Arctic sea ice is nearly back to average global levels for the first time in at least a decade after years of spectacular declines.

The surprise growth at a time of year when ice is normally melting has triggered a blizzard of I-told-you-sos among online climate change skeptics.

But the man whose data is behind the furor says a few weeks of cold weather in one part of the Arctic — not the end of climate change — has skewed the numbers.

"It is not the end of global warming," said Mark Serreze of the National Snow and Ice Data Center in Boulder, Colo., which publishes monthly sea-ice updates on its website.

On Wednesday, the center posted a new graph showing that the extent of ice-covered Arctic Ocean has nearly returned to the 1979-2000 average.

The graph was a significant surprise. Data from the last eight years shows that September sea ice was 22 per cent below that 20-year average. And until the beginning of March, this year's sea ice was on pace to match 2007's record low.

What happened?

It's called freaky Arctic weather. "All of the action is in the Bering Sea," Serreze said. "For the past several weeks, we've been under a rather unusual weather pattern, a cold pattern, that's given us this late spurt in ice growth in the Bering Sea. If you look at the rest of the Arctic Ocean proper, it is very warm."

The Bering Sea, between Alaska and Russia, is caught between two low-pressure systems.

"This is weather," said Serreze. "Don't conflate this with climate." Serreze notes that on the Atlantic side of the Arctic, ice is low.

However, online climate change skeptics have seized on the data as proof that global warming is a hoax. By Thursday morning, hundreds of Twitter posts were referencing Serreze's graph — many linking to a Sydney Morning Herald blog that displays the graph without explanation or context.

"An inconvenient fact," reads one post. "Arctic sea ice back to normal — oops."

Another asked: "Will the hysterical eco-nuts be called out? NO!"

Right-wing think-tanks such as the U.S.-based Heartland Institute had stories on the graph, but without comment from Serreze. Other online articles quoted climate skeptics as if they were the authors of the research.

Even those who accept current climate change theories were fretting.

"Is this another PR problem for global warming activists?" worried one online environmental newspaper.

"Everyone's on this now," sighed Serreze. "What you're seeing now from the usual suspects is that it's the end of global warming, and we don't see it that way."

Serreze points out that the satellite data his graph is based on offers no information on ice thickness. He suggests that most of the recent ice in the Bering Sea is likely to be very thin and won't last.

"Once the winds change, temperatures change, we'll probably lose it pretty quickly."

Serreze said the more important figure is sea-ice minimum, but that won't be evident until the end of the Arctic summer.