The melting Arctic – are we paying attention yet?

Record Arctic Ice melt, 2012.

The annual retreat of arctic ice away from the shores of the circumpolar countries ended on the 16th September.  Timelapse videos of the summer’s melt show the ice morphing like an amoeba as it is pushed around by ocean currents and storms.  All the while, it is being nibbled away at the edges by the warming ocean waters, and bergs are hemorrhaging through the Fram strait between Greenland and Svalbard into the North Atlantic, where they melt.

Figure 1.  Arctic ice melt 2012.  Note ice flowing through the Fram Strait and disappearing into the Atlantic at top right.

At its lowest, the extent[i] of the remaining ice was only 3.41 million km2, the lowest it had been since detailed recording began in 1972.  What is more, this was a record coming hot on the heels of other record low ice years – in 2007, 2005, and so on. 

About 800,000 km2 more ice was lost in 2012 than in the previous record loss.  And if we consider ice volume, the situation is even worse; last year’s minimum volume of 4000 km3was about 75% lower than the minimum volume in 1979 (see Figure 2 below). 

The unprecedented rate of change, especially in ice thickness, is obvious from sea level.  Returning from a 37 day research voyage, David Barber of the University of Manitoba reported that the ice pack was “rotten” all the way to the North Pole: “The multi-year ice, what’s left of it, is so heavily decayed that it’s really no longer a barrier to transportation. You could have taken a ship right across the North Pole this year“. 

Ice free in four years?

David Barber’s depressing observation underscores that what really matters for the future of arctic ice is its volume.  Regardless of area, a thin layer of ice will melt faster – perhaps a lot faster – than a thick one.  Thin ice also has less structural integrity than thick ice, and is therefore more vulnerable to wind, wave and storm action

In spite of the fact that we have observed the thinning of the ice for years, the speed of the current melt has unsettled experts.  David Barber speculates that the Arctic could be ice-free in late summer by 2020, and Peter Wadham of Cambridge University predicts an ice-free Arctic in only four years

Over at Carbon Brief,Verity Payne counsels caution in extrapolating from current trends.  Arctic wind patterns have changed over the last few years, pointing to the possibility that cyclical weather patterns may have produced the record melts. A recent Hadley Centre research note also emphasizes that processes that melt ice are not modeled precisely enough to rule out natural variability as a cause of the current ice decline (Hewitt et al. 2012).

Now I am no climate modeler, but I could not resist fitting a simple statistical model to the volume data published by the Polar Ice Centre.  The polynomial model in figure 2 explains 90 percent of the ice volume variation, and extrapolated forward in time, projects an ice free late summer by 2018 – exactly in the middle of Barber’s and Wadham’s predictions.  


 Figure 2. Ice volume estimates since satellite measurements began.  Data from the Polar Ice Centre are generated using the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS).

Positive Feedbacks

Statistical extrapolation is always dangerous, assuming as it does that current trends and variables that force those trends will continue.  The Cryosphere is also a very complicated place.  The making, breaking and melting of ice interact in complicated ways with ocean temperature and salinity. 

But ice and snow are very sensitive to positive feedback loops – self reinforcing cycles between environmental stimuli and responses that amplify natural phenomena (e.g. melting ice).

Ice reflects between 40 and 80% of incident solar radiation back into space, depending on its thickness and whether or not it is snow-covered.  Melt some of that ice to expose inky-dark open water, and your reflectivity (or albedo) goes down to about 4 to15%.  More radiation is therefore absorbed, heating up the ocean, which in turn melts more ice.  The ice melts from below, where the decline in thickness is tough to observe, which may explain part of our surprise at the “rapid” decline in ice cover. Positive feedback number 1.

The extra heat absorbed by the ocean doesn’t just sit there.  Heat is a flux of energy; it has to flow, and be changed in the process.  In the fall the extra heat absorbed by the open ocean is released back to the atmosphere, mostly as long-wave radiation, which is absorbed by greenhouse gases.  This extra heat is at least partly responsible for the 2 – 5oC increase in autumn temperatures in the Arctic during the last decade. Positive feedback number 2.

The first two positive feedbacks are at the core of “Arctic amplification (AA), which is the tendency of high northern latitudes to experience an enhanced rate of warming.  A more speculative consequence of AA is that it may have reduced the velocity but increased the north-south amplitude of Rossby waves[ii] in the polar jet stream (Francis and Vavrus 2012).  The jet stream has slowed by about 14% since 1980, and like pushing on the end of a rug, this seems to have led to much deeper “folds” in the jet stream, which can lead to cold air being funneled southwards and to warm air being funneled north.  Taken together, these phenomena can lead to more extreme and persistent weather patterns (like the record high temperatures seen in the USA and parts of Canada last summer).

On land as well, earlier spring snow melt on the circumpolar land masses has led to a positive feedback from albedo to temperature (Brown et al. 2010), and is thought to be largely responsible for AA of temperatures during summer.  Positive feedback 3.

