Sea Ice 3-D

Many of us have seen graphs of the dramatic decline of Arctic sea ice over the last 30+ years (since we’ve been monitoring it via satellites), a decline which is unmatched for at least a century and almost certainly for several thousand years. The graph of sea ice extent anomaly (monthly average data from the National Snow and Ice Data Center) shows the change plainly:

Both the extent and area of sea ice coverage in the arctic has declined at an alarming rate — the phrase “death spiral” comes to mind.

There’s far less data about the thickness or the arctic ice cap, but there is a growing body of evidence from satellites, moorings, aircraft, and from submarine measurements of ice draft. It reveals that not only is Arctic ice covering less of the sea, its thickness is also declining.

The good people at the Polar Science Center combine all the available data to form an estimate of Arctic sea ice volume. They have recently made their data available for download; let’s take a look.

Here’s their raw data, estimates of Arctic sea ice volume in thousands of cubic kilometers:

As expected, there’s an obvious seasonal cycle with more ice in winter and less in summer. In addition there’s a downward trend which is rather obvious even with the seasonal changes intact. The annual cycle shows the greatest volume in April, the least in September:

I’ve highlighted both 2010 and 2011 in red, with connecting lines. A clear sign of the continued loss of Arctic ice is that both 2010 and 2011 showed the lowest values on record, for every month of the year. The phrase “death spiral” comes to mind.

We can remove the seasonal cycle by computing anomaly, the difference between a given value and the average for that same time of year. For that purpose, I translated daily values of sea ice volume into monthly averages, then computed anomalies, giving this:

Except for a slight difference in the zero point (they use the 1979-2010 average as a baseline to define anomaly, wherease I’ve used the entire time span), this is essentially the same as the anomaly graph given on the PIOMAS website:

As most of you know, the annual minimum in sea ice extent occurs in September, as does the minimum of sea ice volume. The September extent is in sharp decline and the decrease has actually accelerated:

The September volume is also in sharp decline, one which has also accelerated, and shows its lowest value just last year:

We can divide the volume of sea ice by the area to get an estimate of the average thickness of the sea ice. This too shows both a decline and an annual cycle, with ice thinnest in December and thickest in July:

The reason it’s thinnest in early winter is that new ice has formed but it’s mostly thin, first-year ice, and the dead of winter hasn’t yet thickened it. Even so, the December thickness has declined steadily:

The thickness at its annual maximum in July has also declined:

Meanwhile, thickness during the September volume/extent minimum has declined sharply with the last two years showing a dramatic drop:

The decline in Arctic sea ice extent and area is truly alarming, and the decline in sea ice volume is even more so. The last two years have plumbed new depths in ice cover; the outlook for the polar ice cap is grim indeed.

The phrase “death spiral” comes to mind.


73 responses to “Sea Ice 3-D

  1. Don’t worry Tamino, its only the wind…which obviously has a hidden multi-decadal oscillation waiting to be discovered by some mathturbator.

  2. Perhaps that phrase will soon be coming to a few smaller minds, too.

    Sure hope so–we need the wakeup call. But the consequences surely will be, er, non-trivial.

  3. So wind contributes to sea ice decline, and Watts apparently (I haven’t bothered to read, why bother?) thinks this is somehow evidence against climate science.

    Of course, if sea ice thickness were at (say) 1980 levels, the wind would be blowing between a rock and a hard place.

    Like … duh.

  4. Anthony frames the article such that it is a ‘revelation’ that winds and ice transport fluctuations influence sea ice extent, leading uncritical commenters to assume that WUWT has got some kind of scoop on this.

  5. As a matter of fact, observing a trend and even an acceleration does not prove anything about the cause of variation. It could be the wind, or oceanic oscillations – this even includes the fact that the winds and oceanic oscillations could also be driven by AGW ! if you plot the hourly temperature over the last six months of a year, you will get a graph very similar to that of sea ice over thirty years – without any anthropogenic component or possible extrapolation. Just observing that something is increasing , and something else is decreasing, isn’t a strong piece of evidence – this is much too easy to do cherry picking among a vast amount of possible facts, retaining only those going in the “right” direction. Only a predictive theory has a real scientific content. A posteriori facts are unreliable.

    Among the open questions that a predictive theory should address :
    – if sea ice is so sensitive to GW, why doesn’t Antarctica show the same trend ?
    – why did arctic ice melting accelerate in the 2000’s, when temperatures didn’t particularly grow, and certainly didn’t accelerate ? this would be rather an argument to say that acceleration is due to something else than average global temperatures !
    – why is the september minimum so poorly correlated with other seasons ?

    [Response: Or Leprechauns. You forgot to mention that it might be because of the Atlantic Multidecadal Leprechauns.

    You aren’t a skeptic. You’re in denial, so you’re willing to scrape the bottom of the barrel for any excuse to delude yourself.]

    • Kevin Stanley

      Who are you arguing with, okatiniko? Who has said that the immediate driver of the change in arctic ice is “average global temperatures?” With zero lag, apparently?!?

      Think about that for a second.

      For one thing, references to “average global temperatures” tend to refer to *air* temperature. The thing about sea ice is, the top is bounded by air, and the bottom is bounded by *water*. Do you think the water temperature might be relevant? Do you think people who have dedicated their professional lives to studying climate have somehow missed this?

