One graph to rule them all

Given that I’ve written a number of posts about the so-called “pause”, I thought I would mention a recent paper by Lewandowsky, Risbey & Oreskes called On the definition and identifiability of the alleged “hiatus” in global warming. I don’t want to say too much about it as there are various articles about it already.

Credit : Lewandowsky, Risbey & Oreskes (2015)

Credit : Lewandosky, Risbey & Oreskes (2015)

All I really wanted to do is mention the figure on the right. The top panel is the linear trend for a particular vantage year (end year looking back in time), and the number of years included. The colour is the actual trend, and the dots are those combinations of vantage years and years included for which the trend is statistically significantly different from 0 (no trend). The bottom panel is essentially just illustrating this showing which vantage years and years included have p-values less than 0.05 (the standard test of statistical significance). What it shows is that if you include less than 17 years, trends are typically not statistically significant, while if you include 17 years or more, they are almost always significant – for the time period considered, at least.

So, this figure essentially shows what’s been pretty obvious to many; if you define the “pause” as the trend being statistically consistent with 0 (no trend) then if you consider periods that are too short, you will almost always find a “pause”. Given the natural variability in the data, you need to consider periods of – typically – more than 17 years if you want to determine if there is a non-zero trend, or not. This is essentially what Richard Telford pointed out when he illustrated that Ross McKitrick’s paper was essentially describing a recipe for a hiatus.

So, I think the recent Lewandowsky, Risbey & Oreskes paper is simply illustrating nicely what has been pretty obvious to many people for a while now. This doesn’t mean that the recent period isn’t interesting from a model/observation perspective and that interesting and unexpected things didn’t happen. It’s just clear that there hasn’t really been any kind of “pause”.

This entry was posted in Climate change, ClimateBall, ENSO, Science and tagged , , , , , , . Bookmark the permalink.

74 Responses to One graph to rule them all

  1. Magma says:

    So what you’re claiming is one might need ≥18 years’ worth of data to reliably identify long-term trends in a system exhibiting independent annual, multi-year and decadal variations? What??

    That… that’s just crazy talk.

  2. Nice graph.

    The diagonal stripes are also interesting. There may be long-term variations in how accurate trends are because the amount of data and its quality can change. However, you would not expect strong year-to-year variations in the trend uncertainty. This is a statistical artefact, also the uncertainty has an uncertainty.

    This problem is, again, the worst if you have a short-term-trend fetish.

  3. John Mashey says:

    Indeed, good graph, more comprehensive than the quickie I did, 5,10,15,30 year slopes, but similar idea:
    what’s the regression slope looking back for windows of different lengths.

  4. Magma says: “So what you’re claiming is one might need ≥18 years’ worth of data …

    Let’s pretend you are serious. To be honest, I am surprised that this period is so short. The trends are surprisingly consistent already above 18 years.

  5. dikranmarsupial says:

    An interesting question would be “if you are going to claim global warming has stopped on the basis of a 17 year period without a trend, how many years would you expect to wait before being able to make the claim?”. There is a hidden multiple hypothesis testing issue here, which if you took it into consideration as well you would need more than a minimum of 18 years.

    Statistical power is something that too few really understand well enough to safely argue for the null hypothesis (as those arguing for a pause generally do) rather than against the null hypothesis, which is the usual scientific practice (intended to enforce a degree of self-skepticism).

    I may have mentioned this before ;o).

  6. Tom Curtis says:

    Fairly obviously, the number of years to be sure of statistical significance will depend on the temperature index used. In particular, it will be much larger with RSS and UAH TLT records given the much larger temperature variations from year to year. Consequently those indices will be come the last refuge of scoundrels sorry, the deniers.

  7. I just noticed that the GWPF have altered their logo to make it look more like a pause. I don’t suppose anyone can find a copy of the old version? http://www.thegwpf.org/

  8. john,
    I don’t have a copy of an old one, but I do have an alternative 🙂

    Tom,
    Indeed, and one could also do what Foster & Rahmstorf did, which was to try to remove some of the noise.

