For more than 50 years, scientists have debated whether a massive ice shelf—up to 1 kilometer thick—covered the entire Arctic Ocean during past ice ages, transforming the frigid water into a solid icy surface similar to Antarctica’s Ross Ice Shelf.
The hypothesis dates to the 1970s, when British glaciologist John Mercer and others proposed that during extremely cold periods, continental ice sheets would have extended far into the Arctic Ocean. It gained support in the 1990s when researchers began finding evidence of scouring on the seafloor, indicative of large, kilometer-thick ice running aground.
“We found out that even close to the Norwegian coast, there was still open water, which completely contradicts the hypothesis of a big ice shelf covering the Arctic Ocean.”
But new data published in Science Advances add evidence against such a “pan-Arctic” ice shelf, instead suggesting that seasonal sea ice, rather than a continuous ice shelf, dominated parts of the Arctic Ocean over the past 750,000 years.
“We found out that even close to the Norwegian coast, there was still open water, which completely contradicts the hypothesis of a big ice shelf covering the Arctic Ocean,” said coauthor Gerrit Lohmann, a climate modeler from the Alfred-Wegener-Institut in Germany. However, some experts argue that the results alter merely the timing and location of Arctic ice shelves.
Reading Ancient Algae
Instead of analyzing seafloor scars, the study’s authors looked at what was living in ancient seafloor sediment. They analyzed two sediment cores drilled from the Arctic Ocean between Europe and Greenland, searching for molecules produced by marine algae such as diatoms and dinoflagellates before the organisms died and sank to the seafloor.
Some species of alga grow on the underside of seasonal sea ice, and others thrive in open water. Their presence or absence within sediment deposited at a given time signals whether sea ice was present when they were living. Levels of calcium in the sediment can also indicate the production of marine organisms in surface waters.
By searching for organisms’ unique chemical signatures in dated sections of the cores, the scientists could conclude whether and when a solid ice shelf completely covered the ocean surface.
The results showed evidence of both seasonal sea ice and open water over the past 750,000 years, with one exception, around 676,000 years ago, when the chemical signature of the key marine life decreased for roughly 55,000 years.
On a train home after a funding interview, the study’s first author, Jochen Knies, was discussing the sediment core findings with Lohmann, who immediately recognized that the computational climate model he worked on, the high-resolution AWI Earth System Model, might offer additional data on the sea ice conditions during that time. “We discussed it for hours, maybe disturbing others on the train,” Lohmann said.
After some testing, he found that the model independently predicted that the same regions covered by the core samples would have had open water and seasonal sea ice instead of a continuous ice shelf, even during the coldest periods. “I was fascinated to see that in the time slices that [Knies] was interested in, the sea ice was even partly absent in summer,” Lohmann said of the modeling results. “It was completely the opposite of other hypotheses.”
“I would have assumed that where they found open water, there should have been times when this Arctic Ocean ice shelf moved into the area, and apparently it didn’t,” said Johan Nilsson, a paleoceanographer at Stockholm University in Sweden who was not involved in the new study but has published seafloor evidence of an Arctic Ocean ice shelf.
The Debate Continues
To Nilsson, the results don’t completely refute the possibility of large Arctic Ocean ice shelves; instead, they redefine their possible boundaries. “I think for me, it pushes back the edge of Arctic Ocean ice shelves a bit further north of Svalbard,” Nilsson said.
The authors of the new study “don’t see ice shelves in the Norwegian-Greenland sea, but that doesn’t mean that they didn’t exist in the Arctic.”
Leonid Polyak, a retired paleoceanographer at the Ohio State University who was not involved in the research, said the new study reveals “a very strong set of data.” He noted, however, that the evidence for Arctic ice shelves is strong, and the debate over whether they came together into one pan-Arctic ice shelf is “a bit overblown.”
“Pretty much everyone agrees that there have been ice shelves in the Arctic Ocean. The question is, When exactly did they exist, for how long, and where?” Polyak said. The authors of the new study “don’t see ice shelves in the Norwegian-Greenland sea, but that doesn’t mean that they didn’t exist in the Arctic.”
Lohmann acknowledged that mysteries remain in the Arctic and that the pan-Arctic ice shelf debate may not be settled. “I feel the final word hasn’t been spoken,” he said.
—Andrew Chapman (@andrewchapman.bsky.social), Science Writer