      For another, how could local conditions respond immediately to a global average? Local temperature-related phenomena respond to…local temperature. Are you seriously implying that someone here thinks the opposite?

      Do you really think that you are arguing with people who think that a zero-lag response to air temperature is the *only* factor in the ice melt? No consideration of dark ledes and surface melt ponds decreasing the albedo and absorbing more sunlight?

      Do you think that the question about antarctic sea ice hasn’t been answered? Do you think climate scientists are just sweeping that one under the rug?

      Can you really not think of any obvious physical reasons that the September minimum can vary more than the other seasons? If I give you some hints, do you think it will help? The arctic ocean is roughly a big circle bounded by land. Can you think of any constraints that might place on maximum extent? The average annual temperature is something like -14C. Can you see any reason why a warming trend might not immediately cause big changes in rate of freezing in any but the warmest times of year?

      You write like someone with a lot of faith in his own thinking. Yet you will say something like that bit about comparing a plot of hourly temperature over the last six months of a year with arctic ice over the last 30, without any apparent awareness of having *at best* scored an own goal. Can you see why that example not only fails to prove your point, but fits right in with the case being made by the “other side?”

    • Timothy (likes zebras)

      Your questions seem fairly easy to give reasonable answers to:
      – Antarctica is different, in the sense that sea-ice there surrounds a continent, whereas in the Arctic sea-ice is surrounded by continents. It’s not surprising that the sea-ice response is different, and Antarctic sea-ice extent is already very low at its minimum. There are also good reasons for concluding that the ozone hole has changed the wind circulation around Antarctica in such a way as to have an effect.
      – The acceleration in ice-melt would appear to me to be an obvious example of the ice-albedo feedback in operation.
      – Is it? All seasons of Arctic ice exhibit a strong downward trend.

    • oka,

      You’ve been patient guided to documentation, papers, discussions… now is the time to stand on your own legs on why the Antarctic is responding differently. I’ll give you as leads, the Antarctic Vortex narrowing, Ozone hole, increased water vapor, Land versus Ocean perifery,

      The Irish Leprechaun has a multiheaded cousin in Norway btw.

      Take care.

    • “patiently” was meant in my previous post.

      Meantime, one of my fav items tracked, global sea ice area , continued on a 365 days rolling record low, June month of year too, with only 21 days of data.

    • [Response: Or Leprechauns. You forgot to mention that it might be because of the Atlantic Multidecadal Leprechauns.

      You aren’t a skeptic. You’re in denial, so you’re willing to scrape the bottom of the barrel for any excuse to delude yourself.]

      God, you act like an ass. To respond that way to someone making a good effort at discussion shows you are only interested in furthering propaganda, not discussing science.

      [Response: My purpose is discussing science. Okatiniko’s is not. Neither is yours.

      Clearly you’re so embarrassed by having blundered in such amateurish fashion that you’re clutching at any criticism you can find. I almost feel sorry for you.]

      • Kevin Stanley

        I started reading this blog fairly frequently a few years ago. I would say that Tamino’s tone has grown more severe with certain types of commenters over time…for very clear and understandable reasons. How many times can you respond thoughtfully, taking the time to explain in detail, to the same transparent B*llsh*t? You see the same things over and over…the standard list of denialist tropes, all discredited, but still making the rounds, plus a helping of people who have made up their minds without understanding the current state of knowledge about climate AT ALL. Sometimes both at the same time.

        So I suspect Tamino’s logic goes something like this: The first time someone says something way off base, it makes sense to point them towards good resources for educating them selves, so that if they’re actually curious about reality, they can learn enough to evolve past the silly BS. But after a series of such comments with no sign that the person has made any effort to learn a little before they sound off…they’re either willfully ignorant or dishonest. Not much point in wasting a lot of time with them then.

        Personally I like the approach of continuing to show the errors for the lurkers’ sake. But I can more than understand Tamino’s frustration.

        So Will_Power, you’re a math guy, right? Would it be safe to assume you have some understanding of logic? Both being systems of rules and all….

        Well, if so, can you explain how it follows that if someone is rude (in your opinion), that person is a propagandist? That looks like a giant non-sequitur to me.

      • @ Kevin Stanley

        I read every single comment over at Skeptical Science, so I can well sympathize with Tamino. It takes very little time indeed (usually just one comment) to establish the technical background and ideology of the person posting the comment. After one reads several comments by that poster, one can even often identify them when posting under multiple screen names.

        Tamino displays a very appropriate, learned, level of response to those posting comments. Given the provocation of some commentators, his response is measured and even,

        All the more reason to not have a blog of my own. :)

      • TrueSceptic

        Where’s the propaganda? Do you really think that okatiniko was making a good effort at discussion, or furthering propaganda him/herself?

