In May of 2018, a barrage of earthquakes struck Mayotte, the seismically quiet easternmost island of the Comoros archipelago, which stretches between Africa and Madagascar. After months of investigating the unexpected, intense seismic activity, French scientists discovered a new submarine volcano in the Indian Ocean approximately 50 kilometers east of the island. This new seafloor feature is, by volume, the largest documented underwater volcanic eruption in history, and both volcanic and seismic activities continue today.
In the years since the initial quakes, teams of scientists have refined the picture of the structures below the seafloor. Building on this foundation, marine geologist Nathalie Feuillet and seismo-tectonicist Eric Jacques, both at the Institut de Physique du Globe de Paris, and their colleagues proposed the presence of a large underwater caldera—a volcanic depression formed when a magma chamber drains and collapses—located between Mayotte and the new volcano. A ring of earthquakes in the mantle lithosphere delineates this curious structure at depths where neither earthquakes nor calderas typically occur.
A Timeline of Peculiar Events
Mayotte formed approximately 11 million years ago and is the oldest in the chain of Comorian volcanic islands, said Feuillet in her prerecorded talk for the AGU event. The island slumbered until 10 May 2018, and in the first 2 months of the seismic episode, more than 100 events greater than magnitude 4.5, including a magnitude 5.9 event, rocked the island. Its population, accustomed to stable ground, had only a single seismic station to monitor the sudden crisis.
In November 2018, seismic stations around the world heard a mysterious hum that scientists traced to Mayotte. This 25-minute-long signal vibrated at low frequencies, hinting that magma movement and volcanic activity were the prime suspects to explain the seismic crisis. Back at sea level, Feuillet said local fishermen reported “dead fishes” and a “burned tire smell” during this time.
In May 2019, the first of seven marine cruises confirmed that 5 cubic kilometers of lava had piled onto the previously flat seafloor, constructing an 800-meter-tall volcanic edifice, said Feuillet in her talk. The presence of hydrogen gas in the water column indicated that the nascent volcano was actively erupting into the sea during the scientific cruise. With each subsequent campaign, scientists discovered fresh lava flows, she said.
From Seafloor to Mantle
East of Mayotte’s shores, Feuillet described “the crown,” a circular structure on the seafloor approximately 10 kilometers in diameter and dotted with many cones. She and her colleagues interpreted the structure as the outline of an old caldera. An oceanic ridge stretches eastward, away from the crown, truncated by the fledgling volcano 50 kilometers from Mayotte.
To explore the structures in the lithosphere below the seafloor, Jacques, Feuillet, and their colleagues focused on two distinct clusters of earthquakes. The eastern cluster, located between the suspected caldera and volcano, concentrates earthquakes deep below the seafloor, parallel to the ridge, said Jacques. This cluster includes the inaugural earthquakes from May and June 2018, which Feuillet said may indicate that a dike propagated from west to east before the eruption, pushing magma from a deep western reservoir eastward and upward toward the volcano, similar to the model proposed in a paper published earlier this year.
The donut-shaped western cluster began shaking in the summer of 2018. Earthquakes tell scientists where faults are breaking, so this ringlike cluster implies faults arrayed in a circle mirroring the possible caldera directly above. At depths of 25 to 55 kilometers, these earthquakes nucleated in the mantle below the local depth of the Mohorovičić discontinuity, said seismologist Wenyuan Fan in an email. Fan, an assistant professor of geophysics at Scripps Institution of Oceanography, is not affiliated with the new studies. Earthquakes usually rupture brittle crust, where strain energy can accumulate, he said. Strain cannot accumulate at typical mantle conditions, so most mantle tends to flow, inhibiting the breaks of earthquakes.
“One way to [get mantle earthquakes] would be to bring down cooler materials that can host earthquakes,” Fan said, which happens in subduction zones. But the Mayotte events are not related to subduction, leaving these mantle earthquakes a mystery.
How the western donut cluster connects to the eastern ridge cluster poses another quandary because they are not linked by earthquakes, said seismologist Lise Retailleau, also of Institut de Physique du Globe de Paris but not part of Feuillet’s and Jacques’s studies. Whether the lack of seismic connection between the clusters indicates aseismic magma transfer or some deeper connection hidden from scientists, she said, “we don’t really know.”
Also perplexing are observations of the lithosphere between the proposed caldera and the donut cluster. Retailleau pointed out that the caldera shows no evidence of volcanic activity at the human timescale, even though most of the seismicity recorded since 2018 congregates below it. Even stranger, the upper crustal lithosphere between the surface and the top of the donut cluster lacks earthquakes that indicate breaking, brittle crust. Instead, long-period and very long period seismic events associated with this eruption—events that often imply magma movement—occur just above the donut, said Retailleau. Feuillet said that these events might foreshadow another untapped shallow magma chamber above the brittle rocks defined by the donut cluster. If this magma chamber is active and continues its upward path, Retailleau said it could affect Mayotte and its population of more than a quarter million people. “That’s a big question,” she said.
—Alka Tripathy-Lang (@DrAlkaTrip), Science Writer