The latest head in the sand excuse from climate science: the global warming pause 'never happened'

From the “fighting denial with denial” department comes this desperate ploy and press release written to snare headlines with gullible media. Meanwhile, just a couple of days ago the UK Met office said the global warming pause may continue.

headinsand

Global warming ‘hiatus’ never happened, Stanford scientists say

A new study reveals that the evidence for a recent pause in the rate of global warming lacks a sound statistical basis. The finding highlights the importance of using appropriate statistical techniques and should improve confidence in climate model projections.

From STANFORD’S SCHOOL OF EARTH, ENERGY & ENVIRONMENTAL SCIENCES via press release

An apparent lull in the recent rate of global warming that has been widely accepted as fact is actually an artifact arising from faulty statistical methods, Stanford scientists say.

The study, titled “Debunking the climate hiatus” and published online this week in the journal Climatic Change, is a comprehensive assessment of the purported slowdown, or hiatus, of global warming. “We translated the various scientific claims and assertions that have been made about the hiatus and tested to see whether they stand up to rigorous statistical scrutiny,” said study lead author Bala Rajaratnam, an assistant professor of statistics and of Earth system science.

The finding calls into question the idea that global warming “stalled” or “paused” during the period between 1998 and 2013. Reconciling the hiatus was a major focus of the 2013 climate change assessment by the Intergovernmental Panel on Climate Change (IPCC).

Using a novel statistical framework that was developed specifically for studying geophysical processes such as global temperature fluctuations, Rajaratnam and his team of Stanford collaborators have shown that the hiatus never happened.

“Our results clearly show that, in terms of the statistics of the long-term global temperature data, there never was a hiatus, a pause or a slowdown in global warming,” said Noah Diffenbaugh, a climate scientist in the School of Earth, Energy & Environmental Sciences, and a co-author of the study.

Faulty ocean buoys

The Stanford group’s findings are the latest in a growing series of papers to cast doubt on the existence of a hiatus. Another study, led by Thomas Karl, the director of the National Centers for Environmental Information of the National Oceanic and Atmospheric Administration (NOAA) and published recently in the journal Science, found that many of the ocean buoys used to measure sea surface temperatures during the past couple of decades gave cooler readings than measurements gathered from ships. The NOAA group suggested that by correcting the buoy measurements, the hiatus signal disappears.

While the Stanford group also concluded that there has not been a hiatus, one important distinction of their work is that they did so using both the older, uncorrected temperature measurements as well as the newer, corrected measurements from the NOAA group.

“By using both datasets, nobody can claim that we made up a new statistical technique in order to get a certain result,” said Rajaratnam, who is also a fellow at the Stanford Woods Institute for the Environment. “We saw that there was a debate in the scientific community about the global warming hiatus, and we realized that the assumptions of the classical statistical tools being used were not appropriate and thus could not give reliable answers.”

More importantly, the Stanford group’s technique does not rely on strong assumptions to work. “If one makes strong assumptions and they are not correct, the validity of the conclusion is called into question,” Rajaratnam said.

A different approach

Rajaratnam worked with Stanford statistician Joseph Romano and Earth system science graduate student Michael Tsiang to take a fresh look at the hiatus claims. The team methodically examined not only the temperature data but also the statistical tools scientists were using to analyze the data. A look at the latter revealed that many of the statistical techniques climate scientists were employing were ones developed for other fields such as biology or medicine, and not ideal for studying geophysical processes. “The underlying assumptions of these analyses often weren’t justified,” Rajaratnam said.

For example, many of the classical statistical tools often assume a random distribution of data points, also known as a normal or Gaussian distribution. They also ignore spatial and temporal dependencies that are important when studying temperature, rainfall and other geophysical phenomena that can change daily or monthly, and which often depend on previous measurements. For example, if it is hot today, there’s a higher chance that it will be hot tomorrow because a heat wave is already in place.

Global surface temperatures are similarly linked, and one of the clearest examples of this can be found in the oceans. “The ocean is very deep and can retain heat for a long time,” said Diffenbaugh, who is also a senior fellow at the Woods Institute. “The temperature that we measure on the surface of the ocean is a reflection not just of what’s happening on the surface at that moment, but also the amount of trapped heat beneath the surface, which has been accumulating for years.”