      • @ TrueSceptic

        Dunno if you were referring to me with that one but here goes: okatiniko is one of many who post repeatedly at SkS with variants of “It’s Not Happening”, “It’s Not Us”, “It’s Not Bad” and “It’s Too Hard”. They have an omnipresent narrative and agenda that they work hard to consistently deliver with the goal of nothing more than outright dissembling to confuse and confabulate the background readership which rarely, if ever, posts a comment. These dissemblers seldom can connect the dots themselves, even when it may take all of 0.37 seconds for Google to look up the answer. They offer up an endless stream of unsupported assertions and innuendo which, if called to support with citations and links, they will proceed to ignore. Once in a while you can catch them recycling material (disinformation they previously spread in earlier comments on other posts), so it becomes patently clear who’s who and what’s what.

        Apologies, Tamino, for the OT rant, but when I see the same activities by the same suspects here as at SkS or RC it makes my blood boil at bit.

        [Response: No apology necessary. In fact — thanks.]

      • To respond that way to someone making a good effort at *pretending* to discussion shows quite clearly that Tamino will simply no longer put up with the dimwitted willfully ignorant b*llsh*t that goes on at WTF’sUWT.

      • To respond that way to someone making a good effort at *pretending* to discussion shows quite clearly that Tamino will simply no longer put up with the dimwitted willfully ignorant b*llsh*t that goes on at WTF’sUWT.

        ‘Zackly. At this grisly stage of the climate wars, there is no point in putting up with these pricks any longer. The time when denier trolls might be given the benefit of the doubt is long past.

        I hope Tamino’s fuse will grow even shorter.

      • Daniel,

        Sorry, I should have been clearer. This is a problem with replies only nesting so far. I was replying to Will_Power.

        I’m quite familiar with denier/fantasist* tactics. I agree with everything you say and I understand perfectly when Tamino has little patience with them.

        *They are not sceptics in any true sense, of course.

      • Commenters who come here and spout WTF’sUWT nonsense deserve whatever tamino dishes out.

    • I agree with the others. When okatiniko writes: “Only a predictive theory has a real scientific content. A posteriori facts are unreliable,” I have two reactions. The first is to respond: “Yup.” The second is to ask: “And have you read the relevant theory and research?” From the comment about ice declining while temps didn’t seem to, post 2000, I can only imagine okatiniko has not read the relevant literature. As Eli might suggest, gently: RTFR. I hope okatiniko has been helped by the answers generously provided here. But like the others, I can see why Tamino might be a tad impatient with such questions when it seems the person either hasn’t done the reading that such questions obviously require or is possibly only trying to score points.

  6. Timothy (likes zebras)

    So if we ignore the trend for a moment, and look at the details of recent years, the ice volume data suggests that 2007 was a particularly low year for extent because the winds pushed the ice together to bunch up, but that 2010 only avoided setting a new extent record because the winds did the opposite.

    It would be interesting to see how the PIOMAS numbers compare to the new Cryosat data.

    The decline in September ice volume is remarkable. One is used to seeing noisy data, but in this case I think it is 11 of the last 14 years showing a decline in volume on the previous year, and 8 of the last 12 years set a new record low ice volume.

    Assuming there is a 6th assessment report form the IPCC, and assuming that the PIOMAS data is reasonably accurate, it would seem that there will be some explaining to do then to account for the failure of the climate models to predict such a rapid loss of sea-ice.

    • Andrew Dodds

      I think it’s just a problem of being conservative..

      Despite what some on the denialist blogosphere seem to think, the IPCC has been extremely conservative regarding cryosphere impacts generally – witness the ‘unqualtified’ impact of ice sheet melting on sea level, and the case you just highlighted.

    • IIRC, it wasn’t just ‘bunching’–it was a year of very high ice export. The wind presumably had a role there.

  7. People might be interested that the first Cryosat data set was released at the Paris Air Show:

  8. Should we read anything into the apparent huge swings at the end of the first graph?

    [Response: They’re because the annual cycle itself has changed. Computing anomaly doesn’t remove the annual cycle it removes the *average* annual cycle, so it leaves in place the difference between the present annual cycle and its long-term average.]

    • That’s interesting. It never occurred to me (duh!) that those swings were an artifact. What’s the best way to deal with this? After all, those swings give a misleading impression.

      [Response: One option is to use an “adaptive” annual cycle (one which changes over time). Another is simply to smooth the anomalies on a short timescale (to remove the wild fluctuations). The disadvantage is that either way one loses some information. After all, the *trend* in March is different from that for September, and that difference gets erased.

      As for the “best” way — I suppose that’s a subjective question.]

  9. Extrapolating this decline in Arctic ice volume leads to an ice-free Arctic in summer starting before 2020.

    My understanding is that there is some debate about whether the ice will just smoothly continue to decline, with open water covering a longer period each year, or whether there will be some kind of ‘flip’ or change of state whereby, after a certain point, the ice will not re-form – in other words, can a seasonally ice-free Arctic be stable or is it a transient phenomenon?

    If extrapolating the volume decline is valid then it looks like we might have the answer to that question in 10 years, or not much longer.

    • Timothy (likes zebras)

      Sea-ice does not have any difficulty in forming in places such as the Baltic sea, which for most of the year is ice free. I’m certain that for a long time after sea-ice disappears completely in September there will be considerable sea-ice forming during the Arctic winter. Indeed, this is likely to create a new challenge for forecasters with much interest from shipping companies.

