Slightly Increased 2012 Antarctic Sea Ice Levels No Match for Arctic Declines

The year’s increased Antarctic sea ice levels cannot be seen as undercutting the long-term record decline in Arctic sea ice and the global sea ice decline generally.

Arctic sea ice extent this past September reached the lowest point recorded since satellites first started measuring sea ice in 1978. Arctic sea ice blew past 2007′s record low, ending at 3.37 million square kilometers, roughly half the size of the summer minimum ice cap during the period from 1978 to 2000.

At the same time, sea ice around Antarctica has been approaching a record high. The Antarctic situation has led some to dismiss the dramatic events in the Arctic as a simple fluke, pointing to the growth of sea ice around Antarctica as a counterpoint. What’s missing from that comparison is that the modest growth of sea ice around Antarctica in no way compares to the dramatic 2012 declines seen in the Arctic nor, even more importantly, over the past decade.

The National Snow and Ice Data Center (NSIDC) at the University of Colorado, Boulder, provides an authoritative dataset on both Arctic and Antarctic sea ice cover, using data from NASA satellites. The figure below shows 2012 Arctic and Antarctic sea ice cover (as of October 8th, 2012) compared to the typical values from 1978-2000 for the same days of the year. The typical values are shown as a grey band representing the 2.5 and 97.5 percentile confidence intervals with a dotted line showing the average value.


View larger image

The data make clear that the changes in Arctic and Antarctic sea ice cover are not remotely comparable. While Antarctic sea ice is high, it is barely outside of what would be considered normal based on the 1978-2000 period. Arctic sea ice, on the other hand, is barely half of what it was three decades ago.

To better represent the change in sea ice cover over time, it helps to convert the daily sea ice values for the Arctic and Antarctic into anomalies relative to a 1978 through 2000 baseline period. These anomalies allow scientists to remove the large annual cycle in sea ice to reveal the underlying trends, as shown in the figure below. The tiny red and blue dots represent all daily sea ice anomalies for the Arctic and Antarctic respectively, and the solid lines represent a smoothed fit to the data.


View larger image

Seen side-by-side, the vast differences in changes between Arctic and Antarctic sea ice become readily apparent. The smoothed data actually somewhat under-estimate the magnitude of the change in minimum ice cover, as Arctic sea ice tends to recover to nearly normal levels during the winter somewhat independent of the magnitude of the summer melt.

It is instructive also to compare the smoothed data to more easily see differences in longer-term changes in sea ice cover between the poles:


View larger image

While Antarctic sea ice lately has indeed been on the high side of normal, the situation is in no way comparable to the dramatic declines seen in the Arctic. Furthermore, what is important from a scientific standpoint isn’t so much the record lows seen this particular year, but rather the longer-term trends over the decades. The long-term trends point to a clear and significant downward trend in Arctic sea ice extent, which accelerated starting in the late 1990s. While there is a modest upward trend in Antarctic sea ice, that increase makes up for only a fraction of the decline in the Arctic, and global sea ice as a whole has been decreasing.

The actual data makes it hard to conclude that those wanting to point to the Antarctic as a counterpoint to what is happening in the Arctic may simply be trying to change the subject from the recent unprecedented global sea ice declines.

Zeke Hausfather

Zeke Hausfather, a data scientist with extensive experience with clean technology interests in Silicon Valley, is currently a Senior Researcher with Berkeley Earth. He is a regular contributor to Yale Climate Connections (E-mail: zeke@yaleclimateconnections.org, Twitter: @hausfath).
Bookmark the permalink.

33 Responses to Slightly Increased 2012 Antarctic Sea Ice Levels No Match for Arctic Declines

  1. Joe Witte says:

    Good story. Prof Francis of Ruters points out the potential contextual consequences of an open and warmer Arctic Ocean. While we have the Ozone Hole in the sky over the S Pole we have hole in the Ice in the water over the NPole.
    Joe

  2. Bruce says:

    Zeke, the mean for 2012 is above 2011 will end up above 2007 the way it looks.

    For many weeks in April and May sea ice extent sat near 98% of the 1980s average.

    What do you make of that?

    And what about the cyclone?