While designing a framework that would take temporal dependencies into account, the Stanford scientists quickly ran into a problem. Those who argue for a hiatus claim that during the 15-year period between 1998 and 2013, global surface temperatures either did not increase at all, or they rose at a much slower rate than in the years before 1998. Statistically, however, this is a hard claim to test because the number of data points for the purported hiatus period is relatively small, and most classical statistical tools require large numbers of data points.

There is a workaround, however. A technique that Romano invented in 1992, called “subsampling,” is useful for discerning whether a variable – be it surface temperature or stock prices – has changed in the short term based on limited amount of data. “In order to study the hiatus, we took the basic idea of subsampling and then adapted it to cope with the small sample size of the alleged hiatus period,” Romano said. “When we compared the results from our technique with those calculated using classical methods, we found that the statistical confidence obtained using our framework is 100 times stronger than what was reported by the NOAA group.”

The Stanford group’s technique also handled temporal dependency in a more sophisticated way than in past studies. For example, the NOAA study accounted for temporal dependency when calculating sea surface temperature changes, but it did so in a relatively simple way, with one temperature point being affected only by the temperature point directly prior to it. “In reality, however, the temperature could be influenced by not just the previous data points, but six or 10 points before,” Rajaratnam said.

Pulling marbles out of a jar

To understand how the Stanford group’s subsampling technique differs from the classical techniques that had been used before, imagine placing 50 colored marbles, each one representing a particular year, into a jar. The marbles range from blue to red, signifying different average global surface temperatures.

“If you wanted to determine the likelihood of getting 15 marbles of a certain color pattern, you could repeatedly pull out 15 marbles at a time, plot their average color on a graph, and see where your original marble arrangement falls in that distribution,” Tsiang said. “This approach is analogous to how many climate scientists had previously approached the hiatus problem.”

In contrast, the new strategy that Rajaratnam, Romano and Tsiang invented is akin to stringing the marbles together before placing them into the jar. “Stringing the marbles together preserves their relationships to one another, and that’s what our subsampling technique does,” Tsiang said. “If you ignore these dependencies, you can alter the strength of your conclusions or even arrive at the opposite conclusion.”

When the team applied their subsampling technique to the temperature data, they found that the rate of increase of global surface temperature did not stall or slow down from 1998 to 2013 in a statistically significant manner. In fact, the rate of change in global surface temperature was not statistically distinguishable between the recent period and other periods earlier in the historical data.

The Stanford scientists say their findings should go a long way toward restoring confidence in the basic science and climate computer models that form the foundation for climate change predictions.

“Global warming is like other noisy systems that fluctuate wildly but still follow a trend,” Diffenbaugh said. “Think of the U.S. stock market: There have been bull markets and bear markets, but overall it has grown a lot over the past century. What is clear from analyzing the long-term data in a rigorous statistical framework is that, even though climate varies from year-to-year and decade-to-decade, global temperature has increased in the long term, and the recent period does not stand out as being abnormal.”

###

Debunking the climate hiatus

Bala Rajaratnam, Joseph Romano, Michael Tsiang, Noah S. Diffenbaugh

Abstract

The reported “hiatus” in the warming of the global climate system during this century has been the subject of intense scientific and public debate, with implications ranging from scientific understanding of the global climate sensitivity to the rate in which greenhouse gas emissions would need to be curbed in order to meet the United Nations global warming target. A number of scientific hypotheses have been put forward to explain the hiatus, including both physical climate processes and data artifacts. However, despite the intense focus on the hiatus in both the scientific and public arenas, rigorous statistical assessment of the uniqueness of the recent temperature time-series within the context of the long-term record has been limited. We apply a rigorous, comprehensive statistical analysis of global temperature data that goes beyond simple linear models to account for temporal dependence and selection effects. We use this framework to test whether the recent period has demonstrated i) a hiatus in the trend in global temperatures, ii) a temperature trend that is statistically distinct from trends prior to the hiatus period, iii) a “stalling” of the global mean temperature, and iv) a change in the distribution of the year-to-year temperature increases. We find compelling evidence that recent claims of a “hiatus” in global warming lack sound scientific basis. Our analysis reveals that there is no hiatus in the increase in the global mean temperature, no statistically significant difference in trends, no stalling of the global mean temperature, and no change in year-to-year temperature increases.

The paper is open access, read it here

Advertisements

275 thoughts on “The latest head in the sand excuse from climate science: the global warming pause 'never happened'

      • Statisticians? We don’t need no stinking classic statisticians!
        Off to the Gausian Gulag with you till you confess your coldness to the sancttity of the God of Warm! Good Grief?