    • Andrew Dodds


      The surface layer of the Arctic ocean in summer has a very low salinity (relatively speaking) since it contains mostly molten ice, i.e. fresh water. Hence it easily re-freezes. However, the longer it spends unfrozen, the greater the mixing with salt water and hence, presumably, it is harder to re-freeze.

      There should also be an albedo effect – once the ice melts, the darker ocean water absorbs more heat. This may go the other way in winter, though, with more heat escaping in the dark. This is climate, nothing is that straightforward!

      Mind you, at the rate we are going. it would only take a moderate outlier to make any year ice-free.

      • If I have this right–and am still up-to-date–modeling studies show a seasonally ice-free Arctic Ocean to be unstable. Once we lose summer ice (and absent a significant change in the forcings), we’re pretty much committed to a year-round ice-free Arctic Basin–though there’s a lag time getting there (couple of decades, maybe?–don’t remember and haven’t time right now to search.)

        The Arctic Ocean is landlocked in a sense, but not to a similar degree to the Baltic. Large amounts of heat are advected by oceanic currents into the Basin now; that won’t stop when the ice is gone. So the energy will go proportionately more into heating seawater and less into powering phase changes–freezing will be delayed more and more, winter ice volumes will decline, and spring melts will happen sooner. Of course the albedo feedback only applies in the sunny half of the year, but the water vapor feedback may be a bigger deal anyway.

        I know the prospect of a permanently open Arctic Ocean is a shocking thing to contemplate, and I’m not clear on what the expected consequences are–but clearly they will be far-reaching.

        I’ve got to look up more about those modeling studies. . .

        [Response: I believe a recent study of exactly this issue (I don’t recall the ref) concluded otherwise. If summer ice disappears suddenly then winter ice rebounds quickly — in a few years. But eventually, ice disappears just because it gets too hot.]

      • Yes, Tamino, you are correct. The study is Zhang, Steele & Schweiger, 2010. In their model, winter sea ice growth was enhanced by thinner ice cover, and by decreased ice export (that latter doesn’t make sense to me, but that’s what their results showed.)

        I’ve misplaced the link, but it should be easy enough to Google up.

        It appears I was thinking of Gareth Renowden’s “back of the back of the envelope” calculations, blogged here:

        It’s provocative, and could conceivably have some validity, but peer-reviewed science it ain’t, as Gareth clearly states.

      • Oh, I forgot to add that I also found a study calculating the global result of a seasonally ice-free Arctic Ocean. The full result was paywalled, but the author concluded that the “global-annual mean” forcing would not be more than 0.7 W/m2.

  10. Jonathan Bagley

    I’m interested in hearing your answers to the questions posed by okatiniko.

    • Kevin Stanley

      Well, some of those questions don’t make much sense and imply a deep misunderstanding of the current state of knowledge about climate. Please see my response to okatiniko above.

  11. Reading this, the phrase “death spiral” came to mind…

  12. Igor Samoylenko


    This is getting seriously boring. It is obvious from your posts in the “Chaos” thread and here that you haven’t bothered to look into any of the background to your questions on climate science. Yet you continue to fire them away, based on your own, poorly-informed impression of the issues involved, expecting others to take them seriously and provide “scientifically-substantiated” answers. Why don’t you start with yourself?

    If you want comprehensive, scientifically-substantiated evidence for AGW, start with the IPCC Fourth Assessment report. It is all there, nicely summarised for you – the state of climate science as of 2007. When you are done, come back with questions if you still have any. There are many people here who’d be willing to help with any genuine questions on climate science. Until then, I don’t see why anyone should take you and your questions seriously.

    I have a request for you. Please provide a scientifically-substantiated proof that your questions are not caused by your significant (wilful?) ignorance of the climate science and (are not the result of a random walk).

  13. People wondering about accelerated ice loss without rapid temperature increase may as well ask themselves why they can observe an accelerated ice loss, but no significant temperature increase, in a glass of water with ice cubes placed in the sun. Until the ice cubes have totally melted, that is.

    One has to look at the energy flows, and they can be formidable. Melting one square km ice 1 m thick takes ca 80×10^6 kWh, five million km^2 same thickness takes 4×10^14 kWh = 400 000 TWh. Compare that to domestic energy consumption, for example.

    One small technical point: When the annual cycle changes, a time series has to reflect that to really reflect the changes. We have already seen enough cherry-picking utilizing that phenomenon, with the somewhat delayed onset of spring melting taken as proof of “ice recovery”.

  14. o: if sea ice is so sensitive to GW, why doesn’t Antarctica show the same trend ?

    BPL: Because A) Antarctica isn’t warming as fast as the Arctic, something predicted decades ago, and B) it has massive calving of glaciers coming off the land which replaces the melting sea ice at the edges.

  15. Unfortunately we cant see each others responses… lag between posting and releasing, sure as heck, okadoka has half a dozen giving Antarctic slower decline causes [on the surface, from a pretty picture made from space], duplicate or augmenting. Now if half a dozen can whip out the varying reasons, all valid, what does this make this look like? Time to push the ”ignore” button till it’s seated.