    • Zeke says:

      Hi Bruce,

      As you know, Arctic sea ice tends to recover to around the same winter extent regardless of summer melt. However, the volume of ice has shrunk considerably given that the new freeze is relatively thin (see PIOMAS here: http://psc.apl.washington.edu/wordpress/research/projects/arctic-sea-ice-volume-anomaly/ ). Due to this, its useful to look at the trend in summer arctic sea ice independently from the trend in average annual sea ice.

      The second and third graphs in my article show anomalies for each day for Arctic and Antarctic sea ice, and the overall trends are clear using both all observations as well as summer-only observations (though it is undeniably more dramatic in the summer-only case).

      • Bruce says:

        “As you know, Arctic sea ice tends to recover to around the same winter extent regardless of summer melt. ”

        Not true recently. In Apr/May 2012 was 1 million sq km higher than 2007 and for a short time above the long term average.

        If it was for 2012′s cyclone, the recovery of the maximum in 2012 would have been the story.

        • Gerald Wilhite says:

          Sadly, I have to disagree. I suspect that most observers would agree that if it had increased the AGW propaganda machine would have made sure there was no story at all.

  3. David says:

    Bruce –
    I’ll have to second your question to Zeke… – as I’m quite interested in his response: –
    many weeks in April/May ’12 Arctic sea ice extent near 98% of 80′s average.

    the rather Large Cyclone (how large? – Google for references)…wind driven ice…(vs. ‘melting ice’)

    I’m very curious as to how these realities might be reconciled with his analysis in this article.

    • Zeke says:

      David,

      As I mentioned earlier, much of the arctic refreezes every winter, simply because there is enough time when temperatures are well below freezing. However, given the dramatic summer melts relatively little thicker multi-year ice accumulates, so overall volume has shrunk dramatically over time.

  4. Bruce says:

    Joe, there was an Arctic Ozone hole as big as the Antarctic one in the spring of 2011.

    http://earthobservatory.nasa.gov/IOTD/view.php?id=60598

    Because the stratosphere was really really cold for a long time.

  5. dodgy geezer says:

    “..The actual data makes it hard to conclude that those wanting to point to the Antarctic as a counterpoint to what is happening in the Arctic may simply be trying to change the subject from the recent unprecedented global sea ice declines…”

    I’m curious – why doesn’t this work the other way around, so that those pointing to the Arctic alone, and ignoring the Antarctic, may be simply trying to alarm people by pretending that the loss of ice is greater than it really is…?

    • Zeke says:

      Dodgy,

      Thats why its useful to put them side-by-side and look at the complete picture, something this article attempted to do. The magnitude of sea ice changes between the two really aren’t comparable.

  6. Nullius in Verba says:

    Zeke,

    The reason for making a fuss about Antarctica is not about the size of the effect. It’s about the logic of the argument, and the media bias in reporting.

    The usual argument in public debate for the significance Arctic ice goes something like: global warming predicts a melting of polar ice, we have observed a melting of polar ice, therefore global warming theory is true. This is an argument form known as “confirming the consequent”, and it’s not valid. Sometimes the argument goes further: the melting observed is more than was predicted, therefore the situation is worse than we thought. This also is not valid.

    The Antarctic is mentioned as a logical counter-example to demonstrate the invalidity. Melting was predicted but freezing was observed, therefore global warming theory is false. The melting observed is less than was predicted, therefore the situation is better than we thought. Obviously climate scientists don’t think so, so we need to consider the reasons for the difference.

    IPCC AR4 WGI Fig 10.13 shows projections for future summer and winter sea ice extent anomalies at both north and south pole. It can be seen that the predictions in the current timeframe are similar, and both a slight decline. It’s noticeable that the past is a lot flatter than the record (such as it is) for the 20th century, and this is presumably because the projections show the ensemble averages, without the natural variation on top.

    So the first problem is that we have to separate natural variation from the AGW trend. An obvious possibility is that the Arctic ice decline is mostly natural, with a much smaller anthropogenic component. That’s difficult to do, especially without relying on models, as the IPCC’s chapter on attribution makes clear.

    Assuming that’s been done, then the next problem is that you’ve just falsified the IPCC’s models. The IPCC have calculated the warming/melting that should occur, and it’s wrong. So you have proved the existence of unknown and unmodelled effects whose outcomes you can’t predict. How do you know which way they will push the result? The Arctic suggests more melting, the Antarctic less. On what basis have you accepted one and not the other?

