Arctic Ice June 2011
    By Patrick Lockerby | June 18th 2011 07:02 PM | 25 comments | Print | E-mail | Track Comments
    About Patrick

    Retired engineer, 60+ years young. Computer builder and programmer. Linguist specialising in language acquisition and computational linguistics....

    View Patrick's Profile
    Arctic Ice June 2011

    The sun is the main driver of climate and weather.  The Arctic solar year could be plotted as a graph of the angle of the sun relative to the horizon,  negative in winter and positive in summer.  The graph would be a very neat waveform which varied little in amplitude or in frequency on decadal timescales.  By way of contrast, any graph of any effect of the sun's relative angle will vary from year to year.  Although the sun will warm the air, land, sea, snow and ice, it will never do so in exactly the same way during any two years. 

    In order for any climate or weather pattern to repeat itself in timing or scale, every micro-component of the climate system must behave in exactly the same way during one year that it did in a previous year.  That can never happen.  The reasons why it can't happen are to be found in astronomy and mathematics.

    Our planet's orbit imposes a periodicity on the sun's direct influence on climate, but the various components of the climate system each have their own inertia and each has an element of chaos in its behavior.  The result of these elements of chaos and inertia is that the climate is only broadly periodic.  The climate system exhibits a sensitive dependency on specific ranges of conditions and will change substantially if any one or more of those ranges is changed.

    Fortunately, although our planet's climate system is not perfectly periodic, it is nearly so.  This means that the unpredictability is constrained within limits.  Those limits are set by the physical attributes of the system.  If any attribute is changed by a sufficient amount, the limits of the system will change.  When the climate system moves outside its previous limits it tends to become more chaotic.  The system will still be subject to any controlling conditions which remain unchanged, but will be pulled towards new conditions.  For a time the system will fluctuate between wider limits than before, and then it will enter a new regime with new limits.

    It is fundamental to planetary physics that no body in space can ever repeat its trajectory exactly.  Although the paths of astronomical bodies are broadly dictated by the physical laws of gravitational systems, the  location and velocity of any body at a particular future time cannot be calculated precisely.  As bodies move in response to gravitational attractions, any small perturbation can affect future states.  Minor and seemingly random perturbations can never be predicted without perfect knowledge of every tiniest component in a system.  Since we can never have perfect knowledge, all we can do is to show regions in space within which a celestial object will almost certainly be found at some future time.

    Butterflies and chaos

    In 1952, the science fiction writer Ray Bradbury published a short story: A Sound of Thunder.  It is a story of time travel in which a butterfly has a major role.  It was the mathematician and meteorologist Edward Norton Lorenz, a pioneer of chaos theory, who popularized the idea that a mere butterfly could affect weather1.  The term 'butterfly effect' is now part of our culture.  Lorenz showed that planetary atmospheres exhibit quasi-periodic behaviors which, although fully deterministic, are not predictable beyond a small range. 

    Many physical systems can be described in broad terms by simple mathematical models, but their long term behavior is highly sensitive to initial conditions.  Given that our planet's orbit is such a system, that our atmosphere is such a system, that our oceans form such a  system and that the ice-albedo interaction may be described as such a system it is surprising indeed that we can forecast both weather and climate with the degree of precision required by risk takers2.

    Arctic chaos

    Although the earth's annual orbit is never the same twice, nevertheless a plot of superimposed yearly graphs of sun angle for any latitude will be almost a single plot, almost a single sine wave.  However, a plot of superimposed annual measures of any effect caused in that latitude by solar heating will show the wide variations in effect which arise from atmospheric and oceanic circulation and from many other causes.  The graph below shows this variability quite plainly.

    IJIS AMSR-E sea ice extent, June 17 2011

    The IJIS graphs of ice extent are plotted on an x axis which represents the orbital year, but that is an arbitrary choice of 'signal lock'.  If we would align the graphs so as to superimpose all maxima, or all minima, or all crossings through some arbitrary level, then we would see quite clearly that the wavelength and amplitude both fluctuate.  Sine waves of a fixed frequency would all cross each other at two common points, regardless of amplitude, as in the graph below.


    The effect of inertia in the components of the Arctic climate system is to vary the frequency - the exact time between maximum and minimum ice extent.  The bunching together of the IJIS extent  plots in two places - May and November - corresponds to that variation in frequency.  The variation in frequency is of interest.  The bunching is of no significance in itself: it is merely a displacement of what would be zero crossings in plots of pure sine waves having a common frequency but a varying amplitude.

