Fifty-six million years ago, trillions of tons of carbon found its way into the atmosphere, acidifying oceans and causing the already-warm global climate to heat up by another 5º C (9º F)—an episode known as the “Paleocene-Eocene Thermal Maximum” or “PETM.”

Like today, the warming climate affected the environment on land and in the sea, with extreme downpours and heat-stressed plankton at the base of the food web. Land animals had a high rate of extinction and replacement by smaller species, and there was a mass extinction of tiny shell-making creatures that lived on the sea bed. The hotter climate supported alligators and swamp-cypress forests, like those in today’s southeastern United States, in Arctic latitudes that are covered by ice and tundra today.

Where did all that carbon come from?

Its source has been debated for years, with some scientists blaming the destabilization of methane ice in the seabed and others pointing to the widespread volcanic activity in the North Atlantic at the time. Modeling of the carbon isotope shift points to carbon originating from organic and volcanic sources, but the relative proportions aren’t settled.

A new study published in Nature Geoscience by Professor Christian Berndt of the GEOMAR Helmholtz Centre for Ocean Research in Germany blames underground magma that drove methane and CO₂ from marine sediments into the atmosphere via gassy eruptions dubbed “hydrothermal venting.” Berndt worked with an international group of 35 coauthors on the paper.

Waiting 17 years for a date

The idea that hydrothermal venting played a major role in the PETM dates back to 2004. Seismic images gathered for oil and gas exploration showed that the marine sediments off Norway were pockmarked with thousands of craters that are about PETM age, and other studies have found similar craters near Greenland. But the seismic images couldn’t constrain the time when the craters formed precisely enough to determine if they played a role in triggering the PETM: “That was conjecture, basically,” said Berndt.

To see if the vents could have triggered the PETM, they needed to retrieve samples from them to date them—and that required drilling deep into the seabed that lies below 5,600 feet (1.7 km) of the Atlantic Ocean.

So in 2004, Berndt and several coauthors formally proposed a project to drill and sample a hydrothermal vent, but they had to wait 17 years before drilling finally started in 2021 as part of the International Ocean Discovery Program (IODP). “You have to be patient,” said Berndt.

Berndt and colleagues were aboard the scientific drilling ship “JOIDES Resolution” as it drilled five boreholes into the “Modgunn Vent,” some 200 miles off the Norwegian coast. The crater at the top of the vent is about 1.3 km (4,300 feet) wide and approximately 80 meters (260 feet) deep. Beneath the crater, seismic images show a 400-meter-deep (1,300 feet), chimney-like feeder zone that connects the crater to a sheet of now-frozen magma called a “sill.”

The right time?

“This was bang on at the beginning of the PETM,” Berndt told Ars.

The samples recovered from the boreholes provide “conclusive evidence for hydrothermal venting immediately before the PETM onset,” supporting the “major role” of the vents in the PETM warming, Berndt and colleagues say in their paper. They base this on two lines of evidence in the crater: a globally recognized shift in carbon isotopes that marks the PETM and the presence of a species of plankton that only existed during the PETM.

“The crucial one and the most precise one… is the carbon isotope excursion,” said Berndt.

But these lines of evidence only show up in the sediments that filled the crater after it initially formed; they’re found 10–15 meters up from the crater floor. That distance leaves some wiggle room in tying the crater to the start of the PETM. “That means the vent was formed very shortly before the PETM, and during the PETM, it filled in,” said Berndt.

“The crater is older than the PETM,” agreed Professor Appy Sluijs of Utrecht University, who was not involved in Berndt’s study. But Sluijs points out that the plankton species in these deposits existed throughout the PETM. “The species can therefore not distinguish between the onset or the body of the event,” said Sluijs. In other words, the presence of this species can’t narrow down the time the crater formed to less than a fairly large window.

So how long before the PETM did the vent form?