    And now there are a number of post-hoc hypotheses being proposed to explain the difference. Antarctica is land surrounded by ocean rather than vice versa. It may have something to do with the ozone hole. It may be the polar vortex isolates Antarctica from the rest of the world’s weather better. And that’s fine, that’s how theory progresses, but now you’re back to square one on the validation of the new theory. You can’t use the same data you used to generate the hypothesis to test it.

    Essentially, the purpose in bringing up the record-breaking Antarctic sea ice is to force the climate scientists to have to explain all this: to say that sea ice is not a direct proxy for the truth of global warming catastrophe. This was the same way a couple of snowy winters were used to force an admission that local, short-term weather was not proof of climate change, as was usually implied previously every time there was a heat wave or a drought.

    This sort of thing has a three-fold effect. It exploits the weather-is-climate connection that the greenhouse activists have previously set up to counter the message directly. As noted, it forces the activists to explain more of the science and uncertainties, weakening that link. And it highlights the selectivity about evidence and constantly changing arguments used by the activists in their attempt to interpret every possible outcome as support for their theory. 15 years decline in sea ice is a trend, but 15 years flat line of the temperature means nothing. Arctic low sea ice records are dramatic evidence, Antarctic high sea ice records mean nothing. Hot weather is caused by global warming. Cold weather means nothing. Or as some people are trying to say, cold weather is caused by global warming too.

    The point is not that the record Antarctic ice extent is particularly big. It’s that nobody had even heard about it until the sceptics said something. That’s the problem you need to address.

    • Zeke says:

      Nullis,

      I agree that pointing out individual melting extremes can be problematic, as there is weather noise in the record. However, the overall trend is unambiguously down, and rather dramatically so. I know of no proposed decadal-level natural variability that can account for this decline.

      That said, it doesn’t speak immensely well for climate models that they have underestimated Arctic declines and overestimated Antarctic ones. Climate models are imperfect mechanisms, and scientists should (and are) using these mismatches between predictions and observations to diagnose areas in which models can be improved.

      • Nullius in Verba says:

        Whether there is an unambiguous trend depends on the statistical model you use. If you assume linear plus iid Gaussian noise, you can draw on a convincing-looking trend line, but there are infinitely many alternatives.

        Personally, it looked to me like a step change. There’s a small step down in 1996, then from 2003 to 2007 it ramps down, then from 2007 on it looks level but a lot noisier than previously. The lower mean and larger variance means we’ll get more record lows, but there’s no particular reason to think it will keep on going.

        The fact that the transitions are so sharp (e.g. the variance jumping up in 2007) indicates to me that this isn’t just a gentle linear slide. Something changed.

        Of course, I agree it could be a gentle slide with autocorrelated noise on top. Or it could be some sort of ARIMA with no trends or step changes. Or maybe some sort of Lévy flight model is appropriate, the apparent step being just a bigger random variation.

        The brain finds patterns, which mathematics will then tell you are plausible, but they’re in your brain, not in the data. They’re in the priors. As Feynman said: “The first principle is that you must not fool yourself and you are the easiest person to fool.”

        Unfortunately, the general public has little experience with the range of mathematical possibilities, so when they see the line going down, they assume a trend. Similarly, when they see the line level off (as with global temperatures), they assume the trend has stopped. It cuts both ways.

        So yes, with a suitable trend-plus-noise model, the trend appears to be steeply down. But given that it’s a lot steeper than we were expecting, shouldn’t we at least consider other possibilities?

  7. Nullius in Verba says:

    Incidentally, what do you think of the sharply changed variability at the end of the anomaly record?

    It suggests to me that the monthly normals might have changed.

    • Zeke says:

      Nullis,

      I believe that its primarily due to the relatively consistent winter extent (because temperatures are still far below freezing in most areas) coupled with the large declines in summer sea ice, leading to larger annual oscillations. Volume, at least as measured by PIOMAS, is much more consistently down.

      • Nullius in Verba says:

        Agreed. I suppose what I was wondering was why does that not happen in any year up to 2007, and does in every year thereafter?

