a human hair in width — trapped inside crystals that form as magma cools. By measuring the density of CO2 in these pockets, the team could calculate the pressure at which the bubbles formed, and from that, determine the depth. The volcano's hidden plumbing system became readable.
The team collected samples from two Etna eruptions nearly four thousand years apart. The 122 B.C. Plinian eruption showed magma rising slowly from about 22 kilometers, pausing at shallow depths for weeks, with water as the dominant volatile. The earlier Fall Stratified eruption revealed something entirely different: magma had surged rapidly from 24 to 30 kilometers deep, driven by high concentrations of CO2, and erupted within hours. Same volcano, two distinct mechanisms.
What makes Etna exceptional is that it sits at a geological crossroads where both volatiles genuinely compete — most volcanoes favor one or the other. When CO2 reaches a critical threshold, eruptions originate deep and move fast. When water dominates, the process is shallower and slower. Identifying that threshold could be essential for predicting not just whether an eruption will occur, but what kind.
Gazel's team is now applying the method to volcanoes in Chile, Hawaii, and elsewhere, with an ambitious goal: to eventually study every active volcano on Earth this way. For communities living in volcanic shadows, understanding how an eruption might unfold — not merely that it could — may one day be the difference between preparation and catastrophe.
Inside the magma chambers of Mount Etna lies a competition between two invisible forces. Carbon dioxide and water, both trapped as gases in molten rock, push toward the surface through different pathways and at different speeds. A team of researchers led by Cornell University has now mapped these competing mechanisms with a precision that was impossible just a few years ago, revealing that the same volcano can erupt in fundamentally different ways depending on which volatile wins out.
The work began with a simple question: what makes a volcano explode? Esteban Gazel, a volcanologist at Cornell, describes it using an analogy anyone can grasp. Shake a bottle of soda and the carbonation separates violently; leave it still and you can drink it calmly. Volcanoes operate on the same principle. The magma itself—its thickness, its chemistry—matters less than the gases trapped inside it and how quickly those gases escape. For decades, geologists assumed water was the primary culprit in triggering eruptions. But in 2023, Gazel's group published findings suggesting carbon dioxide could be equally or more important. The question then became: how could they prove it?
The answer came through a technique called Raman spectroscopy, which allows researchers to peer into the tiny crystals that form as magma cools. Inside these crystals are even tinier bubbles—so small they measure between one and ten percent the thickness of a human hair—that contain trapped gases. By measuring the density of carbon dioxide in these micron-scale pockets, the team could work backward to calculate the pressure at which the bubbles formed, and from pressure, determine the depth. Suddenly, the plumbing system of a volcano became readable, like following a map drawn in stone.
Mount Etna, despite its reputation as a gentle giant among volcanoes, offered the perfect laboratory. The team, including collaborators from Columbia University and the University of Hawaii, traveled to Sicily and collected samples from two eruptions separated by nearly four thousand years. The first, which occurred around 122 B.C., was one of the largest on record—a Plinian eruption, named after Pliny the Elder, who witnessed Mount Vesuvius erupt in 79 A.D. The second, the Fall Stratified event, happened nearly four millennia earlier. When the researchers analyzed the crystals from each eruption, they discovered two entirely different stories written in the magma.
In the 122 B.C. eruption, magma had risen slowly from a depth of about 22 kilometers, then paused for weeks at a shallow level between 2 and 5 kilometers below the surface. There, it released gas gradually before finally erupting. Water appears to have been the dominant volatile, controlling the pace and the pathway. The earlier Fall Stratified eruption told a different tale. Magma had shot upward rapidly from 24 to 30 kilometers deep, driven by a much higher concentration of carbon dioxide, and erupted within hours. Same volcano, same basic chemistry, but two distinct mechanisms.
What makes Mount Etna unusual is that it sits at a crossroads where both volatiles compete for control. Most volcanoes lean heavily toward one or the other. Oceanic island volcanoes tend to be carbon dioxide dominated; subduction zone volcanoes, where one tectonic plate slides beneath another, are typically controlled by water. Etna is one of the few places on Earth where the two gases battle for dominance. When carbon dioxide reaches a certain threshold, the eruption originates from great depth and happens fast. When water takes over, the process unfolds at shallower levels and moves more slowly. Understanding this threshold could be crucial for predicting which type of eruption might come next.
Gazel's team is now applying the Raman spectroscopy method to volcanoes in Chile, Hawaii, and beyond. The goal is ambitious: eventually, every volcano on the planet should be studied this way. The data generated could feed into physical models of eruptions, the same models that form the foundation of volcanic risk assessment. For communities living near active volcanoes, knowing not just whether an eruption might happen, but how it might happen, could make the difference between preparation and chaos. Gazel notes that working on Mount Etna carries an extra dimension—the volcano sits at the intersection of geology and history, a place where ancient myths speak of giants buried beneath the earth, and where the plumbing systems themselves seem to echo those old stories.
Notable Quotes
At a certain threshold of CO₂, the eruption will come from very deep and really fast, but when you have a higher threshold of water, then the process is controlled at shallow levels.— Esteban Gazel, Cornell University
This is the data we need for physical models of eruptions that are the base of risk assessment.— Esteban Gazel, on applying the method globally
The Hearth Conversation Another angle on the story
Why does it matter that CO₂ and water trigger different eruption paths? Can't geologists just prepare for any eruption?
The speed and depth of an eruption determine how much warning you get and what kind of damage to expect. A fast, deep eruption from CO₂ gives you hours. A slow, shallow one from water might give you weeks to evacuate, or it might stall and release gas gradually instead of exploding. The mechanism tells you the timeline.
How did they actually see inside the magma? You can't just look at molten rock.
They looked at crystals that formed as the magma cooled. Inside those crystals are microscopic bubbles—smaller than a hair is thick—that trapped the original gases. Raman spectroscopy reads the density of CO₂ in those bubbles, and from density you can calculate pressure, and from pressure you can calculate depth. It's like reading a record written in stone.
So Mount Etna has erupted two completely different ways?
Yes. Four thousand years ago, CO₂-rich magma shot up from 24 kilometers deep and erupted in hours. In 122 B.C., water-rich magma rose slowly from 22 kilometers, paused for weeks at shallow depth, and released gas gradually. Same volcano, same basic plumbing, but two entirely different stories.
Is this technique unique to Etna?
No, they're applying it everywhere now—Chile, Hawaii, and they want to do every volcano eventually. The method works anywhere you have crystals with trapped gases. But Etna was the ideal place to develop it because it's one of the few volcanoes where both volatiles compete, so you can see the contrast clearly.
What does this mean for people living near volcanoes?
Better risk models. Right now, volcanic forecasting is rough. If you know whether a volcano is CO₂-dominated or water-dominated, or where the threshold is between them, you can predict not just if it will erupt, but how—and that changes everything about evacuation planning and preparedness.