    Dikran,
    I think you might have 😉

  9. The Very Reverend Jebediah Hypotenuse says:
  10. Eli Rabett says:

    IT’S THE SURGE MAN, THE SURGE.

  11. verytallguy says:

    Johnrussell40,

    Don’t worry, the logo is such an obviously intellectually dishonest attempt to deceive I doubt they’d ever get any reputable academics to sign up to their advisory board.

    So they’ll never get any attention.

  12. verytallguy says:

    Eli, you have a one track mind

  13. BBD says:

    If you are reading this, Eli, will you *please* do something about BP.

  14. Willard says:

    Was there any graph in Ross’ econometric schtick?

  15. John Mashey says:

    This is rather akin to the cover of The Hockey Stick Illusion, in which the handle is horizontal, thus “proving” the hockey stick made the MWP and LIA disappear … in place of the decling regression line found in MBH99 and IPCC(2001).

    But if somebody bases key claims on an unsupported claim from a dog astrology journal, what would one expect? 🙂

  16. Willard,
    I think his Figures 5, 6, and 7 are essentially equivalent to what was done here, but using only 2014 as the end year.

  17. Willard says:

    I have no idea what “essentially equivalent” means, AT. What I see is that Ross used a Y named “trend magnitudes,” which helps him reveal some kind of decline in HadCRUT95, UHI, and RSS.

    Besides, I’m not sure what Ross means by Overall this analysis confirms the point raised in the IPCC report regarding the existence of the hiatus and adds more precision to the understanding of its length, since he basically introduced a definition. Definitions don’t confirm anything. Nor do they add precision to understanding. At best definitions result from a more precise understanding of a phenomenon. It would be like saying we have a more precise understanding of autism because of our definition of it.

  18. Ethan Allen says:

    AFAIK, Nick Stokes did it better than this paper did a few years ago, interactively even.

  19. Ethan Allen says:

    Ah, I see the Rev already linked to one of Nick’s posts.

  20. Eli Rabett says:

    BB, let him have the last word. If that doesn’t work he can get his own thread.

    (Don’t feed the troll is a good policy)

  21. Willard,
    Oh, I just meant that his Figures 4, 5, and 6 (not, 5, 6, and 7) are equivalent to a vertical strip in the figure here with the vantage year set at 2014.

    Besides, I’m not sure what Ross means by Overall this analysis confirms the point raised in the IPCC report regarding the existence of the hiatus and adds more precision to the understanding of its length, since he basically introduced a definition.

    Oh, yes, I agree. He essentially defined a hiatus and then found one, which was virtually guaranteed given his method.

  22. Tom Dayton says:

    John Mashey, I can’t see your linked graphs with Firefox on Mac.

  23. John Mashey says:

    Tom: try again, although that site has sometimes been flakey lately.
    But, I get it with Chrome or Firefox on WIn8.1 and Safari on iPhone 6Plus,

  24. From a science communication perspective I think it is also helpful to see what Lewandowsky, Risbey & Oreskes are not saying. They are not saying that there aren’t periods of time where it looks like there is some kind of pause or hiatus. Certainly there are quite a few white blue and even purple squares at the bottom of graph A.

    A rational response would be, “OK it looks like there may be a pause in global wamring looking at this part of the dataset for these years. What is actually going on?” Then when you look at the larger picture, over a longer time frame, you see no real letup in the temperature rising. Big Picture == No Pause.

    So yes there are times when there is the appearance of a hiatus, but in the long run it’s not real.

  25. matt says:

    Had not seen that alternative GWPF logo. Great. Should be used more often

  26. mdenison says:

    Arguing about a pause really comes down to the fact that natural variation make warming appear faster of slower at all times scales and some like to make political capital out of it when it is slower.

