Arctic ice
A massive Arctic ice floe north of Europe was studded with cracks and melt ponds on June 30 as the German icebreaker Polarstern continued a year-long monitoring project. (Photo credit: MOSAiC/Alfred Wegener Institute)

Relentless high pressure and cloud-free skies have allowed the Arctic Ocean’s sea ice to plummet to its lowest mid-July extent on record. The persistent pattern sets the stage for what could be unprecedented losses by September – a long-feared next step in the Arctic’s erratic, climate-change–fueled lurch toward a potential “blue ocean” mode.

Early July brought “just the pattern you’d like to see if you’d like to get rid of ice,” said Mark Serreze, director of the National Snow and Ice Data Center. What happens later this summer will hinge in part on whether an entirely different, tough-to-predict ice-destroying mechanism takes shape.

Arctic sea ice extent
The five-day running average of Arctic sea ice extent ending on July 15, 2020, was at a record low for the date in satellite-based data extending back to 1979. (Image credit: National Snow and Ice Data Center)

The core of year-round sea ice that extends from the central Arctic to the north coast of Canada has taken a well-publicized beating lately. In the past half century, human-produced greenhouse gases have led to warming that’s unfolding more than twice as quickly across the Arctic as in the entire Northern Hemisphere. As perennial ice retreats, the exposed ocean absorbs more sunlight than the highly reflective ice surface, hastening regional warming.

Arctic ice now just over half of extent typical in 80s

Arctic sea ice extent plummeted to a record low in 2007, stunning even veteran researchers. The crash marked a new era: No minimum since 2007 has managed to stay above pre-2007 values, and an even lower record was set in 2012.

In 2019, the minimum extent was the second-lowest on record, in a virtual tie with 2007 and 2016.

The Arctic now experiences little more than half the ice extent in September than what was typical in the 1980s. That’s allowed waves from autumn storms to attack Alaska’s Arctic coast with increasing ferocity, putting villages at risk. The delayed autumn freeze-up across the Bering Sea has left Indigenous communities that rely on snowmobile transport marooned for longer periods.

Northern sea route
Northern Sea Route (blue), Southern Sea Route (red). (Image credit: Collin Knopp-Schwyn and Turkish Flame / Wikimedia)

The increasingly open waters of the Arctic have also beckoned shippers, including fossil fuel producers. Much of the Northeast Passage (Northern Sea Route) was ice-free for 93 days in 2019, the longest such period in decades of satellite measurement. This year, a natural gas tanker headed into the passage from the Siberian port of Sabetta on May 18, more than a month earlier than any such departure on record.

How to melt an ocean’s worth of ice

The Arctic’s melt so far in 2020 has been most dramatic in the Laptev Sea, off the coast of central Russia. “It’s the Siberian side this year,” Serreze said.

Unprecedented warmth swaddled much of Siberia during the first half of 2019, leading to a surge in spring wildfires. The heat sent warm river runoff cascading into the sea and helped erode coastal ice sooner than usual. On June 20, the city of Verkhoyansk (latitude 67°N) soared to 38 degrees Celsius (100.4°F), a reading recently confirmed by the World Meteorological Organization as the highest temperature ever reliably measured north of the Arctic Circle.

A study released on July 15 by the World Weather Attribution project found that the Siberian heat of the first half of 2020 would have been “almost impossible” without climate change.

Some of the black carbon – soot and other emissions from far-northern wildfires – can move over the Arctic Ocean, darkening the ice surface and increasing the sunlight’s melting power. One plume of satellite-inferred wildfire smoke extended from Siberia to the North Pole at the start of July, and another plume in mid-July extended to Svalbard, Norway (east of Greenland), according to Mark Parrington of the Copernicus Atmosphere Monitoring Service. This year’s carbon emissions from wildfires within the Arctic Circle from July 1 to 15 were the largest in the 18-year satellite record for that period.