      • This paper is a major breakthrough! It used to be garbage in, garbage out. Now it’s garbage in, polished-garbage out!

      • The study, titled “Debunking the climate hiatus” and published online this week in the journal Climatic Change

        Well with a title that starts with a word like “debunking” it is blattently obvious we are dealing internet trolls and not scientists
        .“Debunking” is argumentative and polemitc, this is an attempt at political point scoring not a scientific study. Do even hope to be taken seriously with a title like that?

      • Debunking is what you do with people talk through the wrong end, see Moon landing hoaxers, or “Moon Landing is an Hoax…”. You debunk fraudulent made-up stories (like the tales of G.E. Séralini who uses “encrypted emails” for fear of Monsanto interference, when Monsanto for years failed to take any action against him beside refuting his crappy antiglyphosate and antiGMOs studies).
        Showing that a scientist is in error is not debunking, it is refuting. Showing that someone who pretends to do science is in fact doing tea leafs reading is a debunking.
        Knowing when to debunk and when to refute is really epistemology 101.
        It may be that current satellite data is not enough to conclude anything about the climate system… anything.

      • The various algorithms of statistical mathematics are thoroughly spelled out in numerous standard text books.
        ” Average ” is obtained by adding all of the members of the data set, and dividing that total by the number of elements in the data set.
        The result is always exact, because all the elements of the data set, are exact real numbers.
        The result (the average) is always correct regardless of the elements in the data set; it works for any finite set of real numbers. It works whether the numbers are unrelated to each other in any way, or whether they are calculated from some closed form mathematical equation.
        The same goes for all the other algorithms of statistical mathematics. they all give a specific result for any finite data set of exact real numbers; and there is no restriction on what those real numbers are.
        Now when I say the numbers of the data set are exact; that is not the same as saying they represent any actual real world value of anything; they are just numbers.
        Where the big mistake is made is in asserting that the results of any of those statistical mathematics algorithms actually mean anything.
        They don’t mean anything except that which they are defined to be in the textbooks.
        So the ” average ” of a data set means just that; it is the average.
        The ” median ” of the same data set is calculated from a different algorithm from ” average ” and usually gives a different result which, is the median of that data set; by definition. And it doesn’t mean ANYTHING else.
        g
        So our new discovery for today is that someone has described a new statistical mathematics algorithm, from those we have all seen before; so it generally gives a different result for a given data set; but it too still means nothing, except that which it has been defined to compute.

      • Lewis, insurance industries are businesses. The aim of a business is to make money and stay in business. The statistics they use attempt to keep the books balanced in their favour. They need be only sufficiently related to actual world events to reliably keep the business in business.
        George is correct in that statistics is only the manipulation of numbers. Re-read George’s text again, he qualifies it quite specifically.

    • “””””…..Statistically, however, this is a hard claim to test because the number of data points for the purported hiatus period is relatively small, and most classical statistical tools require large numbers of data points……”””””
      Well actually you have precisely one data point. The history from circa 1987/8 and 2015.
      We have no way to rerun it to get another data point.
      g

    • I think they are claiming (in effect) that the surface based data sets do not conform to the Nyquist sampling criterion for sampled data systems.
      But we always knew that was so.
      And what is this bunk about the ocean buoys being wrong, and the bucket of water on a ship’s deck being correct ??
      The ocean buoys showed that water temperature and air temperature are different, and are not correlated.
      Well John Christy told us that in Jan 2001.
      g

      • http://www.21stcenturysciencetech.com/articles/ocean.html
        The late Oceanographer Dr Robert Stevenson was a “bucket man” and had this to say when writing a critique of Levitus et al (2000):
        “Surface water samples were taken routinely, however, with buckets from the deck and the ship’s engine-water intake valve. Most of the thermometers were calibrated into 1/4-degrees Fahrenheit. They came from the U.S. Navy.
        Galvanized iron buckets were preferred, mainly because they lasted longer than the wood and canvas. But, they had the disadvantage of cooling quickly in the winds, so that the temperature readings needed to be taken quickly.
        I would guess that any bucket-temperature measurement that was closer to the actual temperature by better than 0.5° was an accident, or a good guess. But then, no one ever knew whether or not it was good or bad. Everyone always considered whatever reading was made to be precise, and they still do today.
        The archived data used by Levitus, and a plethora of other oceanographers, were taken by me, and a whole cadre of students, post-docs, and seagoing technicians around the world. Those of us who obtained the data, are not going to be snowed by the claims of the great precision of “historical data found stored in some musty archives.”

