Arctic

The world has been getting hotter:

nasa1


The globe as a whole has warmed by about 1.1 °C (2 °F) since 1900.

But — not all parts of it are heating up at the same rate. Temperature in the Arctic in particular is rising faster than the global average:

nasa2

The Arctic, according to NASA data, has warmed by about 2.2 °C (4 °F) since 1900.

There are other data sets besides NASA’s, of course, and one which has made a particular effort to get the Arctic right is the data from Cowtan & Way, who used Kriging to estimate its changes. Their estimate of global temperature change since 1900 is, at 1.1 °C (2 °F), similar to NASA’s:

cw1

But their estimate of Arctic heating is considerably bigger:

cw2

It indicates that since 1900, Arctic temperature has risen about 2.9 °C (5.3 °F).

More heat generally means less ice. This is rather obvious in the Arctic lately, as 6 out of 10 months so far this year have set new record lows for the extent of sea ice (2016 values shown in red):

seaice

The extremity of the last couple of months might be even more visible in a plot of monthly average sea ice anomaly (the difference between the month’s value, and what’s “typical” for that month of the year):

seaice_anom

An interesting consequence of the reduction of sea ice is that in the Arctic, it enables more heat to migrate from the ocean to the atmosphere. That decreases the air temperature contrast between the equator and the pole, and that weakens the polar vortex. When the Arctic polar vortex is weaker, Arctic air can migrate southward more easily; this allows warmer air into the Arctic and colder air out. When the cold Arctic air mass dips south, it can bring surprisingly frigid conditions. Not too long ago, the polar vortex allowed cold Arctic air into Siberia — but now it has reached down to north America and we’re shivering through cold conditions because of it.

It’s one of those surprising things, that global warming can make some places colder sometimes. It’s not happening globally — the heat that the U.S. wishes it had right now has invaded the Arctic. But it is commonplace for climate deniers to take any outbreak of cold, anywhere, anytime, and offer it up as evidence against global warming … even when it’s because of global warming. Some who do so are simply opportunists, taking advantage of the simple-minded, while some of them are the simpleminded; anything requiring a hint of “take a moment and think about that” is just too much for them to handle.

Although things like temperature and sea ice extent will always show fluctuations (just because climate changes doesn’t mean we won’t still have weather), they will also continue to show dangerous trends. In the long run, we’ll keep burning fossil fuels, the amount of greenhouse gases in the atmosphere will continue to increase, the globe will continue to heat up, Arctic sea ice will continue to decline.

All the trends will be punctuated by temporary reversals when climate deniers will point and shout about how the trend has stopped — “pause pause pause!!!” Trends will also be punctuated by temporary extemes when climate deniers will point and shout about it all being just “weather weather weather!!!”, with a healthy dose of blaming everything on el Niño thrown in for good measure.

The longer deniers succeed in delaying action, the worse it will be for our kids. Fortunately, they’re a lot smarter and more courageous than climate deniers give ’em credit for.


This blog is made possible by readers like you; join others by donating at Peaseblossom’s Closet.

Advertisements

38 responses to “Arctic

  1. “with a healthy does of blaming everything on el Niño”
    typo alert: should be dose, not does

    [Response: Thanks. Fixed.]

  2. This is a totally geeky comment, but with regard to the mention of the famous geostatistical technique of kriging by Cowtan and Way in the above (spatial best linear unbiased estimation), note it has a competitor now:

    S. Leung, D. Cooley, “A comparison of a traditional geostatistical regression approach and a general Gaussian process approach for spatial prediction,” Stat 2014; 3: 228–239

    This method has been used in, for instance:

    Kopp, Kemp, Bittermann, Horton, Donnelly, Gehrels, Hay, Mitrovicak, Morrow, and Rahmstorf, “Temperature-driven global sea-level variability in the Common Era,” http://www.pnas.org/cgi/doi/10.1073/pnas.1517056113

  3. “with a healthy does of blaming everything on el Niño”

    after being completely ignorant of el Niño when it began their cherry pick but didn’t end it.