    Why not redefine how we estimate the trend by fitting a quadratic to the last 60 or even a 100 years of data. This appears to me to give an admirable fit and shows quite clearly that warming is accelerating. It is warming fast now than at any time in the past 100 years. If you compare the warming rate given over 30 or 17 years to the quadratic you see that these short term trends are sometimes faster or slower. At present it is slower, in the 1990’s it was faster. However when you look at all the data it is clear that true trend now is at a record high and accelerating.

    If a linear trend is deemed appropriate I think the following is important. Using a linear trend must eventually break down since it is self evident that the warming if not linear. So there must be a maximum time scale over which a linear trend can be used and beyond that the next option is a quadratic. Instead of trying to use the shortest time scale possible the longest linear time scale should be identified and that always used if the data is available. I think that time is between 30 to 45 years. Any one using less than the longest time scale available should provide clear justification since not using the maximum available data inevitably reduces precision.

    Given that RSS, the deniers favourite, only has about 35 years of data it is clear that only a linear trend over the full data is justifiable since there is insufficient data to justify any other method. Deniers use only half the RSS data. They have no justification for doing so other than to cherry pick a short term fluctuation.

  27. I really like those graphs you’ve included, ATTP.

    Also along the lines of ‘things we already know, but are well expressed by recent research’, another new article by Justin Farrell in PNAS shows that a focus on temperature trends is one of the two most dominant argumentative strategies of ‘skeptical’ organisations that receive money from ExxonMobil or the Koch foundation.

    These groups have placed an awful lot of their eggs in the ‘hiatus’ basket, which makes me agree with Tom Curtis above, that their attention will be more and more fixated on RSS and UAH TLT data -but even that makes them hostages to fortune. Isn’t it the case that the satellite datasets are more reactive to ENSO, or have the recent changes to them reduced that sensitivity?

    A link to the Farrell article:
    http://www.pnas.org/content/early/2015/11/18/1509433112.abstract

  28. Why not redefine how we estimate the trend by fitting a quadratic to the last 60 or even a 100 years of data. This appears to me to give an admirable fit and shows quite clearly that warming is accelerating.

    Warming is continuing.

    But warming is not accelerating:

    That would stand to reason since rates of radiative forcing increase are not accelerating.

  29. John Mashey says:

    If a time series has a year-to-year variability substantially greater than a year to year trend, it * *has* to have periods of seeming slowdown, or even reversal and ENSO and other ocean oscillations do a good job of ensuring that.

    We don’t have this problem with Keeling curve animation, do we? Why, the trend exceeds the year to year variability. (Of course, it varies pretty predictably by seasons.)

  30. Magma says:

    @ Victor (2nd post from top): the diagonal striping is an artifact of plotting an average with increasing n vs. the end year. In some fields of geophysics similar plots of apparent depth vs. spread are called pseudosections. Without redoing it, I suspect the pattern would resemble inverted Vs if the calculated slopes were based on the midpoint rather than the endpoint.

  31. Magma says:

    Oops. Not inverted Vs… I was still picturing geophysical pseudosections.

  32. mdenison said on November 25, 2015 at 12:26 am,

    “Arguing about a pause really comes down to the fact that natural variation make warming appear faster of slower at all times scales and some like to make political capital out of it when it is slower.

    Why not redefine how we estimate the trend by fitting a quadratic to the last 60 or even a 100 years of data. This appears to me to give an admirable fit and shows quite clearly that warming is accelerating.”

    Yes, the long-term warming underneath the multidecadal variations has been accelerating and still is even as we speak. See my comment

    https://andthentheresphysics.wordpress.com/2015/05/03/the-impact-of-the-hiatus/#comment-54920

    on May 4, 2015 at 7:26 am and some of the subsequent comments by other commenters under ATTP’s post “The impact of the “hiatus”?”, specifically the first of the three graphs I presented, this first graph being of a 60 year running mean covering the warming since the late 1800s, which shows this 60 year moving average tracking the path of a positively accelerated curve (the curve of a convex function on a given interval). That is, all the multidecadal variations are essentially filtered out, exposing that the warming since the late 1800s underneath all those multidecadal variations follows a relatively smooth (and positively) accelerated track. (The other two graphs support this.)

