In northernmost Alaska, permafrost is steadily warming and large amounts of thaw are expected by the end of the century. But for some spots, the thaw is about 70 years ahead of that predicted pace, new research shows.

The soil on the beds of several North Slope lakes has already passed the thaw threshold or is on pace to get there much earlier than the nearby dry permafrost, according to a University of Alaska Fairbanks-led study now online in the journal Geophysical Research Letters.

Soil at the bed of the very shallow lakes that dot the Slope has warmed by 4.3 degrees in the last 30 years, the study found. In five of the last seven years, mean annual lake-bed temperature was above freezing, the study found.

Warmer air temperatures and more snow are cited as reasons for the progression toward thaw.

Those lakes, no more than 1.1 yards deep, have ice that freezes to the bottom each winter, but in recent years that ice layer is getting thinner. Air temperature, meanwhile, is getting higher – with the change more pronounced in winter than in summer – and more heat-insulating snow is falling, the study said

Snowpack is a big driver, said Christopher Arp of UAF, the lead author.

"It makes a big difference, especially if you get snow that accumulated on the lakes early in the winter," he said.

The fastest warming is in lakes that are about 1 yard deep, where soil temperatures in the beds have averaged above freezing for eight of the last 38 years, according to the study. That warming is at an average place of 1.44 degrees per decade, matching the rise in air temperature at Barrow over the past 37 years, the study said.

Warming is happening at a slightly slower pace at the bottom of lakes shallower than 1 yard, but the rate is still fairly rapid, the study said.

In deeper lakes, where only some water freezes in winter, the rate of warming has slowed considerably, but the thaw has already been accomplished, the study said. Temperatures in those lake soils often exceed 3.6 or 5.4 degrees, bringing about sublake permafrost even during the winter, the study said.

"Under lakes where ice doesn't form to the bed, it's actually really warm," Arp said.

The study uses ice thickness and soil temperature measurements taken from 2012 to 2015 at 29 lakes of various depths — making use of fieldwork in late winter, a season not typically used in the past by Arctic scientists — plus measurements taken over longer-term monitoring programs.

Ultimately, Arp said, the result will likely be more lakes with year-round liquid water, with implications for fish and wildlife habitat.

On the bright side, thawing permafrost in wet areas, like on lake-beds, does not emit as much atmospheric carbon as does thawing permafrost in dry terrain, according to a separate study by scientists from Northern Arizona University and an international group of collaborators.

That study, published in the journal Nature Climate Change, found a 10-degree Celsius warming of dry permafrost released three times as much carbon as is released from thawed permafrost in wet areas.

The explanation: Dry permafrost exists in aerobic, oxygen-rich conditions, allowing more activity of the microbes that digest the newly thawed carbon material and send carbon emissions into the atmosphere. Wet permafrost, in contrast, exists in anaerobic conditions lacking the oxygen to support so much microbial activity, according to the study.

Even that silver lining to advanced lake-bed permafrost thaw has a caveat. The carbon emissions from thawed dry permafrost are dominated by carbon dioxide, while the emissions from thawed wet permafrost are more likely to be methane, a gas that is much more powerful at trapping the earth's heat and creating a greenhouse effect.

Correction: An earlier version of this story was missing a decimal point in a temperature statistic. According to a UAF-led study, the soil at the beds of the shallow lakes on the North Slope has warmed by 4.3 degrees in the last 30 years, not 43 degrees.