Water content controls how deep magma is stored under many volcanoes, study shows
Possible key to better eruption predictions overturns conventional theory
Water content controls how deep magma is stored under active volcanoes, a new study shows. Here, lead author Daniel Rasmussen of Columbia University’s Lamont-Doherty Earth Observatory collects volcanic ash from Fisher Crater in the Aleutian Islands. Shishaldin Volcano looms in the distance. (Diana Roman/Carnegie Institution for Science)
Between 40 and 50 volcanoes around the world are currently erupting or in turmoil, putting hundreds of millions of people at risk. However, reliable eruption predictions have long puzzled scientists, largely because they don’t fully understand why magma starts or stops moving below the surface weeks, months or years before the eventual eruption. The results of a new study may bring them one step closer.
The study found that for the world’s most common types of volcanoes, it is the water content of the magma that controls the depth of temporary storage; the more water, the greater the depth. The study challenges the popular theory that magma stops rising when its buoyancy equals that of the surrounding rock. Scientists believe that a deeper stagnation sounds like good news, but it doesn’t seem to reduce the chances of an eruption. Also, water is the main driver of explosive eruptions, so when something ends up being released in so-called “wet” magma, the results can be pretty violent.
“This study links the depth of magma storage to water, which is important because water primarily initiates and drives eruptions,” said the lead author. Dan Rasmussen, who completed most of his research during his Ph.D.candidate Lamont-Doherty Earth Observatory at Columbia University.
In recent years, scientists have used geophysical measurements to determine where magma bodies lie beneath many volcanoes. These objects range from 20 kilometers to almost at the surface. “We already know where the magma is stored — the final resting place where it gathers power before it erupts,” said the Lamont-Doherty volcanologist Terry Plank, a co-author of the paper. “We thought it was a delay, not a stop. Now we know the conditions before the eruption.”
The research began in 2015, when Planck suggested Rasmussen pursue an unsolved question of why storage depths vary from one volcano to another. The study focused on a specific geological setting: so-called arc volcanoes, which sit at the intersection of converging tectonic plates. Arc volcanoes are the most numerous type on Earth, making them the most obvious targets for improving forecasting capabilities.Their storage depths vary from 3 km to 6 km, making up the entire “Ring of Fire” Encircling the Pacific Plate, from the Aleutian Islands in Alaska to the South Pacific.
with a team including study co-authors Diana Roman of Carnegie Institution for Science, Planck and Rasmussen collected ash from eight volcanoes in Alaska’s remote Aleutian Islands. Using boats and helicopters to get around, they encountered rough seas, rough terrain and, on Unimak, the threat of a giant brown bear.
Ash is the main target because it can contain green crystals made of olivine, each about 1mm in diameter. Underground, olivine crystals sometimes trap tiny magma as they form. After the volcanic eruption sent the crystals to the surface, the magma inside them cooled and turned into glass. By analyzing the chemical composition of these tiny magma chunks, the researchers estimated the water content of the magma.
Cleveland Volcano in the Aleutian Islands, one of the most active volcanoes in the United States, was the subject of the study. (Daniel Rasmussen/Lamont-Doherty Earth Observatory)
After estimating the water content of six Aleutian volcanoes, the team combined the data with other estimates of magma water content in the scientific literature to list 62 volcanoes. They supplemented data from more than 100 volcanoes around the world.
Rasmussen, now a postdoctoral fellow at the Smithsonian Institution National Museum of Natural Historysays the Smithsonian database Global Volcano Program “It was the key to compiling these lists, because it’s a very good source of eruption history, and we only wanted to think about volcanoes that erupted recently.” The team focused on recent eruptions, because the magma reservoir didn’t appear to have moved much after the eruption . Therefore, any estimate of depth or water content using recently erupted material has the best chance of accurately reflecting the current state of the volcanic magma reservoir.
The team eventually mapped the estimated magma storage depths of the world’s 28 volcanoes against their respective estimated magma water content. The results were clear: magma with more water content tends to be stored deeper in the crust. The team went on to show that water content is not only related to storage depth, but also to storage depth. They demonstrated this by identifying chemical tracers associated with the formation of hydrous magma in the mantle, which lies beneath the crust.
As for how water content determines how deep the magma is stored, the authors believe it has to do with a process called degassing, in which water dissolved in the magma forms bubbles as it rises through the crust, and the pressure on the magma decreases – similar to how you slowly What happens when you unscrew the cap of a soda bottle. When liquid magma begins to degas, it crystallizes and becomes more viscous—that is, less fluid. The researchers believe that this thickening causes the magma to slow down and stagnate.
That makes the magma less like beer and more like toothpaste, Planck said. But now the newly formed bubbles are trapped in this thick mucus and try to expand under pressure. If something happened to take the pressure off and let everything unleash, it would give the masses even more potential for a blast.
But exactly what it was, and what it might have been, wasn’t clear, Planck said. The authors note that the systems they studied are considered “eruptible” — they “apparently undergo viscous stagnation, rather than viscous death, in contrast to the non-eruptive stagnation intrusions typically observed in the years to decades preceding many eruptions. Events coincide.” In other words, the volcano is ready to erupt. But the eventual trigger or triggers, and how they are detected, remain largely a mystery. “We know what we found, but we’re not quite sure what all the effects are,” admits Planck.
Evidence that water content controls the depth of magma storage overturns the most widely accepted explanation in the field today, that magma rises under pressure through cracks in the crust because lava is more buoyant than the surrounding solid crust. It will then sink at its storage depth as it reaches so-called neutral buoyancy, at which point the magma is no more buoyant than its surroundings.
This project is associated with a Ongoing broader Lamont-Doherty effort In the Aleutian Islands, sophisticated instrument suites are planted on two highly active volcanoes to record in detail the precursors of any eruption. Planck and others hope the research will lead to the development of better forecasting methods, as well as inexpensive arrays that can be used to monitor active volcanoes around the world, many of which are poorly monitored, if they are monitored at all.
The next step, Rasmussen said, is to see if the findings apply to volcanoes in other geological settings.These include so-called hotspot volcanosuch as those in the Hawaiian Islands, and Rift Volcano, such as those in East Africa. Beyond that, Rasmussen said, a bigger question looms: “If magma water content controls magma storage depth, what controls magma water content?”
The study was also co-authored by Mindy Zimmer of Pacific Northwest National Laboratory. Funding and support were provided by the Smithsonian Institution, the National Science Foundation, the Southwest Washington Community Foundation, and the U.S. Geological Survey.
This story is based in part on a press release from the Smithsonian Institution.