  1. Guess I’ll wait to see what Steve McIntyre and friends have to say about this technique. I’m sure they will digest it in detail. Until then, it’ just another climate paper.

    • Originally the subsampling technique, described here:
      http://home.uchicago.edu/~amshaikh/webfiles/subsampling_topics.pdf
      was for estimating parameters in sets of data that were simply too large to handle, by using samples of the data. In contrast it appears that here, they used this technique to estimate what larger sets of data MIGHT look like, using relatively little data. In other words, they made a huge leap from making inferences about large sets of data from samples, to inferring properties of a hypothetical large data set from a small one.
      This is interesting. From what I can tell, in effect they reversed a statistical sampling technique and used it instead to extrapolate.
      I’m not a statistician, but it would take some strong evidence to convince me that the technique is valid for this purpose.

      • This is interesting. From what I can tell, in effect they reversed a statistical sampling technique and used it instead to extrapolate.
        **********************************************************************************
        What’s all the fuss about, if reversing statistical data is alright for Michael Mann (tiljander) then it’s got to be all right…………oh hang on a minute, let me think this through!
        SteveT

    • Yes that is true, but the problem is that as always, they get to grab the headlines first no matter how much BS it is. We are always playing defence.

  2. “Our new technique basically consists of adding one. To almost everything. We get a much happier answer that way. If we used the old techniques, we kept getting the wrong answers, so it’s obvious that pre-additive statistics doesn’t work right with AGW theory.”

    • The part I really like is that you can’t tell the recent period from any previous historical period. So is not this an argument for “earth is recovering from an ice age and has been generally warming ever since the last period of glaciation some 15,000 years ago” Nothing to see here take down the bunting on the stage an tell all those dignitaries attending Paris in December to stay home!

  3. Once again, rather than using the high tech Satellite data, they use the less reliable and less accurate surface datasets — all of which are heavily adjusted by partisans. Simple question — if you applied the same method to RSS or UAH what would the results be?

    • Why would scientists who allegedly have a warming agenda (motivate) and who are able to adjust the data (opportunity) purposely introduce a pause in warming?

      • An apparent lull in the recent rate of global warming that has been widely accepted as fact is actually an artifact arising from faulty statistical methods, Stanford scientists say.

        What happens to all those papers which cited any of the following papers?
        PS I thought we had a consensus on the standstill. So much for consensus eh.