  16. Too quick to enter: [on the surface, from a pretty picture made from space little is seen]

  17. Steve Bloom

    Tamino, you mention “a decline which is unmatched for at least a century and almost certainly for several thousand years,” referring to sea ice quantity, but while it is not possible to know the pace of sea ice decline in the distant past, it’s fair to say that it’s very probably entirely unprecedented. IMHO this is a distinction worth making, since after all the Arctic has been ice-free in the summer for most of its existence.

  18. “Only a predictive theory has a real scientific content.”

    Like this one?–

    “. . . human beings are now carrying out a large scale geophysical experiment of a kind that could not have happened in the past nor be reproduced in the future. Within a few centuries we are returning to the atmosphere and oceans the concentrated organic carbon stored in sedimentary rocks over hundreds of millions of years. This experiment, if adequately documented, may yield a far-reaching insight into the processes determining weather and climate.”

    –Roger Revelle, 1957

    And the result?

    Right, .12 C warming per decade since 1957.

  19. Steve Bloom: …, since after all the Arctic has been ice-free in the summer for most of its existence.

    Arctic was ice free not caused by humans in the past, there has been glaciers miles high, also not caused by humans. Now we have shrinking sea ice caused by humans, time to act, or not?

  20. Armour et al 2011 (in GRL) says:

    “We test sea ice reversibility within a state-of-the-art atmosphere-ocean global climate model by increasing atmospheric carbon dioxide until the Arctic Ocean becomes ice-free throughout the year and subsequently decreasing it until the initial ice cover returns. Evidence for irreversibility in the form of hysteresis outside the envelope of natural variability is explored for the loss of summer and winter ice in both hemispheres. We find no evidence of irreversibility or multiple ice-cover states over the full range of simulated sea ice conditions between the modern climate and that with an annually ice-free Arctic Ocean.”

    This is interesting, but is it rather academic? We don’t yet know what level of atmospheric CO2 results in an annually ice-free Arctic, but the studies I’ve seen suggest that global temperature isn’t going to decline on its own for a very long time anyway, even if we were to stop all emissions and allow nature to bring down CO2 level, which in itself is very unlikely for many decades.

    What I’m thinking is that we could reach the level of CO2 that results in an annually ice-free Arctic before we even know it – perhaps we already have? – and then the fact that Arctic ice would readily recover if global temperature fell again becomes academic, because that won’t actually happen for a very long time. We could say that even if a seasonally ice-free Arctic is theoretically stable, we won’t see it because we’ll have pushed straight past it to an annually ice-free state that will last for the foreseeable future… or perhaps forever, depending on whether we’ve put enough carbon back into the climate system to return the Earth to an ice-free state (which is another very pertinent issue – exactly how much fossil carbon do we have to return to the climate system in order to end up with an ice-free planet, given that the natural rate of sequestration into the Earth’s crust is something like 20,000 times slower than the rate at which we’re currently releasing it?).

    • FWIW, this seems like a very well-taken point to me.

    • Perhaps the most interesting point in my musings on the rate of transition to a permanently unfrozen Arctic Ocean, kindly linked to by Kevin above, is that if I am right, and the current heat budget imbalance continues, then the earliest dates fall within the climate commitment — the 30 or so years of warming “in the pipeline” from current GHG levels. Even if we assume drastic emissions reductions in the near future, we will push well beyond the atmospheric GHG levels within which the Arctic can support ice at any time of year. And that does not bode well for the methane hydrate deposits in the shallow seas off Siberia.

  21. When the summer sea ice finally does disappear, perhaps the worst effect of that – or at least the most annoying – will be when it starts to freeze over again the following winter and we have to listen to the denychopaths telling us that ice is growing at an infinite rate.

  22. “Assuming there is a 6th assessment report form the IPCC, and assuming that the PIOMAS data is reasonably accurate, it would seem that there will be some explaining to do then to account for the failure of the climate models to predict such a rapid loss of sea-ice.”

    It’s not actually the business of climate models to predict ice-behaviour. That’s the business of ice-behaviour models, which is where the failure lies. No blame attaches, since the problem is inherently more complex than climate modelling.

    • Timothy (likes zebras)

      The climate models produce predictions of sea-ice. Climate models incorporate sea-ice models, so if I understand you correctly you are positing that the [assumed] failure to correctly predict the pace of Arctic sea-ice decline is a failure of sea-ice dynamics, rather than a failure to model the effect of GHGs on the climate more generally.

      This is plausibly true, but I’m not sure how you would prove it. Do people run sea-ice models with observed ocean and atmosphere forcing? Would it even make sense?

      I don’t think you can rule out the potential for errors in the ocean/atmospheric circulation in the models from being important.

  23. [edit]

    [Response: It’s not just ironic that you carefully avoid mentioning the effect on Arctic sea ice of temperature change — both air and water — in the ARCTIC.

    The merest glance at Arctic temperature — even looking only at the last decade — reveals the cause of the continuing sea ice death spiral. The merest glance at Antarctic temperature reveals the main reason for the difference in changes of northern and southern hemisphere sea ice. Yet you have used their differences, not to enlighten about interesting regional patterns, but to cast doubt on global warming. You also mischaracterize global temperature with your false claim of a 21st-century plateau.