    Cognitive dissonance extraordinaire.

    • I rather doubt that the “skeptical” proprietors of the porky pie production line were “completely ignorant of el Niño” then, or indeed last weekend!

      Their readership on the other hand….

      • Yes I left out the “pretending to be”. The ones who don’t suffer from genuine cognitive dissonance know they are dishonest.

    • “after being completely ignorant of el Niño when it began their cherry pick but didn’t end it….The ones who don’t suffer from genuine cognitive dissonance know they are dishonest.”
      Is that really what you think, Chris? Or are you speaking out of frustration?
      Have you made an effort to understand their point of view?

      • Sheldon Walker,
        OK, I’m tempted to reply that I cannot see the world from the point of view of the denialists because my head will not fit up my ass. However, I will instead tell you what I see.

        I see that when the data are interpreted in the context of the dominant scientific theories of Earth’s climate, they present an unequivocal picture of a planet warming due to an increase in greenhouse gasses. When I look at the arguments presented by the denialati, I instead see facts removed from context, willful ignorance and appeals along the lines of, “But, Galileo…” I do not even see even the most rudimentary attempt to make sense of the data, but merely a dishonest and scattershot attempt to cast doubt on it. And then there are the conspiracy theorists…

        What, pray, am I missing?

      • Whose “point of view” are you referring to Sheldon? The “skeptical proprietors” or “their readership”? In either case you’ll need to expand upon your point, because I for one don’t understand it yet.

        FYI, I’ve consulted an eminent psychologist on the topic:

        http://GreatWhiteCon.info/2015/05/why-its-so-hard-to-convince-pseudo-skeptics/

      • Just tell us where you talked about the influence of the 1998 el Nino on the global warming trend before 2013 Sheldon.

        Then we’ll know you’re honest and not suffering from cognitive dissonance.

      • Harry Twinotter

        Sheldon Walker.

        “Have you made an effort to understand their point of view?”

        No. And the reason I don’t is an argument based on the assumption all opinions are equal is not a valid scientific argument.

      • Chris,
        I am putting a reply together for you. Because of the time zone difference between Tamino and me, it won’t be ready until tomorrow.

      • “Have you made an effort to understand their point of view?”
        I’m with Harry. Life is too short to look for rational explanations of irrational behaviour. If I could see a clear, logical expression of an alternate hypothesis, it would be worth looking at, but so much “skeptical” stuff is so internally self-inconsistent and illogical that all you can do is point out the problems, not “try to understand it”.

      • Dear Sheldon,

        Is there any chance you could reply to me too?

        TIA

      • Hi Chris (and Jim Hunt),

        you said, “Just tell us where you talked about the influence of the 1998 el Nino on the global warming trend before 2013 Sheldon. Then we’ll know you’re honest and not suffering from cognitive dissonance.”

        I don’t think that I did much talking about the influence of the 1998 el Nino on the global warming trend, before 2013.

        I remember hearing a lot of Alarmists complaining about skeptics starting trends in 1998.

        I have never been a person who believes in censoring science. Like banning people from starting a trend in a certain year. To me, that is a political move, and I am not normally a political person.

        I am a very skeptical person by nature. If somebody tells me that A = B, then I automatically think, “I wonder if there is some circumstance where A is not = B”.

        So if somebody tells me to not start a trend in 1998, the first thing that I will do, is start a trend in 1998.

        I have learnt a lot of things because of my skeptical nature.

        While researching for this reply, I have calculated trends starting in every year from 1990 on, and ending in every year from 1991 on. So I looked at the 17 trends which start from 1998, and end in every year from 1999 to 2015.

        I have some bad news for you, Chris. 1998 is a trivial problem compared to what has happened recently.

        Here are some examples:

        The lowest warming rate for a trend starting in 1998, is the one ending in 2013 (15 years). It has a warming rate of 0.97 degC/century.