In that first half of July, several large-scale weather features ground to a halt across the Northern Hemisphere. One of those was a strong dome of high pressure over the central Arctic that kept temperatures mild and skies largely clear for days on end. Such a sprawling, long-lived high isn’t unheard of, but this one arrived at an especially bad time of year. Although the sun never gets far overhead in the Arctic, it rarely dips below the horizon during late June and early July. That allows a surprising amount of solar energy to soak the region. On the northern summer solstice (usually June 21), the full-day average of solar energy reaching Earth’s atmosphere is larger at the North Pole than anywhere else on the planet. So any weather feature that leads to clear skies can greatly boost the ice-melting ability of the polar sun at this time of year.

Stormy low pressure sometimes dominates the Arctic in early summer, with extensive clouds that block round-the-clock sunlight and blunt its impact on the ice. That’s been the case in several recent years, but not so in 2020.

It’s not just the barrage of sun-driven melt that allows high pressure in early summer to cut into sea ice extent. As air flows clockwise around the high, the ocean and ice surface are pushed toward the right of the wind at any given spot. The result? The ice is packed into a smaller, more concentrated area, which allows sunlight to warm the newly exposed waters.

Satellite image from July 15, 2020
Only scattered areas of clouds can be seen across the Arctic in this NASA visible satellite image from July 15, 2020. (Image credit: NASA Worldview)

Looking at possibilities ahead

Forecast models suggest the ice-baking high will weaken during the last few days of July. By that time of year, though, Arctic ice loss depends less on high-latitude sunlight, which is already on the wane. Instead, atmospheric circulation will take the driver’s seat.

One possible mode of late-summer ice loss would be an Arctic Dipole pattern: High pressure over the Beaufort Sea north of Alaska and Canada, and low pressure toward Europe and Asia, can team up to pull warm air from the Pacific into the Arctic and push ice across the basin toward a watery doom in the far North Atlantic.

It’s also possible that one or more intense Arctic cyclones will push the concentrated ice outward while eroding the edges of the pack with strong wind and big waves. Once dispersed into lower-latitude water and broken into smaller floes, that ice would be more prone to melt quickly.

In August 2012, the strongest such storm on record, nicknamed the great Arctic cyclone, spun across the central Arctic for more than a week. Such a cyclone may not be enough by itself to cause record ice loss, as the clouds and chilly air of a summer cyclone can inhibit some melting. However, one study found that extensive ice damage from the 2012 cyclone may have nudged the Arctic toward its record-low extent that year.

It’s impossible to predict with confidence where or when such a cyclone might form until a few days out. They’re most common in July and August, according to polar researcher Steven Cavallo, of the University of Oklahoma. The most destructive trek for sea ice would be from Siberia into the Alaskan side of the Arctic, he said. For the cyclones to do the worst damage, “you need that preconditioning from the high pressure earlier in the season to melt the ice and make it thin. That definitely seems like the case this year,” Cavallo said. Warm subsurface water pushing in from lower latitudes can also weaken the ice pack from below.

Ice expert Serreze: ‘Old rules are becoming less reliable. … You’re coming into a new realm now.’

In just three days, Cavallo noted, a single cyclone can smash a half-million square kilometers of sea ice – an area larger than California – to smithereens.

As of mid-July, much of the ice between the North Pole and Asia was first-year ice less than 3 feet (1 meter) thick, which makes it especially vulnerable to any potential damage from a late-summer cyclone.

The consensus verdict from 33 groups and individuals, via the Sea Ice Prediction Network, is that Arctic sea ice extent most likely will bottom out in the ballpark of the second-place minimum of 2019. Only two of the forecasters called for a new record-low extent. These forecasts were compiled in June, though, weeks before the accelerated melt of early July, and studies have found limited skill in the seasonal outlooks.

Warming planet soon to trump year-to-year weather vagaries

As the years roll by, summer melt in the Arctic may be driven more by the inexorable momentum of a warming planet and less by the vagaries of year-to-year weather, according to Serreze. “Old rules are becoming less reliable,” he said. “The link between the ice conditions and the atmospheric conditions is still there, but questions are emerging on whether these relationships we’ve looked at in the past are really holding as much as before. You’re coming into a new realm now.”