    Ignorance is reversible. But I don’t believe you’re really ignorant of these facts. I believe you want to *appear* to ask “questions” but it’s just a thinly veiled facade. Even though you know that Arctic temperature has risen much faster than global and continued to do so over the last decade, you’ll still mention only global temperature and lie about its trend to boot. And when any of your errors are revealed (like your faulty, made-up definition of climate sensitivity of malignant design), you dare not admit any mistake, you just move on to the next talking point.

    When I first suggested that you might be more interested in arguing than in learning — I had you pegged. Doubt is your product, it has been your only goal all along.

    So perhaps you’ll understand why I no longer care to publish your comments.]

  24. Tamino,

    Hear hear on throwing Okatino off the bus. I thought he was serious at first, but it became apparent pretty fast that his goal was not to discuss, but to waste time.

    I don’t know how many of these guys are paid to do this, but apparently some really are. I imagine, though, that the majority are self-styled activists some rightist opinion leader called on to throw a wrench in the works. No doubt by doing so they think they’re saving America from a plot to raise taxes, create New World Order socialism, and send black helicopters to take all the white Christian patriots to UN detention camps.

  25. Your third chart from the top erroneously labels volume with km^2 rather than km^3.

    [Response: My mistake.]

  26. @BPL: I don’t know how many of these guys are paid to do this, but apparently some really are.

    Was there not a zombie commenter recruitment drive by one of the denialist think tanks? I think I remember such a thing, but can’t find it. Anyone have a link?

  27. Here is a presentation of Arctic sea ice trends plotted every which way but loose. Blind Freddy can see what is going on here. And yet, once again, here is a denier denying. I am amazed not at Tamino’s being a bit abrupt but that he retains any politeness at all towards these morons.

    Some rudeness towards them here our patience is not inexhaustible.

  28. @Torsten

    No, I’m remembering an actual offer of money for denier spear carriers to spam the comments below any climate article at web news sites.

  29. If I can just interrupt the discussion of denialists (yawn)… ;)

    Does anyone else think the data from Cryosat looks unbelievably thick?

    Here’s the first image from Cryosat pertaining to the thickness of the Arctic sea ice:
    You can download a hi-res image from there. It shows that most of the Arctic sea ice in Jan/Feb is around 3 metres thick with peaks of 5 metres off the Canadian Arctic Archipelago (CAA).

    Another page on the ESA site makes reference to the Alfred Wegner Institute (AWI) validation flights being used to validate the Cryosat mission. However according to the AWI press release the flights in May of three successive years yielded – 2009 average thickness 1.7m, 2010 1.6m, 2011 1.4m. The flight path going across the Arctic Ocean north of the CAA (as well as within the CAA itself).

    The AWI flights are just one set of flights who’s average may be skewed by conditions in the CAA. However whilst most of the NIS Egg Charts are of little use as Jan/Feb conditions yield >2 metre thick categories, those areas of thinner sea-ice are still shown as 2 to 3.5 metres by Cryosat.

    Kwok & Rothrock 2009, “Decline in Arctic sea ice thickness from submarine and ICESat records: 1958–2008”:

    Click to access Kwok.2009.GRL.pdf

    Notably figure 2 c & f.

    Click to access Zwally.pdf

    Notably pages 10 and 12.

    These references based on the ICESat laser altimeter suggest thinner sea-ice than the Cryosat first data shows.

    Finally as a very rough ball-park figure:
    Arctic sea ice volume in April (PIOMAS) ~21k km^3. Area of Arctic Ocean (Wikipedia) ~14M km^2. Which gives an average thickness of (very roughly) 1.5 metres, half what Cryosat shows after the ice growth since Jan/Feb.

    Any ideas anyone?

  30. Michael Sweet

    I read a number of the responses at WUWT at the link at the top of the thread. I do not read a lot there, but it seemed to me that a large fraction (over half) of what I read were concerned about the ice decline and attributed it to humans. Other threads I have read there are usually solidly deniers. Perhaps even the troops at WUWT realize that the ice is telling us a story that cannot be denied. If that is true we can hope that the story will be heard sooner rather than later. The Arctic ice is an easy story to tell: it is obvious without Tamino’s statistics that the ice is in a death spiral, and everyone knows that heat is what melts ice. I think we should emphasize the 5,000 year old Ellesmere ice shelves that have melted. They tell an obvious, undeniable story.

    • Michael Sweet wrote:

      The Arctic ice is an easy story to tell: it is obvious without Tamino’s statistics that the ice is in a death spiral, and everyone knows that heat is what melts ice. I think we should emphasize the 5,000 year old Ellesmere ice shelves that have melted. They tell an obvious, undeniable story.

      Well, while it isn’t quite as obvious, along the same lines it might be worth reminding people when they are in a more reality-based mood that we have had some sea ice in the Arctic for at least the past 13 million years.

      Please see:

      ABSTRACT: Arctic sea-ice extent and volume are declining rapidly. Several studies project that the Arctic Ocean may become seasonally ice-free by the year 2040 or even earlier…. Although existing records are far from complete, they indicate that sea ice became a feature of the Arctic by 47 Ma, following a pronounced decline in atmospheric pCO2 after the Paleocene–Eocene Thermal Optimum, and consistently covered at least part of the Arctic Ocean for no less than the last 13–14 million years. Ice was apparently most widespread during the last 2–3 million years, in accordance with Earth’s overall cooler climate.