  8. citizenschallenge says:

    Nullius says:

    ~a~
    “The reason for making a fuss about Antarctica is not about the size of the effect. It’s about the logic of the argument, and the media bias in reporting.”
    {…}

    ~b~
    “Essentially, the purpose in bringing up the record-breaking Antarctic sea ice is to force the climate scientists to have to explain all this: to say that sea ice is not a direct proxy for the truth of global warming catastrophe.

    This was the same way a couple of snowy winters were used to force an admission that local, short-term weather was not proof of climate change, as was usually implied previously every time there was a heat wave or a drought.”
    {…}

    ~c~
    As noted, it forces the activists to explain more of the science and uncertainties, weakening that link.
    {…}

    ~d~
    15 years decline in sea ice is a trend, but 15 years flat line of the temperature means nothing.
    {…}

    ~e~
    Or as some people are trying to say, cold weather is caused by global warming too.
    {…}
    =====================

    ~ ~ ~

    ~a~
    “the logic of the argument”
    Actual the logic of the climatologist’s argument (as opposed to your fiction) is that atmospheric CO2 has known “insulating” properties – they still don’t have the exact number, but they have been narrowing that range to a point were any intellectually honest person must accept… and worry about!

    You Nullius instead imply climatologists use Arctic melting as their basis for claiming we are inflicting dangerous (so far as society is concerned) levels of global warming to our Earth.

    This is false !
    ~ ~ ~

    ~b~
    Then you bring up extreme local weather events and swap media sensationalism for climatological understanding.

    Climate scientists point to these events not as proof but as being symptomatic of global warming. The proof will be there when it’s too late to do anything about it.

    PS – Climatologists have not been ignoring Antarctica.
    http://www.newscientist.com/article/dn16988-why-antarctic-ice-is-growing-despite-global-warming.html

    http://www.livescience.com/23333-record-high-antarctic-sea-ice-levels-don-t-disprove-global-warming.html
    ~ ~ ~

    ~c~
    Here you claim “activists” are the same as serious scientists.

    PS. Any sincere effort provides a deluge of serious scientific information explaining the nuances of what’s happen.
    ~ ~ ~

    ~d~
    Here you claim that cherry picking surface air temperature data can represent our globe.

    And why ignore what’s happening in our oceans ?
    http://www.uctv.tv/shows/135-Years-of-Global-Ocean-Warming-Perspectives-on-Ocean-Science-23999

    135 Years of Global Ocean Warming – Perspectives on Ocean Science
    Dean Roemmich
    ~ ~ ~

    ~e~
    Here you show a disregard… and disinterest, in understanding our “climate” as our planet’s Global Heat Distribution Engine.

    Namely, that air and ocean currents push air masses around
    and that shoving warm air into cold arctic regions, forces that cold arctic air to go somewhere else.
    ~~~~~~~~~~~~~~~~~~~~~~~~~

    Why not a bit more interest in learning about the ever expanding scientific understanding of the dynamics of our Global Heat Distribution Engine.

    All that hand waving above ignores the known physical properties of atmospheric GHGs and the fact that our planet’s atmosphere is holding in more heat, both on land and in the oceans.

    You misrepresent what climatologists have been reporting to us with a fiction writer’s license.

    PS.
    An Update on Earth’s energy balance in light of the latest global observations
    https://www.dropbox.com/s/8gkbto14cy8ip2y/ngeo1580.pdf

    • Nullius in Verba says:

      Thanks, but just because I don’t mention it in every comment doesn’t mean I have no interest in it or don’t know anything about it. I don’t see how you can deduce my “disinterest” (or “uninterest”) from a couple of comments.

      What I’m complaining about is that people who purport to represent the mainstream science in the media practice that sensationalism, and use the Arctic as proof, and the serious scientists let them. They present events that scientifically-speaking prove nothing as strong evidence to fool the public. And if anyone complains about the media sensationalism, the nuances are pulled out of the cupboard and you act as if that was what was being attacked all along.

      If you want to avoid being associated with media sensationalism, then you need to stop it happening. Global warming may become an issue for the Arctic around 2080, and until then any changes (such as we see now) are more likely to be natural variation. Attributing causes is not yet possible. That’s the science. Stop using it as sensational “evidence” to support the case.