    Overlaying annual graphs helps to show the range within which one might expect to find future ice extent on any given day.  However, a superimposed graph does not help much in seeking out evidence of Arctic regime change.  Chaos theory3 suggests that if the Arctic is undergoing a regime change then we shall see evidence of this, as a stable regime transitions through wild swings to a new kind of stability.  An examination of Arctic ice anomalies from 1979 to 2011 shows - I suggest - evidence of such regime change.

    Ice extent anomalies 1979 - 2011
    full size image available from source: The Cryosphere Today

    The Cryosphere Today graph of ice anomalies is based on the 1979 to 2008 mean.  The overall trend is obviously downward with respect to that baseline.  There is also - I suggest - another trend: a trend towards instability.  From 1979 to 2006 inclusive, each annual anomaly range spans close to 1 million km2, apart from 1996 when the range was closer to 2 million km2.  In 2007 and in each year since, the range has been wider - close to 2 million km2.  This sudden and continued change from average behavior suggests to me that the Arctic is now passing through a period of regime change.  If that is true, then the annual melt pattern for 2011 onwards will resemble the patterns of 2007 through 2010, or else the melt will be more extreme.

    The Arctic ice extent is currently the lowest ever for this date - June 18 2011.  If the current trend continues then the September minimum could easily be lower than in 2010.  There are strong indications that the current trend will continue.

    Arctic climate regime change

    The geographical annual and diurnal ranges of temperature would be partly smoothed away, if the quantity of carbonic acid was augmented.
    But ... I incline to think that the secondary action due to the regress or the progress of the snow covering would play the most important role.

    On The Influence of Carbonic Acid in the Air upon the Temperature of the Ground.
    Svante Arrhenius, April 1896
    Arrhenius was writing about snow cover, not the Arctic.  However, his idea of ice-albedo dominance over CO2 effects would seem to be sound.  Until, that is, the accumulating effects of rising CO2 reduce Arctic winter ice and snow growth to a level below that required for climatic stability.  The continued annual reduction in the Arctic's albedo and thermal mass must lead to a time when it is the effects of atmospheric CO2, rather than albedo, which dominate the Arctic climate system.  That would be a new regime in which the annual temperature range of the Arctic would be less than historic norms.

    In passing from a regime dominated by ice-albedo effects to a regime dominated by the effects of increasing atmospheric CO2, the dynamical system must pass through a chaotic regime.  I suggest that we are witnessing that regime.  Although this year, 2011, saw prolonged low air temperatures in the western Arctic, that has not retarded ice retreat substantially as compared with 2010.  The ice in the Canadian Archipelago, although of slightly greater extent than at this time in 2010, is covered in meltwater pools and is breaking up rapidly.

    Canadian Archipelago June 18 2011

    The ice bridge in Nares Strait was much more robust than last year and lasted much longer.  At this time last year, ice was passing from Lincoln Sea through Nares Strait.  The strait is still blocked at this time, but the ice bridge has collapsed, so the Strait should be a conduit for ice by July.  As an indirect effect, I suggest, of ice transport through Fram Strait, Nares Strait and the Canadian Archipelago, the main pack is relieved of some pressure.  This would allow the pack to spread out more, with consequent enhanced melting.

    Both the NWP and the Eastern passages will almost certainly be as freely navigable this year as they were last year.  Along the Siberian coast, the only ice still largely attached to land is very thin, covered in meltwater pools and in process of breaking up.

    Ice on the Siberian coast

    Much depends on the Arctic weather, but it looks likely that the September ice minimum will be amongst the three lowest.  If the melt in July and August proceeds as it has done on average over the last decade, then the 2007 record minimum may well be beaten.

    [1] - Edward Norton Lorenz,
    Does the Flap of a Butterfly's Wings in Brazil Set Off a Tornado in Texas?

    [2] - risk takers: people who require accurate information about future weather and climate in order to minimise the risk of some form of loss and/or to maximise the chance of some form of gain.

    [3] - for an excellent introduction to chaos theory, I recommend the book Chaos: Making a New Science, by James Gleick.


    Very nice introduction to the chaos theory approach on climate change - or at least variation in ice cover.

    The shipping industry is taking the changing climate seriously and are encouraging better sea ice forecasting in the Arctic. The Arctic route will save lots of money for this transport sector - if we can develop reliable forecasting system. I am only mentioning it to illustrate that adaptation is and will always be a part of the picture.