      L. Polyak, R. Alley, J. Andrews et al (2010) History of sea ice in the Arctic, Quaternary Science Reviews, Vol. 29, Issue 15-16, pp. 1757-1778

      • Andrew Dodds

        Trying to find a decent source, but I’m wondering – what was the state of the Bering Strait at 10ma? Certainly the gap between Greenland and Scandinavia would have been smaller and shallower, so at a guess I’d think that the Arctic would have been significantly less well-connected to the rest of the ocean at that time (may have been a lake at ~20ma). Hence less ice flushing; I’d assume that rivers would also lower the surface salinity.

        It may be the case here that if we go back to Pliocene temperatures, but with modern geography, we lose the sea ice.

    • Andrew Dodds | June 27, 2011 at 1:40 pm wonder about the Bering Strait 10ma. Well somehow the mammoths managed to get across, and that species, origened in the Americas got extinct some 7k years ago, by human hand it’s thought and it’s rather hot coat.

      As for Greenland, one hypothesis is that the Rockies forming a few million years ago are the reason that it’s frozen over, so with that ice-machine NOT in place, venture to think that the Arctic, before another 100 some Milankovic cycles to add since was rather different then.

      • Robert Murphy

        “one hypothesis is that the Rockies forming a few million years ago…”

        They formed tens of millions of years ago.

      • I stand corrected… don’t know why that stuck in me mind:

        … maybe this bit

        In the last 1.8 million years, the ice ages left their mark on the Rockies, forming extensive glacial landforms, such as U-shaped valleys and cirques.

      • Andrew Dodds


        To clarify, 10ma= 10 million years. I was thinking that if/when the Arctic was a closed or nearly-closed basin, presumably with a more freshwater nature, ‘sea/lake’ ice would have more readily formed, even with higher global temperatures, and there would be little if any flushing-out of multi year ice.

        This is speculation, obviously, but it may well be that Arctic ice is less ‘stable’ with regard to global temperatures than at any time in the last 20 million years, due to progressively greater heat transport through both the Bering Straight (the Aleutians are gradually moving south through slab pullback, IIRC) and the Norwegian Sea. So stable sea ice at 10 million years ago (for a given global temperature) may not equate to the same now.

  31. Although less telling than the review I mentioned that synthesizes data from over a hundred different studies, it is also worth noting that while there have been times in past few million years that have seen less ice, there has been sufficient ice and cold temperatures in the Arctic to keep North Atlantic and North Pacific populations of right whale separate, to the point that the the symbiotic “whale lice” crustacea that inhabit them diverged 6 million years ago, becoming the distinct species that they are today.

    Please see:

    North Atlantic and Southern Ocean populations of all three species are reciprocally monophyletic, and North Pacific C. erraticus is well separated from North Atlantic and southern C. erraticus. Mitochondrial clock calibrations suggest that these divergences occurred around 6 million years ago (Ma), and that the Eubalaena mitochondrial clock is very slow. North Pacific C. ovalis forms a clade inside the southern C. ovalis gene tree, implying that at least one right whale has crossed the equator in the Pacific Ocean within the last 1–2 million years (Myr).

    Kaliszewska ZA, et al (2005) Population histories of right whales (Cetacea: Eubalaena) inferred from mitochondrial sequence diversities and divergences of their whale lice (Amphipoda: Cyamus). Mol Ecol.; 14(11): 3439-56.

  32. I may have read incorrectly, but to me the part about ice volume reads as if PIOMAS is composed of actual measurements.

    [Response: You don’t read incorrectly, you see the situation through the filter of your own bias.]

    “There’s far less data about the thickness or the arctic ice cap, but there is a growing body of evidence from satellites, moorings, aircraft, and from submarine measurements of ice draft. It reveals that not only is Arctic ice covering less of the sea, its thickness is also declining.

    The good people at the Polar Science Center combine all the available data to form an estimate of Arctic sea ice volume. They have recently made their data available for download; let’s take a look.”

    To me these two paragraphs imply that PIOMAS is composed of actual measurements of thickness.

    But as I understand it, PIOMAS is only a model driven by temperature, wind, SSTs etc. … not actual measurements of thickness.

    [Response: The model is *built* using actual thickness measurements. Some — not me to be sure — might interpret your statement as implying that the PIOMAS people never laid eyes on an actual thickness measurements in their lives.]

    True, PIOMAS has some decent verification vs actual thickness measurements. But as far as I am aware, there is no good data set of thickness measurements over the last 3 decades. There’s IceSat data for the mid 2000s and now we have Cryosat.. but before that it’s pretty sparse. Just enough to conduct some validation.

    [Response: You’re just making stuff up. I myself analyzed over 2,000 data points of ice draft measurements from submarines over the time period 1975 to 2001, covering a vast area of the Arctic. And that’s just the “tip of the iceberg,” so to speak, since there’s also data from satellites, moorings, aircraft, and yet more submarine data. When you offhandedly say “There’s IceSat data for the mid 2000s and now we have Cryosat.. but before that it’s pretty sparse” you’re just conjuring up a fantasy designed to discredit PIOMAS.]