    I also have pointed out that Tamino at his blog Open Mind has shown that the warming is accelerating. I give a link to the details of how he did it starting with links I give at comments such as

    https://andthentheresphysics.wordpress.com/2015/05/30/hmmm-entering-a-cooling-phase/#comment-57068

    on May 30, 2015 at 2:22 pm under ATTP’s post “Hmmm, entering a cooling phase?” By the way, I presented at this comment a long-term 30 year running mean along with that 60 year running mean. Compare the two. (I like these kinds if graphs because, as I pointed out, I think they are a lot more understandable for the common person than the more advanced statistical graphs.) Anyway, here’s the link to Tamino’s post at Open Mind, posted on May 19, 2015:

    https://tamino.wordpress.com/2015/05/19/response/

    He was “debating” – actually, educating – someone who claimed to have a bachelor’s in math who claimed that the warming was not accelerating. Please read the comments, since Tamino continues to give throughout the comments relevant information. One of the commenters, Barton Paul Levenson, gives some especially relevant information.

    Actually this “debate” takes place in the comments under two posts by Tamino. The post I just gave is the second post, and the first post is this one, posted earlier on May 14, 2015.

    https://tamino.wordpress.com/2015/05/14/temperature-data-update/

    This “debate” ended in the comments of this earlier post. I think these two posts by Tamino and the relevant comments under these posts are part of what should be must reading. It’s an education as to how this “debate” over whether the warming underneath the variations is accelerating is just that, not a real debate at all.

  33. anoilman says:

    Anders… clearly that’s a fake.. I have my sources.

  34. Nick Stokes says:

    T Eddy,
    “But warming is not accelerating:”
    I don’t think you are using the latest data there. There is a facility here which will draw such plots for you, using latest data. You can choose datasets, rescale etc. Here is what it shows for those indices individually.

    The recent warmth has been pushing the whole curve up. As you’ll see, for years starting 1970-1995, it sits at about 1.5°C/Cen, which is about expected. Your 2C/Cen relates to future decades.

  35. Nick Stokes says:

    As the Rev and others noted (thanks), Moyhu maintains a page here which has the functionality of those graphs, but monthly resolution, and data is kept up to date. It uses start year instead of trend duration on the y-axis, though duration is shown with faint white lines and on the right axis. It’s interactive, so you can click and get data about the point, and choose dataset and significance options. There is an active timeseries (with trend line) which responds. It uses Ar(1) autocorrelation noise model.

    Here is the plot of GISS trends since 1989:

    And here is the plot with trends that fail significant difference from zero masked:

    It is a bit different from the above, partly because of monthly, and partly because of the autocorrelation model.

  36. Nick,
    That’s a really useful tool. Thanks.

  37. Willard says:

    > But warming is not accelerating

    I think you mean that the warming acceleration is not accelerating, TE, unless you’re willing to state that you can stop a car with this kind of “not accelerating” trend.

    Incidentally, we call that kind of driving technique an American stop around here. You bring the acceleration to a hiatus, and then you accelerate again. The car never really stops moving.

    Calling something that never stops moving a “hiatus” is kind of strange, isn’t it?

  38. anoilman says:

    Officer! I totally bowed to the stop sign.

  39. Magma says:

    Officer: Do you know why I pulled you over, sir?

    Driver: Because you’re an alarmist?

  40. anoilman says:

    Arguing the economics of speeding is Richard Tol;

  41. Willard says:

    Another kind of hiatus:

  42. JCH says:

    I just assumed TE made graphs with all the data. Would there be any point to his not doing that? Skeptical minds do not want to know, so don’t answer.

  43. JCH says:

    Willard – as a ExxMob shareholder, I need to tell you that climate research is a dynamic, nonlinear process.