  9. Mack says:

    citizenschallenge,

    “PS.
    An Update on Earth’s energy balance in light of the latest global OBSERVATIONS”

    No CC this has not been observed. What is observed is an incoming yearly average solar irradiation of about 1366w/sq.m. and this has been observed since 1902.( by water flow measurement, then and lately by sattelite) . Your geometrically arrived at figure of 340.2w/sq.m. of incoming solar irradiation in that Earth’s energy budget diagram is just as wrong as the Kiel Trenberth diagram here….
    http://www.ipcc.ch/graphics/ar4-wg1/jpg/faq-1-1-fig-1.jpg
    Both have an “incoming yearly average?” at the TOA of about 340w/sq.m which is incorrect. To understand this you are going to have to read the science here…
    http://jennifermarohasy.com/author/nasif-s-nahle/

    • Mack says:

      I notice Trenberths diagram has 342w/sq.m. and your latest has 340.2w/sq.m.
      Shiftin’ the decimal about ain’t gonna fool nobody C.C. ;)

      • Mack says:

        What you have is misunderstanding and concoction “IN LIGHT OF the latest global observations” Note it is not “following” or “as a result of” but just “in light of”.

  10. RickA says:

    One thing that confuses me about the flap over the Arctic vs. Antarctic ice loss is that Arctic melting cannot raise sea level (as it is floating on water), while antarctic melting can raise sea level, because most of the ice is on land.

    Isn’t that correct.

    Granted – massive ice melting in the Arctic has something to say about warming temperatures (at least in part), but very little to say about sea level rise – right?

    So doesn’t this latest news actually lower the concern about SLR? Zero plus lower sea level rise = lower sea level rise – right?

    • Nullius in Verba says:

      Not quite.

      Sea ice doesn’t affect sea level significantly. (Technically, there will be a very slight change due to salinity differences, but it’s not detectable.) We have been talking here about both Arctic and Antarctic sea ice, so there’s no difference there.

      People elsewhere have also been talking about land-based ice in both the Arctic (Greenland) and the Antarctic, although they didn’t come up in this particular discussion. If the mass of land ice at either pole were to increase or decrease then sea level would decrease or increase respectively. (Although really it’s not sea level as such we should be potentially concerned about but its rate of change. Otherwise the land moves to keep up with it.)

      However, people are concerned about the Arctic sea ice for other reasons besides sea level. That’s another topic, though.

  11. citizenschallenge says:

    Mack,
    who is Kiel Trenberth?

    And… can you give some links or other references to your claim:

    “No CC this has not been observed. What is observed is an incoming yearly average solar irradiation of about 1366w/sq.m.”

    I would like to understand what you are trying to explain to me.

    • Mack says:

      Citizenschallenge,
      Sorry I wrote in a little bit of haste. It was Jeffery Kiehl and Kevin Trenberth. Trenberth is the older of the two and therefore his minion Kiehl is less culpable, shall we say, of this “travesty”.(my words) :)
      From your link …..
      “At the top of the atmosphere, this balance is monitored globally by sattelite sensors that provide measurements of energy flowing to and from the Earth. By contrast, observations at the surface are limited mostly to land areas. As a result, the global balance of energy fluxes WITHIN THE ATMOSPHERE or AT THE EARTH’S SURFACE CANNOT BE DERIVED DIRECTLY FROM MEASURED FLUXES, and is therefore uncertain.”
      So what they are literally saying C.C. ,(or confessing), is that they are not taking into account any direct measurement of fluxes in the atmosphere or on the land. All the readings they consider is sattelite readings at the TOA. The sattelite readings at the TOA are in reality about 1360 w/sq.m.of incoming radiation. This is the “solar constant” and is a yearly average for the whole globe at the TOA. It cannot be divided. It has to remain simply as that, and that has simply got to be translated down to about 340w/sq.m.average yearly solar radiation at the Earth’s surface.
      You just have to keep reading Nasif Nahle or you might go back to Searing Heat, Raging Wildfires Offer “Teachable” Summer Moments by Bud on Sept. 5th on this site and read my comments.

  12. citizenschallenge says:

    RickA,
    Is sea level rise all society has to worry about?
    ~ ~ ~

    Oh incidentally, about that erratic sea level rise…
    I recently watched an astounding video (well personally, it was astounding) lecture by Jerry Mitrovica PhD.
    given at Harvard University April 2nd, 2011

    “The Fingerprints of Sea Level Change”
    http://www.youtube.com/watch?v=RhdY-ZezK7w

    You might find it interesting.