    It would also seem worthwhile that PIOMAS just revised the model in a major way so as to reduce the trend in volume decline by 30% or so. If they are still making such large revisions, I highly doubt the model can be said to be particularly accurate at this point.

    [Response: Upon revision the model trend estimate changed from -3.6 km^3/decade to -2.8 km^3/decade. That’s a 22% reduction, but you call it 30% — an exaggeration in order to discredit PIOMAS. And by the way, I estimate the 95% CI estimate using the latest data as from -2.1 to -3.6 km^3/decade so the error range from the new model includes the estimate from the old model. I highly doubt that your claims or opinion can be said to be particularly trustworthy at this point.]

    • Some other considerations when comparing old and new. From the PSC site: “Comparisons of PIOMAS estimates with ice thickness observations show reduced errors over the prior version. The long term trend is reduced to about -2.8 10^3 km3/decade from -3.6 km^3/decade [sic] in the last version. Our comparisons with data and alternate model runs indicate that this new trend is a conservative estimate of the actual trend. ”
      So, the current trend (-2.8) is regarded by its authors as a conservative estimate (ie that the downward slope is, if anything, greater than that being reported). I don’t recall version 1 making any such disclaimer. While I couldn’t put a number to it, this suggests that the -2.1 to -3.6 10^3 km3/decade 95% CI Tamino calculates should probably be a smidge higher – that the version 1 results are more clearly within the 95%CI of version 2 than is apparent at first glance.

      Prior to the release of version 2 and its results, PSC only published a graph of their output, not the raw data, so several readers at Neven’s Arctic Sea Ice blog tried to interpolate the source data from the graph, with fair results. Subsequently, Dr Zhang made output for September available which showed a flatter trend than the September data from the published graph . Dr Zhang explained that the graph was built from ensemble results, while the output he had provided was from the run that produced results most consistent with actual ice thickness measurements. That trend was 0.8 x 10^3 km^3/decade flatter than the one derived from the published graph – exactly the same difference that we see between version 1 and version 2. That suggests to me that while the ensemble results have changed somewhat, there is very little difference between the best results from version 1 and the ensemble results from version 2. It appears that the change is not so much significantly different output, as more internal consistency between runs.

      I find it a strange notion that making a revision to a model shows that the revised version is not accurate. That seems quite a non sequitur to me – I mean, if I cleaned up a dataset and reduced the error rate from 10% to 0%, by Andrew P’s logic, that would prove that the cleaned up data was unreliable (an argument we have all heard with regard to temperature datasets, of course). The only conclusion I can draw from the development of a new model is that the OLD version was imperfect. Who’da thunk it?

  33. Was looking at this North Pole Webcam picture with spreading melt-ponds. Reading how CryoSAT-2 is going to determine ice-tickness, they’ll have something to work on to get it right in the summer months. It’s a major issue for all the Sea Ice Trackers anyhow… the melt-ponds that is.


  34. Rob Dekker

    Moderator : please ignore previous post. Too many spelling mistakes. Here is a better one :

    I just bet $10,000 (with William Connolley) in part based on PIOMAS (and TOPAZ) ice volume reports.
    So you can imagine I was shocked to see the first Cryosat ice thickness map.

    In my opinion, the Cryosat map does not make much sense at all. Not only is the projected ice thickness inconsistent PIOMAS and TOPAZ numbers, but it also seems to be inconsistent with past ICEsat and submarine measurements in Jan/Feb, and even with in-situ measurements of the Cryosat validation team itself (made later, in April) .

    There is a note from NSIDC Julienne Stroeve on WUWT which is revealing what may be at the root of the excessive ice thickness presented with this map :

    Dr Haas at the IUGG meeting showed 2 locations in his presentation regarding airborne validation sites used for comparison with the Cryosat map. Unfortunately both were over mutliyear ice. My understanding thus far is that none of the firstyear thicknesses were validated and those regions appear to be too thick based on thermodynamic growth rates and observed winter temperatures. I am sure the Cryosat team is busy doing more validation from their recent field campaign, so we should know more hopefully soon.

    I agree with Sekerob that we should give the Cryosat team at least a full melt season time to calibrate the system. After all, measuring ice thickness with cm accuracy on a solid ice slab (Jan/Feb) with a previously uncalibrated satellite system is simply asking for trouble.

    I’m just disappointed that the Cryosat team was not up-front with a uncertainty measure when they presented this map.

  35. Another chart from the [my] playgrounds, not seeing anything similar already in circulation [from recollection]. This one of global sea ice area, where the Antarctic [horizontal] growth is oft used as an offset ”escape” argument, most all know is just waiting on kicking in hard. Build on Cryosphere Today data, computed 1 & 2 Std.Deviation and plotted on top various reference years to get an idea how the world sea ice is doing. Seems the path has closely followed 2007 so far and short periods been above, some below.

    What I was wondering was while looking at this is if there is a known regular delay between world temperatures and response of the annual sea ice cycle. Specifically, 2010 was the hottest year we know on modern record [fractional and contested by the Aint Truthies]. I’d imagine that the departure of 2010 set the base for the 2011 showings.

    Any takers to comment [before gaffing myself into the ignorant corner] ?