  44. Nick Stokes, your graph does not indicate exactly what trends are being represented.

    The peak trend through the last complete year ( 2014 ) is from 1974 and trends of shorter durations are less, which indicates deceleration.

  45. Marlowe Johnson says:

    On a totally unrelated note, WGIII techophiles (I’m looking at you Mr. Mashey) might find this post by Brett Victor to be of interest. http://worrydream.com/ClimateChange/

  46. Joshua says:

    Off topic –

    But I never let that stop me before, so no reason to start now…and I thought some folks here might find this stuff interesting.

    http://fivethirtyeight.com/features/science-isnt-broken/?ex_cid=endlink#part4

    Speaking of good graphs, IMO, that interactive graphic is really good.

    And I liked this too:

    http://fivethirtyeight.com/features/not-even-scientists-can-easily-explain-p-values/

  47. Kestrel27 says:

    I thought of posting this on ‘Notalotofpeopleknowthat’ but at the risk of getting some hard brickbats it seemed more worthwhile to put my head in the lion’s den here than to sing to a chorus of people who disliked the Lewandowsky paper as much as I do. The ‘discussion’ part of the Lewandowsky paper seems to me to be entirely illogical. He may or may not be right that there was no ‘pause’ but the reasoning at the end can fairly be described as bizarre. Before going on I want to say that this is aimed at the reasoning in that part of the paper; it is not intended as an attack on the AGW theory since I understand neither the science behind it nor the statistics discussed in this thread.

    The reasoning in the paper seems to be as follows. First, he records that in the numerous papers discussing the ‘pause’ or ‘hiatus’ there is no agreement on what it is or consistency about what period it covers. Then there is a discussion of exactly how long a trend has to be before it is significant in statistical terms and therefore, I presume, to be taken as significant in the public arena. Concluding that none of the evidence points to a pause that is ‘significant’ he comes to the view that whatever pause there was, and I invite people to spot the logical fallacy, cannot be shown to exist and therefore does or did not exist. Despite that he has to find some way of talking about it, whatever it may be, and chooses the word ‘fluctuation’, which is probably as good as any, before reverting to ‘hiatus’ again later.

    The patronizing conclusion in the paper is that as the pause didn’t or doesn’t exist, being merely a fluctuation, climate scientists shouldn’t talk about it because that would encourage the general public to get the wrong idea. Given that he accepts the existence of a fluctuation it seems to me that the paper is only in part a ‘science’ paper; the ‘discussion’ is about the way the science should be presented. But the conclusion comes uncomfortably close to saying that scientists shouldn’t take part in discussions about the facts.

    If you question this consider a reverse example. Let’s assume that from 2016 there is a fifteen or so year period over which the earth warms by more than the models currently suggest; this would of course be a statistically insignificant period. We could call it the super warming interval (SWI). In about 2032 would Lewandowsky be writing a paper about the conclusions to be drawn and the appropriate language to be used in describing the SWI. Is he going to write a paper saying the SWI is just a fluctuation like the previous cooler fluctuation ending with 2015 and that it shouldn’t be talked about. Erm . . . no, I don’t think so! That paper, if written, would be likely to come from a contrarian, though he would of course be just as wrong as Lewandowsky is now. I’m pretty sure that Lewandowsky, on the other hand, would be saying that this was conclusive proof of the AGW theory and in reality would probably have been saying it for years.

    At the end of the day – glimpse of the obvious warning – the science has to fit the facts and not the other way round. It seems elementary to me that if if there is a period of stable or cooling temperatures then the longer it lasts, so becoming significant in both statistical and common sense terms, the more it raises questions about the science behind the AGW theory. By the same token if there is a period of more or less steady warming then the longer it lasts the more its existence supports the validity of that theory. The proposition that climate scientists shouldn’t discuss such fluctuations is surely tendentious and I am sure that as this thread shows there is, thankfully, no chance that his advice will be followed. There will always be such discussion and so there should be.