    • RickA says:

      Is sea level rise all society has to worry about?

      Of course not.

      It is however, one of the few certain bad consequences of a warming world.

      Many of the other predicted effects are ill-defined (at least to me).

      Changes in precipitation patterns. Dry places get dryer and wet places get wetter. But overall, more moisture in the air, so more total precipitation in a warming world. What does that mean for me – I have no clue. However, if you live close to sea level, it is very easy to look out 100 years, and imagine what an extra 39 cm might mean (or an extra 60 cm).

      • Nullius in Verba says:

        Is it so certain that it will be a bad consequence?

        Coastlines shift back and forth, land rises and subsides, all the time. There are normal background geological processes that result in so many of us living so close to sea level – mountains erode and the rivers deposit flat land when the water reaches sea level and slows down, while at the same time being the reason it is so fertile. It is no coincidence that so much of it matches sea level so precisely, though the sea has only been where it is for a few thousand years, nor that we live there. And if you’ve ever seen an archaological dig, you know that in many places the land surface rises. The question is, why should we suppose these longstanding geological processes will now fail?

        It’s entirely possible to imagine that 39 cm, or even 60 cm, might pass without notice. And in any case we are long used to dealing with shifts in coastline, and can even engineer it to a large extent.

        • RickA says:

          Well, to be honest – not for me. I live in Minnesota. Global warming will probably be a net plus for my region. Warmer temperatures, and supposedly more precipitation. Should be good for farmers.

          It might not be so great if you live in the Florida keys.

          • Nullius in Verba says:

            People who live in the Florida Keys are pretty relaxed about such things. They would be aware that mangrove swamps trap sediment, building up mud banks and raising land levels. There are some locations with erosion problems, but these are primarily due to wind-driven waves and the rapid ocean currents and are long-standing features.

            That’s what I mean. Everyone looks at the Keys and sees the altitude above sea level, nobody notices the mangroves and how they are affecting it. Coastal geology is dynamic.

  13. N.D.Babaji says:

    Collecting measurements across multiple geographic regions for a general picture increases the likelihood of an accurate assessment that the anthropogenic injection of climate forcing effects is the root cause of ice volume loss. That is, net losses in sea ice volume rather than cover estimates are more to the point. Loss in sea ice volume leads to earlier ice cover loss in the summer season (arctic). Incoming solar radiation impinges on dark sea surface over an increased period, increasing energy intake to the Arctic sea (esp. area surrounding Greenland, generally considered part of the Arctic as well, though unfortunately not sea ice).
    Increased net sea temp increase sea volume, increase percent of measured Greenland sheet in thaw state (in excess of 90% for 2012) represents increased loss of Greenland ice sheet volume, directly adding to sea volume.
    Costs include population displacement, arable land loss, and increased storm damage to population centers. Costs of these effects are appearing than projected earlier predicted (projections are being moved up rather than out). Damping the future costs will require economic adjustments that should not be expected to provide return on investment within a 10 year horizon.

    The injection of anthropogenic positive energy absorption factors as the forcing input to climate change are only being disputed by less than 2-sigma of researchers degreed in climatology related fields (less than 3-sigma when compensated outliers are identified and excluded). To that end the identification of anthropogenic forcing inputs as the driver of the trends may be set aside to pursue more effective investment of debate, such as the ranking of best cost-benefit return to dampen the injection factors.

    HA!

  14. Leonard Weinstein says:

    Zeke,

    Carbon particles, mainly from China’s coal plants, have been implicated as a significant cause of a portion of Arctic melting. In addition, studies show that more exposed ocean at very high latitudes result in more clouds forming (from evaporation/condensation), with the blocking clouds reducing net ocean solar absorption. The first of these two effects is likely a large fraction of the cause for more melting, but the second results in not much change in Earth’s net albedo, despite some high latitude warming. The fact of large scale melting at the Arctic also appears to be a long term cyclical effect (e.g., the edge of Greenland was GREEN during the Viking occupation, and seas relatively open). Since the global average temperature has not increased significantly over the last 15 years, why do you make a big issue of Arctic ice melt?