Researchers found hidden layers of ancient earthquake damage inside a single tube of deep ocean mud

Radiocarbon tests on a 16-inch sediment core taken off Japan found seven hidden layers, and several line up with Nankai earthquakes since 1605.

In 2012, scientists aboard the research vessel Tansei-maru pulled a sediment core 16 inches (40 centimeters) long from the Kumano Trough off central Japan, under about 6,820 feet (2,080 meters) of water.

The mud looked the same from top to bottom, and it was logged as homogeneous. Radiocarbon measurements taken less than half an inch apart have now turned up seven disturbed layers inside it.

A University of Tokyo team links most of those layers to underwater slope failures during historical Nankai Trough earthquakes. Every match comes from this one core, and the authors word them cautiously.

Taking core samples

The Kumano Trough lies off the Kii Peninsula. It is a forearc basin, a depression that formed as the Philippine Sea Plate slid beneath the Eurasian Plate.

Its floor sits around 6,560 feet (2,000 meters) down, below a steep slope and a shelf less than 6 miles (10 kilometers) wide.

The crew took core KT-12-34-MC01 with a multiple corer in the western part of the basin, far from the mouths of the Shingu and Owase submarine canyons.

The sample is clayey silt. Near 1.6 inches (4 centimeters) down, it shifts from brownish black to olive black, and the naked eye finds nothing else.

X-ray CT scans made with help from Kochi University showed only slight changes in brightness. One brighter band, between about 4 and 6 inches (10 and 16 centimeters), hinted at a past disturbance.

Older carbon above younger mud

Lead author Akiko Omura, Yusuke Yokoyama and three colleagues at the Atmosphere and Ocean Research Institute measured radiocarbon in the total organic carbon of the mud.

They worked down the core in steps of 0.2 to 0.4 inch (0.5 to 1.0 centimeter), using the single-stage accelerator mass spectrometer at the institute.

Undisturbed seafloor mud gets older with depth. Some horizons in this core broke that rule, and their carbon tested as much as 700 years older than the mud beneath them, well past the average measurement error of about 100 years.

The authors read those reversals as the mark of turbidity currents, fast underwater flows of sediment that can start when a slope fails. Such a flow lifts old seafloor material and drops it on top of younger mud.

"We found several new geological layers that were not visible before," Yokoyama, a professor at the institute, said in a University of Tokyo statement. He said the redeposited sediments match the material below, which left "the chemical and structural features" indistinguishable.

Organic carbon in this basin tests older than the mud around it, so the team needed a second clock. They picked planktonic foraminifera, tiny shelled ocean organisms, in samples of just 0.75 to 2.35 milligrams, and M. Okitsu of Palynosurvey Co. identified them.

Seven paired depths gave offsets of 700 to 1,109 years between shell dates and carbon dates. Near the top, shells from about an inch down carried a radiocarbon level that, checked against coral records from Ishigaki and Kikai, placed that mud after 1965.

The corrected timeline puts the whole core at roughly 600 years of deposition. The oldest dated part, about 15 inches (37 to 38 centimeters) down, goes back to around 1440 C.E.

Hoei, Keicho and Ansei

Omura's team then set the layers beside a list of documented earthquakes. According to the paper, layers 1 and 2 may date to the 1944 to 1946 earthquakes and to the Ansei Tokai earthquake of 1854, but it does not say which layer goes with which.

The authors tied layer 3 to the Hoei earthquake of 1707, shown at magnitude 8.6 on the university's map. That layer overlaps the bright band from the Kochi scan.

A shift in carbon isotopes appears near 6 inches (15 centimeters), where it begins, and the paper says the earthquake might have changed conditions at the site.

The paper offers two options for layers 4 and 5, the Hoei event or a minor submarine earthquake. It says layer 6 could have come from the Keicho earthquake of 1605.

The Kumano River runs directly into this basin, so the team had to consider floods. M. Saito and N. Suzuki at the National Museum of Nature and Science in Tokyo helped run the stable carbon isotope tests that check for land plants.

Carbon-to-nitrogen ratios came back near 8. Carbon isotope values stayed between minus 20 and minus 22 per mil in normal and disturbed layers, which the paper reads as marine organic matter and evidence against a flood origin.

Layer seven stays unassigned

The paper reports data from this one core and gives no estimate of the size or location of any event. Omura's group presented the work at the 25th International Radiocarbon Conference in Krakow, Poland, in 2025.

Two layers were not fully invisible. Layer 3 lines up with the bright CT band and layer 5 with a pale strip of fine laminae, while scans showed nothing matching the others.

On the paper's chart, the deepest layer carries a question mark. The authors did not assign it to any earthquake.

The university statement lists severe floods, storms and submarine earthquakes as possible causes of disturbed mud. It also mentions previously unknown earthquakes and written records from the 13th century onward.

The paper connects the layers to historical disasters after the 15th century, and the core stops near 1440. Whether any layer records an earthquake missing from the written history remains open.

Buried ridge shaping Nankai

On Nov. 9, 2026, the Japan Meteorological Agency posted its regular Nankai Trough assessment and reported no notable change in the odds.

The same notice says a magnitude 8 to 9 earthquake there has a high probability within 30 years, and that about 80 years have passed since the 1944 and 1946 events.

JAMSTEC researchers recently described a buried ridge that may shape where the next Nankai rupture stops. Off North America, deep-sea cores hold about 3,100 years of Cascadia earthquake history in turbidite beds that geologists can see, unlike most layers in the Kumano sample.

Tsunami work has also turned to mud. Patrick Sharrocks of the University of Leeds led a study showing that fine sediment thickened the front of the 2011 tsunami.

According to the university, Yokoyama now wants samples from across the Kumano Sea and beyond. The JSPS and JST CREST funded this round, which began with a 16-inch tube of mud once logged as uniform.

The full study was published in the journal Radiocarbon.

-- -

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.