    In conclusion I am baffled, given the faith Lewandowsky must have in the science, as to why he thought it necessary or appropriate to write the discussion part of the paper as he did. Where did his confidence go?

  48. Kestrel,
    Two things. Firstly, you’re kind of right that we have to be careful of making some kind of big deal out of a period that might warm faster than the long-term trend. Variability can work both ways, it could speed thing up and slow things down. On the other hand, I think you’re slightly missing a key point about this paper. A crucial point is that if you consider less than about 17 years when trying to determine if there is a trend that is statistically different from 0, then typically you might find that the trend is statistically consistent with 0, even if we are warming relatively fast (< 0.2 degrees/decade). Hence you really can't use periods of about 17 years, or less, if you're trying to determine if there is some kind of warming trend, since the year-to-year variability can easily mask such a trend.

  49. We do not have to go to the future. In the 90ies the warming was faster than expected. No political pressure group named that period as far as I know and scientists just wrote that the warming was larger due to natural variability.

  50. Nick Stokes says:

    T Eddie,
    “The peak trend through the last complete year ( 2014 ) is from 1974 and trends of shorter durations are less, which indicates deceleration.”
    Well, I mistakenly cut off the recent period while processing. Here is another version, up to about 2010. I’ve shifted it so that the x-axis is the zero, so you can see all trends are firmly positive. You can generate for yourself here.

    Recent warmth is raising the right end of the graph faster than further left, so starting from years around 2010, you get a trend of about 3. If that is your criterion for acceleration, there is plenty. But it isn’t mine. You can see what is left of the “pause”. A region around 2000 where the trend is just 1 °C/cen in NOAA and GISS, less in HADCRUT.

    You can see the pause in proportion in the top right corner of the triangle plot I showed above. Brown means zero trend, so there was a short time when selected trends dropped below zero. It’s the blob I’ve labelled Pause. We’re rapidly leaving it behind.

  51. anoilman says:

    Kestrel27: I really can’t tell where you’re going with any of that.

    First, The temperature projections used do not show weather. So… before you can start talking about error or hiatus, you have to consider the difference between the projections and actual temperatures. Namely, weather (ENSO) volcanoes, and solar variance. This is not a new concern;
    http://iopscience.iop.org/article/10.1088/1748-9326/7/4/044035

    Second, In answering your question of what the hiatus is… it is and always will be any short term period in which you can see declines in year on year temperatures. aka ‘down the up escalator’. Its not a new meme, only a newish name.
    http://skepticalscience.com/still-going-down-the-up-escalator.html

    The Down the Up Escalator meme has been a moving target for some time since denial sphere is always attempting to include 1998, a monster temperature spike, in their data. First it was 5 years, then 10, now 17? Whatever… As soon as we have a new record temperature year, they will be resetting the duration required for their meme.

    The take home message is don’t look at short term data. If you want to look at short term data, understand the differences between it and the climate (Foster Rahmstorf). If you don’t want to understand that, then you should consider the statistical significance of viewing short term trends in the big picture (Lewandowsky, Risbey & Oreskes).

  52. Willard says:

    Your best comment to date, Oily One.

  53. Magma says:

    @ Willard: I don’t know. The Tol economics of speeding ranks high with me.

  54. BBD says:

    Yes, I laughed at that too. 🙂

  55. BBD says:

    Kestrel

    In summary:

    Pretending that nat var means low CS and nothing to worry about = intellectually dishonest tripe.

  56. BBD says:

    I should add that the problem of presentation would not exist but for the vociferous misrepresentation of science for political ends by the so-called ‘sceptics’. That’s why Lewandowski and others have felt compelled to write as they do.

    Blaming them for something for which you and your fellow ‘sceptics’ are entirely responsible is distasteful.

