Normally in the Arctic, the ocean water keeps freezing through the entire winter, creating ice that reaches its maximum extent just before the melt starts in the spring.
Not so this year.
As of Tuesday, sea ice had stopped growing for two weeks. Sea ice extent -- the areas with at least 15 percent ice coverage -- hit a winter maximum of 14.214 million square kilometers (5.488 million square miles) on Feb. 9, and has stalled since, according to daily reports from the National Snow and Ice Data Center in Colorado.
If there is no more growth, the Feb. 9 total extent would be a double record that would mean an unprecedented head start on the annual melt season that runs until fall.
"If this was the maximum in sea ice extent, it would be the earliest that we've ever seen and it would also be the lowest maximum that we've ever seen, by a long shot, actually," said Mark Serreze, director of the National Snow and Ice Data Center. "If this was the maximum, that would be pretty big in terms of what is happening in the Arctic."
Up to now, the lowest winter ice-extent maximum in the satellite record dating back to 1979 was hit last year, when ice extent reached 14.54 million square kilometers (5.614 million square miles) on Feb. 25, according to NSIDC records. The earliest seasonal winter maximum was reached in 1996, when sea ice hit its peak extent on Feb. 24 of that year, according to the center's records. Normally, ice extent reaches its maximum in early or mid-March; between 1981 and 2010, the average maximum date was March 12.
Serreze and his colleagues are not prepared to declare that the 2016 maximum has now been reached and those new records set, however.
"I think it's too early to make any kind of call, although it is certainly a very perplexing winter," he said.
Extent could bounce up if weather conditions change from what has been a pattern of unusual Arctic warmth, he said. "There's still a fair bit of winter left in the Arctic," he said.
But even if ice starts growing again this winter, that would be only thin ice that disappears quickly once the melt season starts, Serreze said. "You don't have any time left in the season to grow thick ice," he said.
The extraordinarily low ice for February, which follows a record-low January ice-extent average, has put extent in a "hole" for the start of the 2016 melt season, he said.
"Where will we end up in September? It depends so much on the summer weather pattern," he said. "But the odds are we're going to have a very low September."
The most notable lack of winter ice has been on the Atlantic side, near Norway's island of Svalbard, according to the center. Ice is also low for this time of year in the Bering Sea and Pacific regions, with direct impacts on Alaska, experts said.
"There isn't a lot of ice in Cook Inlet. There's almost nothing in the Bristol Bay region. And that's remarkable," said Tony Fischbach, a U.S. Geological Survey wildlife biologist who compiles a daily sea-ice report.
Residents of Northwest Alaska, posting on the Alaska Native Tribal Health Consortium's Local Environmental Observer network, have provided online evidence of warm conditions and sparse winter sea ice. Michael Sloan of Nome sent photos and a description of a rainy Dec. 30, when water puddled on what he said was an unusually small amount of intact ice near the shore. Millie Hawley, president of the Native Village of Kivalina, made similar observations in another Dec. 30 post, with a photo of slush on the local lagoon.