Future surprises.

So much for the positive feedbacks that have already been observed.  If the ice continues to melt and oceans continue to warm, there may be other, particularly nasty feedbacks waiting in the wings.

The potential of melting permafrost to release CO2 and methane is well known.  Less well known is the potential for a warming Arctic Ocean to release the methane trapped in gas clathrates[iii] and sub-sea permafrost.  If the clathrates were to melt, the resulting release of methane would give a massive one-time boost to global warming (the “methane gun” hypothesis). 

Methane releases from the bed of the Arctic ocean have been observed, but have only contributed a small amount to the global release of methane so far.  On the other hand, regional warming on land, partly resulting from sea ice loss, could release some of the gigataonnes of methane and CO2 trapped in terrestrial permafrost.

Warming could also accelerate melting of the Greenland Ice Cap.  While melting sea ice makes little difference to sea level, the disappearance of the Greenland Ice Cap could raise sea levels by at least half a meter.

Then there is the prospect that the infusion of all that fresh melt water into the ocean could weaken the northern edge of the Thermohaline Circulation, which is part of the vast oceanic conveyor belt carrying warm tropical waters northwards into temperate climes.  The conveyor belt is maintained by the sinking of dense saline water whose salinity is enhanced by evaporation and annual sea ice formation.  Meltwater from the Greenland ice sheet and enhanced melting of sea ice could weaken this circulation, causing a decades long cooling of up to 8oC (Vellinga and Wood 2001).

But will it happen?

The proximate reason for the speed and extent of this year’s melt appears to lie with a cyclonic storm that struck the Arctic during August.  This storm broke off large and small segments of decaying ice, accelerating the melt and pushing some of it into warmer waters (Figure 3).

Figure 3.  The action of an August’s cyclone on ice is shown in false colour in this video, courtesy of NASA/Goddard Science Visualization Studio. The visualization shows wind strength and direction; the fastest winds are in red and the slowest in blue.

Needless to say, climate skeptics jumped on this storm as justification for minimizing the importance of this year’s record melt.  Some believe that cyclical weather fluctuations have contributed to the current breakup, and believe that these patterns will eventually reverse themselves.  And the Hadley Centre emphasizes the multiple sources of uncertainty that afflict the modeling of sea ice and Arctic weather in Global Circulation Models (Hewitt et al. 2012).

All this uncertainty leads to competing theories about ice loss, which are summarized by Judith Curry over at Climate etc.  These include:

  1. The greenhouse induced ‘spiral of death’,  in which the positive feedback to albedo produces further melting, and rapid disappearance of late summer ice.
  2. Natural variability alone:  A guy called Joe Bastardi[v] argues that simultaneous occurrence of warm phases in the Pacific Decadal (PDO) and Atlantic Multidecadal Oscillations (AMO) have pumped large volumes of warm water into the Arctic.
  3. The climate shifts hypothesis:  This is basically a hybrid theory that suggests that cyclical changes (as in Theory 2) combined with CO2 forcing have produced the observed breakup. 

Both the second and third theories imply that the high melt rate since 2007 could be reversed “If natural variability is dominant, the sea ice extent could increase if the AO stays predominantly negative, the PDO stays cool, and the AMO switches to the cool phase.

However, Theory 3 is also consistent with the possibility that summer sea ice cover has “flipped” into a new stable state (Livina and Lenton 2012).  Ecosystems that transition from one alternative stable state to another have a hard time making the return transition unless forced to do so by strong forcing variables.  This may mean that thin, small summer ice sheets may become a new normal, that reverse forcings could flip it back to its “normal state”, or that more catastrophic ice loss is to come (Livina and Lenton 2012)[vi].

What does it all mean…..?

Over at Climate etc Judith Curry is strangely complacent about the long-term effects of low summer sea ice.  Among other things, she believes that “For the next two decades, natural variability will trump any direct effects from AGW by a long shot”.  She goes on to list all the factors that could lead to a recovery of late summer ice volume.  These include:

  • ·         Reduction of the sea ice export through the Fram Strait,
  • ·         Reduction of warm water inflow from the Atlantic and Pacific,
  • ·         Fewer clouds in winter and/or more clouds in summer,
  • ·         Less snow fall on ice in autumn and more in spring,
  • ·         No rainfall on snow covered ice before mid June, and
  • ·         Fewer storms in summer causing  ice breakup and more storms in autumn/early winter causing ice ridging/rafting.

Curry goes on to list a number of reasons, some logical and some highly speculative, as to why the loss of late summer ice might not matter.

Judith Curry and other proponents of natural variability may well be right, and ice cover may rebound in a few years.  But given the evidence, there is at least a 50 percent chance that they are wrong and that the trend in Figure 2 will continue into the future, leading to the complete absence of late summer ice within, say, a decade.  In fact I would bet 100 bucks that that is what will happen.