  57. Kestrel27 says:

    Thanks for the helpful comments. I particularly appreciate the trouble anoilman has gone to and his explanations are telling. I tried to make clear in my first comment that my criticisms were directed at the the conclusions in Lewandowsky’s paper rather than the scientific analysis which I am, unfortunately, in no position to debate. The only expertise I would lay any claim to is with words and language; not a lot of use in grappling with the complexities of climate change science and statistical analysis.

    In reply to BBD, I understand the frustration with the nature of the debate that scientists like Lewandowsky must feel but it is he who is responsible for the way he chooses to conduct the debate and not his opponents. I simply feel he has made the wrong choice. However wearying it may be patient repeated explanation, taking account of all the facts openly, is the right one. You say I’m a sceptic but actually I don’t have enough knowledge to place myself firmly in either camp; I want to understand more but lack of a scientific background means it isn’t easy. I’m not arrogant enough to think that anything I say, as a private individual with no public persona, is likely to influence Lewandowsky or any other climate scientist in any way.

  58. Willard says:

    If you insist, Krestel32, then perhaps you have better than play spot the fallacy:

    Concluding that none of the evidence points to a pause that is ‘significant’ he comes to the view that whatever pause there was, and I invite people to spot the logical fallacy, cannot be shown to exist and therefore does or did not exist.

    You go first, and quotes might be nice.

  59. If you want to discuss semantics, then a philosophy blog may be a better place than a science blog where people are interested in understanding nature.

  60. BBD says:

    In reply to BBD, I understand the frustration with the nature of the debate that scientists like Lewandowsky must feel but it is he who is responsible for the way he chooses to conduct the debate and not his opponents.

    Rubbish. L et al. is a response to a vociferously-aired contrarian meme. Its nature is unavoidably determined by the thing to which it is responding. Otherwise L et al. would be arguing irrelevantly at right angles to the topic.

    The problem – as ever – lies with the contrarians and they should have the intellectual honesty to own it.

  61. anoilman says:

    Kestrel27: None of this is new. I merely explained what the scientific community has been saying for a long long long long long time now. The only difference is the verbiage of the denial sphere which intern forces the scientific community to rebut.

    Of course its a 17 year hiatus… the ‘pause’ is past us now.

    Here’s James Hansen looking at the 10 year ‘pause’ in 2012;
    http://www.columbia.edu/~jeh1/mailings/2013/20130115_Temperature2012.pdf
    “The 5-year running mean of global temperature has been flat for the past decade. It should be noted that the “standstill” temperature is at a much higher level than existed at any year in the prior decade except for the single year 1998, which had the strongest El Nino of the century. However, the standstill has led to a widespread assertion that “global warming has stopped”.

    Examination of this matter requires consideration of the principal climate forcing mechanisms that can drive climate change and the effects of stochastic (unforced) climate variability.”

    If this year (or next) doesn’t beat 1998’s temperature, deniers will have to extent (move the goal posts) the length of the hiatus. If these years are the hottest, then 2017 will the start of a new ‘pause’. This is how the game is played with deniers.

    Its also not a new meme for someone to show up here and say they can’t understand any science, and that they think there’s something wrong with the science, and those who actually do it. This is what you are claiming. Care to explain you can possibly be skeptical of global warming science while simultaneously being unable to understand any of it? Where are you going with all that?

  62. Nick Stokes, as the paper indicates, trends more brief than 17 years, tend to not be statistically significant, so that would exclude trends since 1999. Trends 17 years and longer indicate deceleration.

  63. TE and his cherry-picks again.

  64. anoilman says:

    Lucifer (aka Turbulent Eddie): The dilemma is what to cherry pick next, right? Of course that comes with the baggage that you need to blindly stupidly not understand what it means or what it contains or if its even significant.

    Dates and times chosen are all relative to the point you want to make, right?

  65. BBD says:

    OilMan is a barrel of laughs at the mo.

