Hot Times in the Arctic

by Bill Chameides | May 4th, 2010
posted by Erica Rowell (Editor)

Permalink | 11 comments

Because so much of the Arctic Ocean’s surface used to be covered by ice in the summer and no longer is, scientists wonder whether the disappearance of that ice has had a climatic impact and whether the Arctic’s warming can be tied to the melting sea ice.

To climate scientists, positive feedbacks can be a bête noire.

Most of us think of “positive feedback” as a good thing — a pat on the back from a friend, spouse, boss, or, if you aspire to Idolhood, Simon. But when used in climate science to refer to a natural process that amplifies a potent warming effect, it can quickly quicken the pulse of a climate guy like me.

An Unsettling Wild Card of Climate Science: How Strong Is the Feedback?

The biggest uncertainty in climate scientists’ ability to predict how much something will cool or warm the climate (including greenhouse gases) is our poor understanding of the strength of additional processes that act to amplify or dampen that cooling or warming. We call those processes feedbackspositive feedbacks for those that amplify and negative feedbacks for those that dampen.

The positive feedback that probably gets the most publicity these days is the one involving water vapor. A climate perturbation like an increase in carbon dioxide (CO2) raises temperature a little, which causes water vapor concentrations (which depend upon temperature) to increase. Because water vapor is the atmosphere’s most important greenhouse gas, higher concentrations of it cause a further increase in temperature, which leads to another increase in water vapor and … well you get the picture: the initial warming is amplified.

An example of a negative feedback goes like this: an increase in CO2 acts as a fertilizer that leads to more photosynthesis and more rapid removal of CO2 from the atmosphere, dampening the initial CO2 increase.

Sorting out the relative strengths of positive and negative feedbacks is a difficult and complicated task, but we’ve been learning a lot. Through studying the present climate as well as climate changes over geologic time, climate scientists have been able to get a reasonable handle on things. We know that overall the feedbacks are positive and we can constrain how positive.

This in fact is how climate scientists have determined the 2-4.5-degree Celsius range of the temperature increases we can expect with a doubling of atmospheric CO2. If the climate is relatively insensitive to rising CO2, meaning the positive feedbacks are relatively weak, the temperature increase is expected to be on the low end of the range. If the opposite is true and feedbacks strongly amplify temperature, we could be at the high end. (For more, read this discussion of climate sensitivity.)

Here’s Where Ice Comes In

One of the most famous feedbacks (at least among climate scientists) involves ice. Ice is white and so it is highly reflective. The icier the Earth’s surface, the more reflective it is to the Sun’s rays. We call this reflectivity albedo — the more ice, the higher the Earth’s albedo; conversely the less ice the lower the albedo.

The climate connection comes about because the icier the Earth’s surface — and the higher its albedo — the less solar energy the planet absorbs and the cooler it will be. Conversely, with less ice and thus a lower albedo, the warmer the planet will be.

The feedback comes about because, as we all know, the warmer it is, the less ice there is and vice versa.

How Positive a Feedback Is the Ice-Albedo Effect?

The ice-albedo feedback occurs when something — say a rise in CO2 — causes the planet to warm up a bit, melting some ice and thus lowering the albedo. These conditions increase temperatures, which lower the albedo, which warms the planet, and so forth. As you might have guessed, this is a positive feedback that amplifies the initial climate perturbation.

The ice-albedo feedback, central to the theory of climate change, explains how a relatively small perturbation to the climate system can result in large changes. For example, ice is believed to have played a central role in the Earth’s swings from ice ages to warm interglacials: by amplifying small perturbations in the amount of solar heat deposited in the northern middle latitudes due to changes in the Earth’s orbit, the feedback brought about the huge climate swings we see in the Earth’s climate record.

How the Ice-Albedo Feedback Relates to Current Climate Change

As climate scientists watch the current warming trend, questions about the ice-albedo effect arise: is the feedback already operating? Will it begin to dominate the warming trend, and if so, on what kind of time scale?

A good deal of attention to the climatic effect of melting ice has focused on Arctic sea ice since so much of the Arctic Ocean’s surface area that was formerly covered in summertime by ice no longer is. Has the disappearance of that ice had a climatic impact? Moreover, while temperature increases in the Antarctic have been modest, the Arctic’s temperature increases have been about twice as large as the global average. Can the Arctic’s large warming be tied to the melting sea ice?

You’ve probably heard or read of some climate scientist warning that this might be happening. But it was all speculation. Not anymore. A new paper published in the journal Nature by James Screen and Ian Simmonds of the University of Melbourne provides data to back up that speculation.

The authors analyzed the most recent observational data (1989 to 2008) from the European Centre for Medium-Range Weather Forecasting. Some of the key differences between this data set and those used by previous investigators include a higher spatial resolution, improved numerical model physics, and an improved correction for a known bias in the satellite data. 

The two major findings of their analysis were that:

  • most of the Arctic’s atmospheric warming for this period has occurred near the surface, a finding indicative of heating from decreased surface reflectivity (i.e., melting ice); and
  • this amplified temperature at the surface correlates spatially with changes in sea ice extent, especially within the Arctic basin north of 70 degrees.

