Published online 18 September 2008 | Nature | doi:10.1038/news.2008.1119


Permafrost that lives up to its name

Ancient Canadian ice survived previous warm periods.

PermafrostUndergound ice can last a long timeDuane Froese, University of Alberta

A 740,000-year-old wedge of ice discovered in central Yukon Territory, Canada, is the oldest known ice in North America. It suggests that permafrost has survived climates warmer than today's, according to a new study.

"Previously, it was thought that the permafrost had completely disappeared from the interior about 120,000 years ago," says Duane Froese, an earth scientist at the University of Alberta in Edmonton, Canada, who is the author of the study published today in Science1. "This deep permafrost appears to have been stable for more than 700,000 years, including several periods that were warmer and wetter."

The relict ground ice is located in the Klondike region of the Yukon and was exposed by gold mining in the late 1990s. Froese first identified the site in 2000, but assumed at the time that the ice and surrounding permafrost were relatively young. It wasn't until 2005, when a large rainstorm uncovered a layer of volcanic ash, or tephra, on top of the ice that he and his colleagues were able to estimate its age.

The rate of decay

“We have to work on an explanation of this”

Vladimir Romanovsky
University of Alaska in Fairbanks

Fission-track dating, which quantified the damage done to the tephra's glassy particles by the decay of the uranium and other radioactive materials they contain, put the ash's age between 680,000 and 800,000 years.

"It's pretty exciting," says Jim Beget, a tephra expert at the University of Alaska in Fairbanks, who visited the site in 2005, but was not involved in the study. "Based on the exposure, the ash is above the ice and it seems to be a valid date."

Others are more sceptical. Vladimir Romanovsky, a geophysicist who is also at the University of Alaska in Fairbanks, says that although he can imagine that some very local areas of permafrost could have survived the previous interglacial warm period about 120,000 years ago, 740,000 years seems too old. If the age is correct, "we have to work on an explanation of this", says Romanovsky.

If the finding is corroborated and other similar sites are identified, climate modellers could use the information to improve their understanding of permafrost dynamics under warmer scenarios. Most models show the permafrost to be "more unstable than it really is", says David Lawrence, a climate researcher at the US National Center for Atmospheric Research. Permafrost depth, surface vegetation and the lay of the land all influence the speed at which the permafrost degrades.

Holding back the fears

Global circulation models predict that high northern latitudes will experience the greatest greenhouse warming, with some models estimating a 7–8 °C rise in temperature in the Arctic if nothing is done about rising carbon emissions. The warmer temperatures could thaw the permafrost, which would boost microbial activity. The microbes' respiration would release carbon from previously frozen soil into the atmosphere in the form of further carbon dioxide and methane, a more powerful greenhouse gas.


In an article in the September issue of Bioscience, Ted Schuur of the University of Florida and his colleagues estimate that there are more than 1,672 billion tonnes of carbon stockpiled in permafrost up to three metres deep2. Earlier calculations estimated that 48 billion tonnes of carbon could be released from Canadian permafrost over the twenty-first century if the mean annual temperature increased by 4 °C.

The old-ice finding suggests that some pockets of permafrost may be more resilient to climate change. "But it doesn't mean that we shouldn't expect very severe changes in permafrost if this predicted warming does happen," says Romanovsky. 

  • References

    1. Froese, D. G., Westgate, J. A., Reyes, A. V., Enkin, R. J. & Preece, S. J. Science 321, 1648 (2008). | Article | ChemPort |
    2. Schuur, E. A. G. et al. Bioscience 58, 701–714 (2008).
Commenting is now closed.