    🙂

  66. Kestrel27 says:

    In response:

    Wradill.The logical fallacy is that because the pause is not significant in scientific terms it does not exist in ordinary parlance or, therefore, at all. The quote is ‘We conclude that the evidence does not support the notion of a “pause” or “hiatus” as an identifiable phenomenon that is implied by standard dictionary definitions and common understandings of these terms (my emphasis)’. I entirely accept, in the light of the comment from Andthentheresphysics and anoilman that had Lewandowsky limited himself to saying that there was no pause in scientific terms his reasoning would have been logical – but he didn’t.

    Victor Venema. I’m not entirely sure whether you are being serious. It was Lewandosky in his paper, not me, who introduced semantics when he argued that climate scientists should find new language to refer to the pause, hiatus, fluctuation, temporary blip or whatever. Simple definition of semantics from Merriam-Webster: ‘the meanings of words and phrases in a particular context’. Philosophy is another matter.

    BBD. I simply disagree. If I insult you that is my responsibility. How you react to the insult, possibly by turning the other cheek, possibly by insulting back or taking even more drastic action is yours; it is not right to argue that I have compelled the choice you make.

    anoilman. I largely agree with what you say but not with your last paragraph. I may not understand science but that doesn’t mean I can’t identify what seems to be poor reasoning in matters that are secondary to the science. I’ve said that I’m not firmly in the sceptic camp.

  67. Kestrel27 says:

    I’m sorry, the emphasis I meant to underline in the first paragraph didn’t show up but I hope it’s clear.

  68. Kestrel,

    I entirely accept, in the light of the comment from Andthentheresphysics and anoilman that had Lewandowsky limited himself to saying that there was no pause in scientific terms his reasoning would have been logical – but he didn’t.

    No, I think he’s saying the opposite. He’s saying that in scientific terms, warming has not stopped. The problem is that by using the term “pause” or “hiatus” people have implied that it has. Therefore, the strict semantic meaning of the terms is not consistent with the scientific evidence.

  69. BBD says:

    Kestrel

    BBD. I simply disagree. If I insult you that is my responsibility. How you react to the insult, possibly by turning the other cheek, possibly by insulting back or taking even more drastic action is yours; it is not right to argue that I have compelled the choice you make.

    So if you spout nonsense and I point it out, I am responsible for your nonsense?

    Be serious 😉

  70. Kestrel27 says:

    and Then There’s Physics. Thanks. I don’t want to be unfair and on reflection I think you are right about what he meant to say; I don’t think he expressed himself at all clearly though.

    BBD Oh dear! That is nothing like what I was saying, in fact it is pretty much the opposite. In the unlikely event that I spouted nonsense I would be entirely responsible and you would have nothing to do with it. My final response I think.

  71. BBD says:

    Kestrel
    So while not questioning the science in L15 you are concerned about the tone?

    Oh dear!

  72. anoilman says:

    Kestrel27 says: “I may not understand science but that doesn’t mean I can’t identify what seems to be poor reasoning in matters that are secondary to the science. I’ve said that I’m not firmly in the sceptic camp.”

    To be clear, I think you’re firmly in the septic camp.

    If you can’t understand the math and the science, I’m not sure how anyone can possibly illuminate this to you.

  73. Willard says:

    > The logical fallacy is that because the pause is not significant in scientific terms it does not exist in ordinary parlance or, therefore, at all. The quote is ‘We conclude that the evidence does not support the notion of a “pause” or “hiatus” as an identifiable phenomenon that is implied by standard dictionary definitions and common understandings of these terms (my emphasis)’.

    In the quote, Lew & al simply (well, not that simply, for it’s still academese) claim that the evidence doesn’t support it as “an identifiable phenomenon” if you approach it “in scientific terms.” In other words, there’s no scientific evidence that we’re having a paws, if by “paws” you mean a paws. The last part of that sentence should have been omitted because such appeal to common sense definitions can only be seen as petty. Now we can also see that it can add unnecessary confusion.

    More generally, I don’t think Lew & al can claim that non-scientific treatments of the paws don’t exist, since they study it.

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s