        I am only going to give you the warming rates for trends of 10 years or longer, because some of the shorter ones are negative (cooling), and we don’t want that, do we.

        The warming rate for a trend starting in 2001, and ending in 2011 (10 years) is 0.57 degC/century.

        The warming rate for a trend starting in 2002, and ending in 2012 (10 years) is 0.21 degC/century.

        The warming rate for a trend starting in 2002, and ending in 2013 (11 years) is 0.29 degC/century.

        The warming rate for a trend starting in 2002, and ending in 2014 (12 years) is 0.63 degC/century.

        Hopefully you are beginning to appreciate that 1998 is not a major issue. Let the skeptics start there trends in 1998, it might stop them noticing that there are much lower warming rates that have occurred recently.

      • I don’t think that I did much talking about the influence of the 1998 el Nino on the global warming trend, before 2013.

        OK. So you don’t deny that those who were completely ignorant of el Niño when it began their cherry pick but didn’t end it and then suddenly knew all about el Niño in 2016 were displaying cognitive dissonance (apart from those who were just plain dishonest).

        Thank you for not disagreeing with my point.

        By the way, the complementary cherry pick to using an el Niño for the starting year is using a la Nina for near the ending year/years (2011/12 e.g.) Since it doesn’t suit their purpose, global warming denialists/minimizers (like yourself Sheldon) don’t have much interest in talking about la Ninas.

      • @Sheldon Walker: Thank you for your post. It was revealing, and, so, communicated well.

        What your post showed is that the problem in question has to do with climate data only as a secondary point but, rather, the primary problem and point is a problem of epistemology. In particular, your “skeptical” approach treats the systems in question as if they are deterministic or, equivalently, systems about which legal reasoning can be relevant in analysis, when, with an enormous about of observational and physical evidence, as well as physical theory, these are stochastic systems.

        So, to illustrate with reducto ad absurdum, assessing the mass distribution of numbers of colored balls in a barrel of randomly mixed colored balls, say Blue, Green, and Red should proceed with a sampling plan which is consistent with the stochastic nature of the question, and there are many schemes which deal with this well, such as the Pólya urn model and its relatives which actually have substantial applications. (See the references at the link.) In the case proposed, it is as if the observer likes Red and Green balls, but dislikes Blue ones, and, so, begins to pay attention to Red balls and Green balls, discounting Blues, and then reporting the result as if it were representative of the population in the barrel.

        Indeed, in the case in hand, the assessment seems to applaud some kind of analysis based upon an arbitrary choice from the space of all possible assumptions, arguing that only by repeated sampling from such a space can some kind of notion of “truth” be obtained. This is similar to positing that the only true posterior distribution which can be had for a Bayesian analysis is by averaging over all possible instances of prior distributions. That might be true if the sampling distribution of the data were known well (sometimes called the likelihood function) and, in the limit is equivalent to arguing from uninformative priors. What’s worse is the argument (even at Wikipedia) that using an uninformative prior is somehow equivalent to the classical statistical view of setting aside priors and doing something suboptimal like maximum likelihood estimation or analysis. (See Stein’s paradox.) And, as it turns out, there are serious problems with using such uninformative priors, and, so, even if one sets aside the willful ignorance of the proposed procedure and its inefficiency for inference, so are there problems with it.

        I’m sure many others will have something to say about this proposal. From my perspective, it is so outlandish, continuing to pay attention to it is a waste of both mental energy and time when there are so many more important things to do.

      • Hi Sheldon,

        You haven’t answered my implicit question yet though. I think you should!

        Why It’s So Hard to Convince Pseudo-Skeptics

        Be sure to watch the video until the punch line. Have you read Bill the Frog’s magnum opus yet for that matter? I think you should!

        How to Make a Complete RSS of Yourself (With Sausages)

      • Sheldon,
        Your reply makes it clear that you have learned nothing. The World Meteorological Organization has shown that trends over a period less than 30 years are too noisy to convey any real information. You are a prime example of stupidity sent to college.