        Dr. Judith L. Lean – Geophysical Research Letters – 15 Aug 2009
        “…This lack of overall warming is analogous to the period from 2002 to 2008 when decreasing solar irradiance also countered much of the anthropogenic warming…”
        doi:10.1029/2009GL038932
        __________________
        Prof. Shaowu Wang et al – Advances in Climate Change Research – 2010
        Does the Global Warming Pause in the Last Decade: 1999-2008?
        “…The decade of 1999-2008 is still the warmest of the last 30 years, though the global temperature increment is near zero;….The models did not provide answers to the physical causes for warming pause. The mechanism still remains controversial….”
        doi:10.3724/SP.J.1248.2010.00049
        __________________
        Dr. B. G. Hunt – Climate Dynamics – February 2011
        The role of natural climatic variation in perturbing the observed global mean temperature trend
        “Controversy continues to prevail concerning the reality of anthropogenically-induced climatic warming. One of the principal issues is the cause of the hiatus in the current global warming trend.”
        doi:10.1007/s00382-010-0799-x
        __________________
        Dr. Robert K. Kaufmann – PNAS – 2nd June 2011
        “…Given the widely noted increase in the warming effects of rising greenhouse gas concentrations, it has been unclear why global surface temperatures did not rise between 1998 and 2008. We find that this hiatus in warming coincides…”
        doi: 10.1073/pnas.1102467108
        __________________
        Dr. Gerald A. Meehl – Nature Climate Change – 18th September 2011
        “There have been decades, such as 2000–2009, when the observed globally averaged surface-temperature time series shows little increase or even a slightly negative trend1 (a hiatus period)….”
        doi:10.1038/nclimate1229
        __________________
        Met Office Blog – Dave Britton (10:48:21) – 15 October 2012
        “We agree with Mr Rose that there has been only a very small amount of warming in the 21st Century. As stated in our response, this is 0.05 degrees Celsius since 1997 equivalent to 0.03 degrees Celsius per decade.”
        metofficenews.wordpress.com/2012/10/14/met-office-in-the-media-14-october-2012
        __________________
        Dr. James Hansen – NASA GISS – 15 January 2013
        Global Temperature Update Through 2012
        “…The 5-year mean global temperature has been flat for a decade, which we interpret as a combination of natural variability and a slowdown in the growth rate of the net climate forcing…”
        columbia.edu/~jeh1/mailings/2013/20130115_Temperature2012.pdf
        __________________
        Dr. Virginie Guemas – Nature Climate Change – 1 March 2013
        “…Despite a sustained production of anthropogenic greenhouse gases, the Earth’s mean near-surface temperature paused its rise during the 2000–2010 period…”
        doi:10.1038/nclimate1863
        __________________
        Professor Masahiro Watanabe – Geophysical Research Letters – 28 June 2013
        “The weakening of k commonly found in GCMs seems to be an inevitable response of the climate system to global warming, suggesting the recovery from hiatus in coming decades.”
        doi:10.1002/grl.50541
        __________________
        Met Office – July 2013
        The recent pause in global warming, part 3: What are the implications for projections of future warming?
        ….Executive summary
        The recent pause in global surface temperature rise does not materially alter the risks of substantial warming of the Earth by the end of this century.”
        Source: metoffice.gov.uk/media/pdf/3/r/Paper3_Implications_for_projections.pdf
        __________________
        Dr. Yu Kosaka et. al. – Nature – 28 August 2013
        Climate change: The case of the missing heat
        Sixteen years into the mysterious ‘global-warming hiatus’, scientists are piecing together an explanation.
        Recent global-warming hiatus tied to equatorial Pacific surface cooling
        Despite the continued increase in atmospheric greenhouse gas concentrations, the annual-mean global temperature has not risen in the twenty-first century…”
        doi:10.1038/nature12534
        __________________
        Dr. Kevin E. Trenberth – Nature News Feature – 15 January 2014
        Climate change: The case of the missing heat
        Sixteen years into the mysterious ‘global-warming hiatus’, scientists are piecing together an explanation.
        “The 1997 to ’98 El Niño event was a trigger for the changes in the Pacific, and I think that’s very probably the beginning of the hiatus,” says Kevin Trenberth, a climate scientist…
        doi:10.1038/505276a
        __________________
        Dr. Gabriel Vecchi – Nature News Feature – 15 January 2014
        “A few years ago you saw the hiatus, but it could be dismissed because it was well within the noise,” says Gabriel Vecchi, a climate scientist……“Now it’s something to explain.”…..
        doi:10.1038/505276a
        __________________
        Dr. Jana Sillmann et al – IopScience – 18 June 2014
        Observed and simulated temperature extremes during the recent warming hiatus
        “This regional inconsistency between models and observations might be a key to understanding the recent hiatus in global mean temperature warming.”
        doi:10.1088/1748-9326/9/6/064023
        __________________
        Dr. Kevin E. Trenberth et al – Nature Climate Change – 11 July 2014
        Seasonal aspects of the recent pause in surface warming
        Factors involved in the recent pause in the rise of global mean temperatures are examined seasonally. For 1999 to 2012, the hiatus in surface warming is mainly evident in the central and eastern Pacific…….atmospheric circulation anomalies observed globally during the hiatus.
        doi:10.1038/nclimate2341
        __________________
        Dr. Young-Heon Jo et al – American Meteorological Society – 24 October 2014
        Climate signals in the mid to high latitude North Atlantic from altimeter observations
        “…..Furthermore, the low-frequency variability in the SPG relates to the propagation of Atlantic meridional overturning circulation (AMOC) variations from the deep-water formation region to mid-latitudes in the North Atlantic, which might have the implications for recent global surface warming hiatus.”
        http://dx.doi.org/10.1175/JCLI-D-12-00670.1
        __________________
        Dr. Hans Gleisner – Geophysical Research Letters – 28 January 2015
        Recent global warming hiatus dominated by low latitude temperature trends in surface and troposphere data
        Over the last 15 years, global mean surface temperatures exhibit only weak trends…..Omission of successively larger polar regions from the global-mean temperature calculations, in both tropospheric and surface data sets, shows that data gaps at high latitudes can not explain the observed differences between the hiatus and the pre-hiatus period….
        http://dx.doi.org/10.1002/2014GL062596
        __________________
        Dr. Hervé Douville et al – Geophysical Research Letters – 10 February 2015
        The recent global-warming hiatus: What is the role of Pacific variability?
        The observed global mean surface air temperature (GMST) has not risen over the last 15 years, spurring outbreaks of skepticism regarding the nature of global warming and challenging the upper-range transient response of the current-generation global climate models….
        http://dx.doi.org/10.1002/2014GL062775
        __________________
        Dr. Veronica Nieves – Science – 31 July 2015
        Recent hiatus caused by decadal shift in Indo-Pacific heating
        Recent modeling studies have proposed different scenarios to explain the slowdown in surface temperature warming in the most recent decade…..
        http://www.sciencemag.org/content/349/6247/532.short

