In this photo provided by Dirk Notz, taken April 24, 2009, ice floats in the Arctic near Svalbard, Norway. A new study has found evidence of an ancient methane “explosion” in the Barents Sea near Svalbard. (Dirk Notz via Associated Press)

Nearly 12,000 years ago, a sudden burst of methane surged from the bottom of the Arctic Ocean, leaving an array of gigantic craters in its wake. And a slow trickle of gas continues to leak from the seafloor to this day.

The findings, described in a paper published Thursday in the journal Science, are the latest of recent discoveries pointing to Arctic methane explosions in the ancient past. And the findings have inspired debate among scientists about whether such an event could happen again and how it could affect the climate.

“The authors show exciting new evidence that this methane can be trapped in the sediment and released in large bursts, with a much greater potential of reaching the atmosphere,” said Stephen Grasby, a research scientist with Natural Resources Canada who was not involved with the new research, in an emailed comment.

This matters because methane is a potent greenhouse gas — its warming effect on the atmosphere is up to 30 times as strong as that of carbon dioxide over a period of 100 years or so. So the discovery of any new methane sources or leaks on the planet is a point of interest with climate scientists.

But the climate implications of underwater methane leaks are murkier. First, these leaks or “seeps” are by no means rare. There are hundreds of known seeps throughout the ocean floor. They’re the result of a natural process, in which methane forms as organic matter decomposes at the bottom of the sea and gradually bubbles back up from the sediment into the water column.

Scientists generally believe that the methane leaking from these seeps never makes it to the surface of the ocean, instead dissolving in the water on its way up. But some suggest that an explosion, of the type described in Thursday’s paper, could produce enough force to send some gas straight up to the surface and into the atmosphere, with potentially climate-warming consequences.

The new paper describes one such event that occurred about 12,000 years ago in what is now the Barents Sea, a region of the Arctic Ocean stretching between Norway and Russia. There, at the bottom of the ocean floor, stands a collection of more than 100 craters, some as much as 3,000 feet wide and nearly 100 feet deep. The researchers believe they were formed by sudden rushes of methane from the seafloor.

Methane, although most commonly observed in gas form, can sometimes become trapped at the bottom of the ocean in very deep or cold regions, freezing into a solid substance known as a methane hydrate. It can remain trapped this way indefinitely until something destabilizes it.

The area where the craters are located was once covered by extensive glaciers, which placed enormous pressure on the land beneath them, helping to form the methane hydrates and keep them stable. But about 15,000 years ago, the ice sheet began to retreat, gradually releasing some of that pressure. As this happened, the space in the ground where conditions were right for the hydrates to remain stable grew thinner and thinner.

“The thinning and the retreating of the ice sheet led to increased concentration of gas hydrates in increasingly shallow layers below the ice sheet,” explained Karin Andreassen, the new paper’s lead author and a professor of marine geology and geophysics at the Center for Arctic Gas Hydrate, Environment and Climate at the Arctic University of Norway.

At the same time, temperatures in the region were increasing. And as the ice continued to retreat and the enormous pressure was relieved, Earth’s crust began to bounce back into shape. These factors further contributed to the compression of the hydrates in the Arctic sediment and began to cause mounds to rise up at the bottom of the newly formed sea. Finally, the pressure became too much, the gas burst forth, and the mounds collapsed into the craters we observe today.

“I think it was probably like a lot of champagne bottles being opened at different times,” Andreassen said.

Although there’s no evidence that other sudden bursts have occurred in the region since, there appears to still be a substantial reservoir of gas there. To this day, methane and other hydrocarbons continue to slowly vent from the seafloor — and probably have been since the explosion occurred, Andreassen said.

She added that there’s evidence of other intact hydrates still in the region — mounds that have yet to burst — and what will happen to them in the future remains unclear.

As for the methane explosion, it’s not the only such event to have occurred in ancient Arctic history. Grasby, the Natural Resources Canada scientist, recently published research describing a cluster of rocky mounds in the Canadian Arctic that researchers believe were also caused by an ancient methane explosion, although much further in the past. Grasby’s mounds were likely formed about 100 million years ago, probably after a sudden warming period in Earth’s history destabilized the hydrates.

His paper raised the question of whether modern-day Arctic warming could be priming the region for another methane explosion in the future and whether that gas could make it to the atmosphere. The concern is that this scenario could lead to a dramatic amplification of global warming, even potentially triggering a climate feedback loop, in which more warming causes more methane to be released.

But some scientists have contested this idea. When Grasby’s paper was released in April, geophysicist Carolyn Ruppel of the U.S. Geological Survey told The Washington Post that the findings did not substantially change our understanding of what happens to methane when it vents from the ocean floor, and she has generally taken a skeptical position on the theory that unstable methane hydrates could lead to runaway climate effects.

Andreassen, too, cautioned that these theories have no concrete support for the time being, noting that it would be a “dramatic conclusion” to assume that ancient methane explosions had a significant effect on the climate without substantial extra work to back up the idea. Still, because methane hydrates do still exist in the region — and conditions in the Arctic are rapidly changing, thanks to the climate change — it’s a scenario deserving of continued research.

But because we haven’t observed any similar explosions in the present day — not yet, anyway — Grasby also agrees that there’s considerable uncertainty remains about exactly what happens when such an event occurs.

“The novel study points out that we still have a poor understanding of the mechanisms of how methane hydrates melt and release gas, and the potential of that methane to reach the atmosphere,” he said.