    Regardless of how bad we manage our planet....
    Bente Lilja Bye is the author of Lilja - A bouquet of stories about the Earth
    I was wondering how the recent Icelandic volcanic eruption would affect solar input for the arctic this year - it was reported to reach 60,000ft so if it put much sulphur dioxide up it might take some time to fall out, keeping things cooler. Then of course there was the ash which would darken any snow/ice it fell on, helping it melt!
    More variables for this years melt, equivalent to very many butterflies pulling in different directions?


    Meltwater at the North Pole.

    Hi Patrick,

    An excellent article, again.

    I would query whether the abnormally blue ice in the Canadian Archipelago and the East Siberian Sea is so coloured because of meltwater ponds.

    Certainly, in the Canadian Archipelago, there are substantial patches of blue ice that are apparently land-based glaciers, and water does not pool very well on a gradient.

    I'd suggest that this blue ice is actually ice that is completely deficient in snow-cover, which is unusual enough in itself, but not necessarily covered in melt ponds. It should have a higher albedo than snow-covered ice.

    Whether the absence of snow cover is due to an early melt, or a lack of precipitation over winter, I simply don't know.

    I am absolutely certain that you have much more experience in examining satellite images than me, so would be interested in your opinion.

    Yes the light blue is seen both north of Canada ,Alaska and Siberia. I have never seen to that extend anytime earlier in my time of satellite image observations. But I dont believe it is meltponds, then maybe more melt seas, because their size, it could also be the ice is so thin without cracking and the non existing snow cover off course

    At least 2 melt ponds are seen next to webcam # 1 at the pole:

    Patrick, I've been waiting for your ice update, good as usual. It is so interesting that june extent this year almost follows (and leads) extent in 2010. But, on the other hand, last year we had big drop in Hudson bay area. This year, most noticeble drop is in Kara sea. Also, there is a little bit more ice in Beaufort sea, but on the other hand, we less ice in Chukchi and Laptev sea. Also, at least for today, we have quite warm waters in those areas, mainly in Chuckchi sea (which can speed up melt in that area).

    @Espen: According to Neven's blog (and NOAA web page), these are the earliest melt ponds:

    @Espen: Neven has also created nice animation of Nares strait ice bridge breakup:

    The "Nares Train" will soon be on the move, on the latest modis images a coast to coast crack is clearly seen across Kane Bassin.

    If compare this year to 2010, we see that now the ice is melting much more active in most areas.

    Somehow I doubt the different (underestimated) figures coming from normal sources, my feeling and observations from satellite images tells me something else, are we more in this category?
    I think this summer is very warm in the Arctic - 5-degree anomaly (pink) in the Arctic is becoming more permanent.

    Now there are melt ponds around webcam #1 and rotten snow around webcam #2 at the pole, so the spring has finally arrived up there!

    Hi Patrick

    When comparing the The Cryosphere Today :

    I can not see that sea ice extend/area is at the same level as last year, as figures show, to me it looks like there is a lot less ice cover this year than in 2010!


    It looks to me that the break up of ice at Barrow, Alaska is in progress it may already have happened or it will happen very soon, watching the today's first modis images from the scene.


    USCGC Healy is now entering the Bering Sea:

    To follow it by web cam:
    Or position:

    Hi Patrick

    I have some questions I hope you can answer: When I look at The Cryosphere Today: Compare Daily Sea Ice
    I can confused because sea ice cover (area/extend) on June 24 this year compared with June 24 1010, looks much much less than the figures shown from several other sources. And The Cryosphere Today images to me looks more or less what I can see directly from the Modis images, so how come there is such a big difference at least in my oppinion?
    Regards Espen

    It's interesting to see you devote so much time to describing the many natural cycles we can see clearly from a personal time frame, [daily, monthly (lunar), annually] and note some possible longer term perturbations with respect to astronomically movements. Chaos, such as clouds and wind then cover the rest. It's unfortunate that your assessment ignores the longer term thermal cycles that have been observed in sun and ocean behavior, which have cycle frequencies that are much more difficult to assess in our personal time frame, roughly 11 years and 20 to 30 years. There are even evidence of cycles with frequencies of hundreds of years and 10's to 100's thousands of years . The interplay of all of these cyclical behaviors, each with different time constants and amplitudes create climate behavior. With such cycles evident in the data, it's always amazing to me to see behavior extrapolated as a linear function. It's like drawing a line between the temperature reading at 6 am and 11 am on any given day and using the line to predict the temperature for the rest of the day. The line will be close at noon and 1 pm, but the prediction won't look very good for 9 or 10 pm.