      • Sheldon, if you aren’t afraid of getting cooties from reading actual science, go read the following:

        https://skepticalscience.com/record-hot-2015-glimpse-future-global-warming.html

      • Since it doesn’t suit their purpose, global warming denialists/minimizers (like yourself Sheldon) don’t have much interest in talking about la Ninas.

        By the way Sheldon, thanks for not mentioning la Nina when you made this statement:

        The warming rate for a trend starting in 2002, and ending in 2012 (10 years) is 0.21 degC/century.

        You have proven my point so far.

  4. Whilst the MSM have been concentrating on the “Arctic Report Card” there was an even more interesting (IMHO) press conference at AGU yesterday:

    Arctic Sea Ice News from AGU

    The R/V Lance spent 6 months frozen into the sea ice north of Svalbard, and observed a variety of “surprising weather” events during that time. By way of one example:

    Winter storms caused the sea ice to drift so fast that it increased mixing of the water beneath the ice. Deeper, warmer water was mixed up closer to the sea ice, causing it to melt from below despite winter air temperatures that were below freezing.

    It seems there’s heat migrating from the ocean to the bottom of the ice in winter, and also heat migrating from the atmosphere to the (apparently increasing) snow cover:

    One winter storm raised the air temperature from -40 F to +32 F in less than 48 hours, while the moisture in the air increased 10 times. All of these factors significantly warm the surface of the snow, even in mid-winter, and slow the growth of ice.

  5. You do realise all this information is going to be shut down.
    America is going to go to the { We do not do Science } paradise.
    That will be wonderful and we all can rite a note for the great, great grand kids explaining how we protected them by burning FF so they could have a wonderful life.
    Are we honestly moving to a world where the Educated Idiots are honestly in charge to me it seems so.

  6. I’ve noticed that some fake sceptics have been using Ole Humlum’s ‘climate4you’ interpretation of the HadCRUT4 Arctic (70-90N) data to claim that temperatures were warmer there in the 1930s and 40s than they are now: http://climate4you.com/images/70-90N%20MonthlyAnomaly%20Since1920.gif

    I notice that this plot is ‘area weighted’, and suspect that the apparent discrepancy with the other Arctic temperature providers is the result of HadCRU’s limited Arctic data in the early part of their record. My understanding is that HadCRU doesn’t infill regions where no data are available. I suspect therefore that the early part of the HadCRUT Arctic record mostly relies on temperatures from relatively low latitude (i.e. warmer) stations.

    Area weighting would, I imagine, extrapolate these warmer temperatures out across an entire region, the bulk of which would in fact have been much cooler.

    I wonder if Tamino, or anyone else here, understands this better than I do? It might make an interesting post. Also, it would be nice to have an expert source to quote should I ever get into a discussion about this with someone who misuses the HadCRUT4 Arctic data.

    • There are a number of issues with HadCRUT4 in the Arctic. You’re correct that higher monthly anomalies in the early part of that record are really just due to data scarcity causing increased noise. Which is obvious if you look at the amount of negative anomalies which also happened around that time, compared with zero negative anomalies over the past decade. Using decadal averages to smooth over the noise reveals that the most recent is over 0.5ºC warmer than the 1930s-40s peak.

      Your point about lower latitude stations is probably not right I would think, because data is anomalised before averaging, so stations being in warmer locations shouldn’t make a difference. However, anomalies lie at the heart of other issues with HadCRUT4. The basic problem is that most of the sea surface area north of 70ºN has no valid baseline reference temperature, due to lack of historical measurements in those locations. Instead HadCRUT4 (or rather its component HadSST3) simply compares to a value of -1.8ºC all year round. To illustrate the problem, there were recent recordings of Arctic temperatures 30ºC above normal. At that time of year 30ºC above normal means about 0ºC in absolute temperature. HadCRUT4 would compare the 0ºC figure with the -1.8ºC baseline to give a +1.8ºC anomaly, even though the actual anomaly was +30ºC. Aside from biasing the annual average one perverse result is that July is shown to be the fastest warming month in HadSST3 for 70N-90N, even though it’s likely the slowest warming month in reality.