Furthermore, there seems to be no reason why ongoing warming and the sea / ice albedo feedback would not produce an ice-free ocean earlier and earlier in the season as time goes by.  Such a loss of summer ice would amplify the risks posed by the other positive feedbacks I described.  If that is the case, there is a good chance that the loss of Arctic ice during the summer will mark a tipping point for the global climate.

And do we care?

Even if we could know for sure that the loss of summer ice would continue unabated, would we or our political “leaders” care?  Surveys from around the world suggest that most people are woefully ignorant, not just of the science of global warming, but of science in general.  There is no reason for Canadians to buck that trend.  Furthermore, the Arctic is remote enough to be another country, and ecologically, it might as well be another planet for all we know about its biological communities.  And in the mainstream media, record low Arctic Ice extent was relegated to the back pages, and a 25 second item on CBC radio news. 

So if melting Arctic ice is not a hot topic in the Twittersphere, we should not be surprised.

Politicians may be better informed than the public, but may be even more ignorant of what is going on in the Arctic.  According to the CBC, Federal environment Minister Peter Kent “was perturbed” about the record melt, but mostly because he was worried that unpredictable reformation of the ice would mess up navigation: “It is a concern and it is not going to be reversed any time soon,…..We realize that climate change is a significant contributing factor and we have to adapt.”

Unfortunately, part of that adaptation is taking part in the unseemly gold rush mentality that has seized circumpolar nations, as they vie for the rights to bring climate-changing Arctic oil up from the depths and into your gas tank. 

In what moral universe do we exist when politicians can see the effects of climate change, but think only of exploiting the opportunities that come with it? 

And if the record ice melt of 2012, coming out of left field as it did – doesn’t qualify as proof that we needed drastic reductions in greenhouse emissions yesterday, what will?


Brown, R., C. Derksen, and L. Wang. 2010. A multi-data set analysis of variability and change in Arctic spring snow cover extent, 1967–2008. Journal of Geophysical Research 115, D16111: 16 pp.: doi:10.1029/2010JD013975

Francis, J. A., and S. J. Vavrus. 2012. Evidence linking Arctic amplification to extreme weather in mid-latitudes. Geophysical Research Letters 39, L06801: 6 pages. doi:10.1029/2012GL051000, 052012. 

Hewitt, H., S. Bacon, D. Feltham, C. Folland, K. Giles, T. Graham, E. Hawkins, D. Hodson, L. Jackson, S. Keeley, A. Keen, S. Laxon, A. McLaren, M. Menary, M. Palmer, J. Ridley, A. Scaife, D. Smith, M. Srokosz, A. West, R. Wood, and A. Schweiger. 2012. Assessment of Possibility and Impact of Rapid Climate Change in the Arctic. Hadley Centre Technical Note 91, UK Meteorological Office: Hadley Centre, Exeter, UK. <> Accessed on Oct 4th, 2012.

Kinnard, C., C. M. Zdanowicz, D. A. Fisher, E. Isaksson, A. d. Vernal, and L. G. Thompson. 2012. Reconstructed changes in Arctic sea ice over the past 1,450 years. Nature 479: 509-513. 

Livina, V. N., and T. M. Lenton. 2012. A recent bifurcation in Arctic sea-ice cover. The Cryosphere Discussions 6: 2621-2651. 

Vellinga, M., and R. A. Wood. 2001. Global climatic impacts of a collapse of the Atlantic thermohaline circulation. Hadley Centre technical note 26, UK Meteorological Office: Hadley Centre, Exeter, UK.


End Notes

[i]     There is a difference between extent and area.  Arctic ice is a little like Swiss cheese.  There are irregular openings and linear open water leads called polynas that are included in the satellite measurement of what is “ice”.  This is ice extent.  Estimates of “ice area” try to factor out the open water with in the ice pack, but this is quite difficult to do.  Therefore, when newspapers report on extent, they are reporting on the area of ice plus the area of open water contained within the ice pack.

[ii]     Atmospheric Rossby waves are loosely defined as “meanders” in high-altitude winds.  They are large waves in the polar jet stream and the westerlies that extends from the middle to the upper troposphere. They are also associated with mid-latitude cyclones at ground level.

[iii]    Methane molecules trapped in lattices of frozen water.


[v]    Joe Bastardi is a weather forecaster writes over at a site called “The Patriot Post: the voice of essential liberty”.  This places him firmly in the “Republican skeptic” wing of climate skepticism.  Among other things, he believes that CO2 has little to do with global warming and that we are in for several decades of global cooling

[vi]    Skeptics love to point out to the short time series in the satellite record and have anecdotal “records” that ice extent has been lower in the past.  On the other hand, Kinnard et al (2011) use “proxies” from ice core records, lake sediments tree ring chronologies, and two historic series of sea ice observations to find that the “duration and magnitude” of the current loss of sea ice has no historic precedent for at least 1,450 years (Kinnard et al. 2012). 

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s