Fresh magma is refilling the reservoir under Japan’s Kikai caldera, site of the Holocene’s largest known eruption

A seismic survey under the sea south of Kyushu has found a large, partly molten body of rock beneath Japan’s Kikai caldera, and the researchers behind it say it is the same magma reservoir that fed the largest known volcanic eruption of the Holocene about 7,300 years ago. Kobe University and the Japan Agency for Marine-Earth Science and Technology, known as JAMSTEC, report that magma has since been injected into that reservoir anew. The authors add that this does not mean another catastrophic eruption is imminent.

Save more, waste less: The Cool Down’s newsletter sends smart, simple changes that help you and the planet in your everyday life.

Subscribe to The Cool Down →

We may earn a commission if you sign up.

The finding reached a wide audience on Oct. 23 when ScienceDaily reposted it, but the work itself was published on March 27, 2026, in the journal Communications Earth and Environment. Readers meeting the story for the first time should treat it as a seven-month-old paper newly in circulation, not as a fresh volcanic development.

What the airgun survey measured

The team fired powerful airgun arrays to send seismic pulses through the crust beneath the seafloor and recorded the pulses on ocean bottom seismometers, an approach described in Kobe University’s release. Waves travel at different speeds through solid rock and through rock that contains melt, so the slowdown reveals where molten material sits. According to the published paper, the line ran 175 kilometers and used 39 seismometers, and it found a zone where seismic velocity falls by more than 15 percent, with a maximum of 22 percent.

The authors read that zone as a large reservoir at a depth of 2.5 to 6 kilometers. Their best estimate of the melt fraction is 3 to 6 percent, which they say could be limited to 10 percent at most, and they give a volume of about 220 cubic kilometers.

Why the researchers call the magma new

Two lines of evidence point to re-injection. The first is location: the low-velocity zone lies beneath the site of the giant eruption, which Kobe University’s Nobukazu Seama, a co-author, put in plain terms. “Due to its extent and location it is clear that this is in fact the same magma reservoir as in the previous eruption,” he said, according to the ScienceDaily repost of the release.

The second is chemistry. A lava dome has been building near the center of the caldera for about 3,900 years, and the paper lists its volume at more than 32 cubic kilometers. Its composition, along with that of other recent activity, differs from the material the giant eruption threw out, which the authors take as a sign that the magma was injected from deeper in the crust or mantle. Seama described the newer material as “likely newly injected magma.”

For scale, the paper puts the 7,300-year-old eruption at about 160 cubic kilometers of dense-rock equivalent, within a range of 133 to 183. A 2024 seabed study from the same Kobe group had mapped its deposits over more than 4,500 square kilometers around the caldera, and found that the underwater part of the pyroclastic flow could travel long distances, even uphill. The study’s authors noted that deposits laid down in the sea are preserved well, while on land erosion removes much of the evidence of such an eruption.

The hedges built into the finding

The paper is careful about what it can and cannot say. The reservoir’s size and melt content depend strongly on an assumed laboratory-derived parameter, so the authors say exact values may not be determinable, and their synthetic tests suggest the true reservoir could be larger, with a higher melt fraction, than they report. The re-injection model rests on two assumptions: that the reservoir of the giant eruption sat in the same place, and that later melt was re-injected into it.

The authors frame any role for this reservoir in a future giant eruption as a possibility rather than a conclusion. In the words of the ScienceDaily text, the replenishment process “does not mean that another catastrophic eruption is imminent.” They suggest the model could also help explain shallow magma reservoirs at Yellowstone in the United States and Toba in Indonesia.

Kikai is not quiet in the ordinary sense. The Smithsonian Institution’s Global Volcanism Program entry for Kikai describes a caldera 19 kilometers wide that is mostly submarine, with Satsuma-Iwo-jima on its northwest rim, and lists the highest point on the island, Iodake, at 681 meters and describes continuing low-level activity there, including gas-and-steam plumes and volcanic earthquakes, under a level 2 alert on Japan’s five-level scale. That ordinary activity at the island is separate from the deep reservoir the survey imaged, and the sources read do not link the two.

Seama said the underwater location lets researchers carry out systematic, large-scale surveys. He also said that understanding how such large quantities of magma accumulate is necessary to understand how giant caldera eruptions occur, which is the stated motivation for imaging a reservoir that, by the authors’ account, has received new melt.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

Save more, waste less: Subscribe to The Cool Down →

We may earn a commission if you sign up.