    The block at the Nares Strait that was stronger than you anticipated showed up with a uniform blue tint, just as the ice in the northern throat of the strait near the Peterman glacier now has a blue tint. It's the composite ice that's been broken and refrozen that still looks white. The blue tint of the ice throughout the archipelago is NOT an indication of inherent weakness. Note that the MOSAIC images and the CIS plots still show the archipelago ice as FAST ICE. It is not broken and loose the way the CT data would suggest. The CT images and composition estimated are biased low due to the coloration and shadows from clouds. CIS data indicates the Canadian Arctic is progressing at a long term normal rate with current ice levels at a 10 year high.

    The Nares Strait remains blocked and will not be leaking old ice from the basin until mid July. That's precursor to higher ice area for 2011.

    @ Crashex

    Cycles? Chaos? Higher ice area for 2011?

    Quite frankly you display the capacity to learn quite a lot from Logicman, as it is evident there is a a lot you don't know.
    Now that's not necessarily your fault, as you may be a late-comer to the science of Arctic Sea ice and analysis. I recommend the following links by Tamino for better learning about the Arctic Sea ice and things cyclical:
    Arctic Sea Ice 3-D
    More Magical Cycles



    What are you saying?

    You don't believe there are any natural cycles? Day v. Night, Summer v. Winter, El Nino v. La Nina, Glacier v. Interglacial.
    Which ones are bunk?
    Or, is it when Tamino says that a 30 year cycle can't be proven statistically because there is insufficient data, you assume that there is no such cycle. Those are two different things, right; statistical proof v. existence. You pefer the 1979 is a Magic Year that happens to correspond to both the start of satellite data, the through of the long term temperature record and, fortuitously, the "tipping point" for the new climate dominated by CO2 behavior.

    You don't believe that Chaos Theory influences the predictability of the weather and climate?

    Do you think that the blockage of the Nares Strait for 3 months longer than 2010 creates a feedback that leads to less ice in 2011? What would that be? I figure that if the ice stays in the arctic basin longer and doesn't get transported south to a warmer environment, it's less likely to melt. You think I have that wrong? It's clearly not the only thing that would influence the ice area for the year, but it does push it in that direction, right?

    Do you think that the ice persisting throughout the archipelago longer this year than last year is an indicator of faster melt? Doesn't the lower Albedo of the ice over that region that persists through June, past the solstice, necessarily mean that more solar energy is reflected and less becomes absorbed by the water? That's just the other side of the feedback coin that I've heard so much about, i.e. more exposed water due to ice melt causes the sun's heat to be absorbed and enhances further melting.

    I know I have a lot to learn. And I read through most of Tamino's stuff, as well as the rebuttals to it.

    I think we all have a lot to learn.


    With regards cycles. Aside from those driven by the Earth's tilt, there aren't actually many real cycles in the atmosphere/ocean system.

    There is a lot of quasi-periodic behaviour, the ENSO being a case in point. ENSO is not cyclic in the same way the seasons are, which is to say that whilst the fourier transform of the ENSO shows a peaking in the region of 4-6 years there isn't any mechanism per-se that demands a cyclical rebound. Indeed paleo-evidence shows that the ENSO has been 'stuck' in El Nino state for prolonged periods in the geological past. Another case in point is the AMO whose periodicity is merely apparent, and is actually the outcome of the 1940s warming, and the post 1975 warming causing changes in SST. Modelling studies and examination of changes in radiative forcing show the causes of those 2 warmings to be rather different.

    The ice-ages are clearly down to changes in the Earth's orbital configuration. Roe has shown that the first derivative (w.r.t. time) of insolation at 65degN correlates very closely with ice-sheet changes. However whilst the individual components are cyclic (e.g. precession, obliquity) the outcome in terms of the Earth's climate is not as the individuals factors aren't harmonics.

    What is happening in the Arctic is clearly due to human activity; largely CO2 and other greenhouse gasses, and soot. But it is being amplified by resultant changes such as ice-albedo, intra-ocean and ocean/atmosphere heat fluxes, atmospheric humidity, changes in winds (due to the Arctic Oscillation and Arctic Dipole).

    For what it's worth, as far as I'm concerned Nares Strait is irrelevant. The volume of sea-ice flowing through it is negligible in the scheme of the Arctic. I'm expecting this year to be 4.5M (+/-0.5M) km^2, longer term there will be no recovery to a pre 2007 state.

    The sea ice at Kimmuret crashed today, and the grass is green now:

    I think we could easily get ice free below 75 north this season.

    Thank you all for your comments.  I apologize for not being able to respond point by point, but I have taken all points on board.

    My July article is now published: Arctic Ice July 2011.