      In HadCRUT4 this bias doesn’t affect the 1930s-40s as much as recent highs because there were very few SST measurements in that earlier period. The early record appears to be mostly a land-average, which transitions to land+SST over time up to the present.

      • paulskio,

        Thanks for your illuminating response. I guess HadCRU’s Arctic treatment is one component contributing to the fact that it’s long term global warming trend is lower than GISS. No wonder the fakes target HadCRUT4 Arctic, though!

  7. “Not too long ago, the polar vortex allowed cold Arctic air into Siberia — but now it has reached down to north America and we’re shivering through cold conditions because of it.”
    Beware for it is not interesting. Winter in America used to be cold. This wave is no record breaker like the one of December 1983 when Arctic Sea Ice was more or less intact.

    There is no increase in midlatitude cold waves: there is only decrease. In both number and intensity.

    Arctic Amplification is creating ‘stuck pattern syndrome’. It may send some cold to mid latitudes, but this is nothing new; what is new is that these recent cold waves are increasingly physically unable to set anything near cold records.
    The true change, and the true consequence of Arctic Sea Ice demise, is e.g. the record hot autumn in the US and especially the margin that record was set.

  8. David – That’s the very issue that Cowtan & Way are endeavouring to address, and BEST as well to some extent. For much more on all that see this “interesting post” by Kevin Cowtan:

    http://GreatWhiteCon.info/2015/02/the-telegraph-is-wrong-again-on-temperature-adjustments/

  9. At first glance the arctic seems to have the same trends as the globe, just magnified. And we are getting “crazy” behaviour in the arctic more often. Like sea ice right now, and the sea ice minimum in 2012.

  10. @David, I cannot answer you on Humlum’s site, climate4you site since they don’t seem to give detailed documentation of their methods for how the data there is prepared for presentation, and don’t track that, as far as I can tell, back to their own published papers. (Admittedly, I’m not familiar with climate4you, and I really don’t want to invest time understanding it.) I do know that Humlum, et al published papers have data sourcing, processing, and interpretive flaws, and these have been carefully documented (open access) by Benestad, Nuccitelli, Lewandowsky, Hayhoe, Hygen, van Dorland, Cook, and see especially the details in their supplementary material for that article.

    Accordingly, without justification by Humlum, I’d be very careful taking anything on that site at face value. I can tell you, for example, that Humlum’s comments on data smoothing are not sophisticated, and in some cases wrong. (In particular, they act and write as if there were no way to assess how overfit a smooth is, in other words, how much risk it might have for failing to represent out of sample data. There are at least three ways.) I’ve read the same from, for instance, science denier William Briggs and wrote a blog post about it.

  11. SW: So if somebody tells me to not start a trend in 1998, the first thing that I will do, is start a trend in 1998.

    BPL: And if someone tells you not to divide by zero, the first thing you’ll do is divide by zero. Right? Because that’s the arithmetic equivalent of the statistical statement you just made above.

    And perhaps if someone tells you not to stick an aluminum fork in the light socket, the first thing you’ll do is stick an aluminum fork in the light socket?

  12. Sheldon, starting a trend analysis from 1998 without correcting for el nino is like a paediatrician monitoring a child’s growth and starting with the height recorded on the wall chart while the child was standing on a chair.

    It is not skeptical or revealing. It is plain dumb.

  13. Perfessor Whitehouse is far to modest. Using the Whitehouse method of trendpoint analysis, it has been conclusively shown that global temperatures have been trending down since 1973. https://therationalpessimist.files.wordpress.com/2012/02/skeptics-view-of-global-warming.jpg
    We could all learn a few things about data analysis if we consider the perfessor’s work. Exactly what we can learn is left as an exercise for the astute or ambitious student.