Notes from The Gathering #5: Arctic sea ice: is it tipped yet?

The summer of 2007 was apocalyptic for Arctic sea ice. The coverage and thickness of sea ice in the Arctic has been declining steadily over the past few decades, but this year the ice lost an area about the size of Texas, reaching its minimum on about the 16th of September. Arctic sea ice seems to me the best and more imminent example of a tipping point in the climate system. A series of talks aimed to explain the reason for the meltdown.

Sea surface temperatures were warmer this past summer also; I forget how many standard deviations the temperature was off the trend, but it was definitely anomalous. The region of the meltback is just inside the Bering Strait, where warm water flows in from the Pacific, but in the analysis of Steele et al. this inflow of comparatively warm water was not particularly anomalous in 2007 relative to other years. It could be that the exposure of the sea surface to the atmosphere by the melting ice could have an impact, although the meltback is so late in the solar heating season (September) that this effect seems of limited explanatory value also. Bit of a chicken and egg problem here.

Melting ice can be seen from space, I believe as puddles sensed by the QuickSCAT satellite. The puddles are most abundant in mid-summer when the sunlight is strongest, and by mid-September when the ice meltback was the strongest, the melting season was largely over. Apparently the reason for the disappearance was an anomalous weather system which generated a strong jet of surface winds blowing straight over the pole southward toward the Atlantic ocean, a “Polar Express”. A research ship frozen into the ice in 2006 crossed the Arctic in about a year, about three times faster than the transit time of the Fram in the 1890’s. To summarize, the ice cubes in the freezer tray didn’t melt because the freezer is broken exactly, but because the ice cube tray fell out of the freezer onto the warm floor.

The disappearance of the ice was set up by warming surface waters and loss of the thicker multi-year ice in favor of thinner single-year ice. But the collapse of ice coverage this year was also something of a random event. This change was much more abrupt than the averaged results of the multiple IPCC AR4 models, but if you look at individual model runs, you can find sudden decreases in ice cover such as this. In the particular model run which looks most like 2007, the ice subsequently recovered somewhat, although never regaining the coverage before the meltback event.

So what is the implication of the meltback, the prognosis for the future? Has the tipping point tipped yet? When ice melts, it allows the surface ocean to begin absorbing sunlight, potentially locking in the ice-free condition. Instead of making his own prognosis, Overland allowed the audience to vote on it. The options were

  • A The meltback is permanent
  • B Ice coverage will partially recover but continue to decrease
  • C The ice would recover to 1980’s levels but then continue to decline over the coming century

Options A and B had significant audience support, while only one brave soul voted for the most conservative option C. No one remarked that the “skeptic” possibility, that Arctic sea ice is not melting back at all, was not even offered or asked for. Climate scientists have moved beyond that.

202 comments on this post.
  1. Tenney Naumer:

    I am respectfully requesting some help here from scientists about a recent comment on the Dot Earth blog concerning the effects from the loss of Arctic sea ice.

    This was the comment (a sort of question-and-answer session with nonsensical answers from Sashka who you may have already seen here on realclimate):

    May 9th,
    12:27 pm

    Re: #13 by John McCormick

    JM: What does all that open Arctic Ocean absorbing heat mean to the Western North American climate?

    S: It is absorbing in summer and losing during winter. But the answer is – almost nothing. Two reasons: (1) the area is very small; (2) the extra heat can much easier go up to outer space than laterally to midlatitudes.

    JM: Does anyone have any interest in climatalogical impact of sea ice melt or is it all about breaking records?

    S: Yes, this is what actually interests climate scientists.

    JM: We have a world of people reliant upon Western North American agricultural production which is reliant upon predicatable weather (precip and temperature).

    S: Climate change in Arctic will have negligible effect on agriculture and precisely zero on weather prediction.

    — Posted by Sashka

    This is a link to the nytimes page:

    Thank you in advance for any help you can give.

    [Response: This appears to be referring to some recent work that indicates that disappearing summer sea ice affects Pacific storm tracks (making them go further north) and could affect rainfall in western North America. This is a relatively new result, and you’d want to see some replication in different models before you took this as gospel. It isn’t completely out of the bounds of possibility though. – gavin]

  2. Tenney Naumer:

    OK, thanks, Gavin, I will see what I can find on that.