Whatever happens over the remainder of 2020’s melt season will be tracked with new precision thanks to the MOSAiC project, billed as the largest polar expedition in history. An instrument-laden German icebreaker has been drifting across the Eurasian side of the pole since last fall. The ship was moored to a large floe, with sensors scattered across the ice up to 40 miles (25 km) away and teams of scientists rotating on and off board every few months.

Despite obvious signs of melting by June (see image at top), the floe continued to serve as MOSAiC’s home base.

“The thickest area of the floe, which we dubbed the fortress, has for the most part weathered the deformations in the spring quite well, and continues to offer a good basis for our research camp,” according to MOSAiC lead Markus Rex, of the Alfred Wegener Institute at the University of Potsdam, in a news release on June 18. “We’ll be able to continue working here well into the summer. But, with the extensive summertime melting that has now already begun, we’ll need to keep our instruments and installations highly mobile, and be ready to adapt to changing circumstances. Later in the summer, it may be necessary to relocate the camp – it all depends on how the ice conditions develop.”

Julie Regnery and Amy Macfarlange on MOSAIC expedition
Julia Regnery and Amy Macfarlane look toward MOSAiC’s host ice floe before they start their exploring tour on the project “fortress”, the thickest part of the floe, on June 17, 2020. (Photo credit: Lisa Grosfeld, via MOSAiC/Alfred Wegener Institute)

Pegging an ominous benchmark

The startling Arctic ice loss of 2007 suggested that a nearly ice-free Arctic Ocean – typically defined as less than 1 million square kilometers of ice, or less than 10% of the basin, and sometimes dubbed a “blue ocean event” – could arrive far sooner than earlier thought. But just how soon?

In a 2013 paper, James Overland of the NOAA Pacific Marine Environmental Laboratory and Muyin Wang of the University of Washington laid out the predictions from three schools of thought. Simply extrapolating trends in sea ice volume from the early 1980s to the late 2000s would have led to a blue ocean event as early as 2020. Variations in weather and climate have kept that frightening option at bay, at least so far.

A second camp emphasizes the Arctic’s ability to recover from a year of moderate ice loss, and also the importance of major-loss years such as 2007 in pushing the Arctic ice down a stairwell of successively greater long-term depletion. This view suggests that several more summers of major loss, perhaps separated by 5 or 10 years, could end up producing a blue ocean event at some point between 2030 and 2040.

Global climate models have long been conservative in predicting the demise of Arctic ice. On average, they’ve tended to produce a blue ocean event no sooner than 2040, and often not until late this century. And that’s assuming that greenhouse emissions continue unabated (the Intergovernmental Panel on Climate Change RCP 8.5 scenario), which will perhaps end up being overly pessimistic.

Also see: Sea ice video addresses key questions

Whether the year-round ice has decades to spare remains an open question. Dirk Notz, of the University of Hamburg/Max Planck Institute for Meteorology, led a recent review of the latest crop of modeling in support of the next IPCC assessment. These models (CMIP6) continue to span a wide range of outcomes for Arctic ice, said Dirk and colleagues. Although the models still tend to underestimate the amount of ice loss produced by a given amount of warming, most of the models now produce a blue ocean event prior to 2050, even if major emission cuts are put into place.

“Available evidence suggests that scientists have been conservative in their climate projections, with a late bias in dates for change,” Overland and Wang wrote in their 2013 paper.

In a recent email, Overland continued to cast his lot with the middle-range rather than the longer-range scenario. “I am still in the 2030 range,” he said. “Present ice is just too thin not to be influenced by the gradual warming trend plus a wind event.”

For information on YCC’s “Eye on the Storm,” visit this page. To keep up with news on timing for launch of the blog, sign up for announcements. Got questions? Send an email to EyeOnTheStorm.

Bob Henson is a meteorologist and journalist based in Boulder, Colorado. He has written on weather and climate for the National Center for Atmospheric Research, Weather Underground, and many freelance...