Conclusion: Hotter Times in the Arctic Sooner?

It’s important to note that previous investigators looking at older versions of this data set reached different conclusions, and others have argued that the type of data used by Screen and Simmonds cannot be used for trend analysis in part because of the bias they claim to have corrected for. So I’d like to see confirmation from independent studies.

Nevertheless, Screen and Simmonds, by providing credible empirical evidence of the operation of the ice-albedo feedback, raise the stakes on climate change by at least one measure. In typical, understated science-ese the authors dryly conclude: “The emergence of strong ice–temperature positive feedbacks increases the likelihood of future rapid Arctic warming and sea ice decline.”

Now consider this: climate models typically underestimate the amount of Arctic sea ice melting that has occurred so far. Like I said, positive feedbacks can be a bear, or was that a bête noir?

filed under: Arctic, carbon dioxide, carbon dioxide emissions, climate change, faculty, global warming
and: , , , , , , , , , ,


All comments are moderated and limited to 275 words. Your e-mail address is never displayed. Read our Comment Guidelines for more details.

  1. MattN
    May 6, 2010

    …Dr. Gray is having a hard time finding any evidence of a positive feedback with water vapor:

  2. MattN
    May 6, 2010

    All evidence that I’ve come across (and there’s not all that much) indicates ice levels in the 1920s and 1930s is very, very similar to current levels. Example: Which leads me to believe this is all a natural cycle, driven by the oceans.

    • Bill Chameides
      Jun 1, 2010

      MattN: Yes, as Ken pointed out, there was a low stand in sea ice extent in the 1920s and ’30s. What is different now is that there has been a consistent downward trend since the 1970s.

      • Bill Chameides
        Jun 7, 2010

        More to the point, it is not at all clear that sea ice declines in the 1920s and ’30s approached the declines we are seeing today. See this:

      • MattN
        Jun 1, 2010

        What was the 20-30 year trend leading up to the low point of the 1920s and 30s? Do we have any good data? No. Certainly not as good as modern data. We weren’t monitoring it with satellites, or airplanes in the 1890s-1900s (like we have been since the late 70s). You have produced ZERO evidence the downward trend since the late 1970s of Arctic sea ice is in any way different than it was in leading up to the low of the 1920s. We don’t have the data to say that. And again, aren’t you the least bit encouraged that each of the last 2 years we have had marked gains in summer Arctic sea ice extent? Did you know that sea ice VOLUME in the Arctic has increased 25% in the last 2 years? See: You’re a real “glass half empty” kid of guy, aren’t you Doc?

  3. Ken Towe
    May 5, 2010

    “an increase in CO2 acts as a fertilizer that leads to more photosynthesis and more rapid removal of CO2 from the atmosphere, dampening the initial CO2 increase.” Yes, and that’s a good thing. But only if the photosynthetic product is kept away from oxygen… essentially ‘buried’ (as in an undisturbed tree or its wood products put into homes etc.) can the feedback be meaningful, long term. Its carbon also acts as fertilizer (food) for heterotrophs…bacteria to humans. Most is rapidly recycled back to CO2 by their respiration, especially in the oceans. It is not long term. Re. ice sheet loss, do we know how the current Arctic loss compares with that in the 20s? Is it “deja vu all over again”?

    • Bill Chameides
      Jun 7, 2010

      Ken: You are right that the fertilizer feedback only works if the carbon stored in the biosphere is retained for a long, long time. That is thought to be generally the case for forests before they reach climax. With regard to Arctic sea ice extent in the 1920s and ’30s: Yes, the anecdotal and limited observations from that period do indicate that this was a time of sea ice retreat. But as much as today? Without global coverage, it’s hard to say. Here is one analysis from the Climate Data Access System that suggests the recent loss of sea ice is considerably larger than that of the 1920s and ’30s.

      • MattN
        Jun 7, 2010

        …we absolutely CANNOT conclusively say that the current ice level in the arctic is unprecedented. We have compelling evidence that extent/area in the 20s/30s was very similar to what we are seeing today.

  4. MattN
    May 5, 2010

    The amount of warming you get from the doubling of CO2 is THE dispute between the two sides. We all agree the planet has warmed. We all agree CO2 is a greenhouse gas, and we all agree that CO2 concentration has increased. Taken by itself with no feedbacks, all the papers I have seen show 1.6-2C range for doubling of CO2. Water vapor is the point of contingency. Is it a positive or negative feedback? Some very smart scientists are on record as saying the overall feedback from water vapor is negative due to the increase in cloud cover (totally ignored by models) and the real warming from the doubling of CO2 is much much less than the amount indicated by the models.

    • Bill Chameides
      Jun 7, 2010

      MattN: You are correct that the effect of doubling CO2 depends to a large extent on the water vapor feedback, but you are wrong about the sign being negative. Measurements of atmospheric water vapor and the Earth’s radiative balance show pretty conclusively that it is positive.

©2015 Nicholas School of the Environment at Duke University | Box 90328 | Durham, NC 27708
how to contact us > | login to the site > | site disclaimers >

footer nav stuff