      • But Jimbo, the 97% consensus is a rather limited one, and does not accommodate every hypothesis in climate science. For instance, one can find paper after paper claiming that climate change is the cause of extreme weather events; but then, you can find an equal number of papers claiming that climate change will moderate weather to such an extent that extreme weather events will disappear.

    • “Once again, rather than using the high tech Satellite data, they use the less reliable and less accurate surface datasets”
      Yes that is a big problem for credibility with this paper, how can any honest scientist totally ignore the Satellite data without at least acknowledging it’s existence and explaining it’s impact or why it is not relevant.
      Absent that the study is just a waste of taxpayer $$$, but what is new.

    • Simple, 1998 was far warmer then any year sense. The “scientists” doing this study appear to think temperature readings before other readings somehow affect current readings. Nonsense, T is what it is, period.
      1998 was far warmer then any year sense.

    • if you applied the same method to RSS or UAH what would the results be?

      Or indeed if they analysed from 1970 instead of 1950 what would their results be? I feel this paper is far from “robust”.

  4. “By blending fake data with massively adjusted, homogenized, and infilled data, no one can say we reached our conclusion first, then invented some new techniques to prove it.”

  5. Funny how everything “faulty or noisy” only happens in the cooling or neutral direction.
    Cut the funding cut the nonsense.

    • If surface temperature starts trending upwards they will say that the pause (that never happened) has now ended. 😉 Heads we win, tails you lose. This is why it’s now known as Climastrology.

      • If surface temperature starts trending upwards they will say that the pause (that never happened) has now ended.
        I suspect you are right. And then they’ll claim that they never claimed there wasn’t a pause. And the media won’t look into things because memory hole and incompetence.

      • I sorta like Climate Astrology. “You will meet someone interesting” is replaced by “you will experience warmer temperatures………someday.”

  6. There are statistics and damn lies!! If you have one foot in boiling water and the other foot in iced water, statistically you should be quite comfortable.

    • The best summation of such methods appeared in the letters page of a newpaper (The National Observer) in 1891
      “Sir, —It has been wittily remarked that there are three kinds of falsehood:
      the first is a ‘fib,’ the second is a downright lie, and the third and most aggravated is statistics. It is on statistics and on the absence of statistics that the advocate relies…”

  7. “Using a novel statistical framework that was developed specifically for studying geophysical processes such as global temperature fluctuations”
    Translation, we kept torturing the data until it eventually told us what we wanted to hear.
    Reminds me of the “novel” statistical tricks used to create the original hockey stick.

  8. They seem to ignore the satellite data.
    They ignore their own buoys and use what, bucket measurements and ship intake measurements from where?
    They still utilize the buggered up data from land based instrument readings with the problems of UHI’s, and cherry picked locations.
    Garbage in, garbage out.

    • They use only what’s useful in backing their forgone conclusions and call it “science”. they need to be prosecuted for their crimes against humanity as well as slander and libel against all honest scientists. They can not truely believe their lies!

  9. They developed a technique that when used on unadjusted date, it showed no pause.
    Then they used it on data that had been adjusted in order to decrease the size of the pause, and once again, it showed no pause.
    And in their minds this proves that their technique must be valid?

  10. This is like them saying we drank one beer on monday, two beers on tuesday, three beers on wednesday, thursday and friday but because our beer consumption was rising earlier in the week we actually drank four beers on thursday and five beers on friday. Let them try getting those expenses through the accounts department without a receipt for those 3 extra beers.

  11. “the Stanford group’s technique does not rely on strong assumptions to work. If one makes strong assumptions and they are not correct, the validity of the conclusion is called into question,” Rajaratnam said.

    My god, the blindness is astounding. The assumptions in his paper are that his “new and improved” method is the indisputably correct method and the relationships between all factors describing temperature at a given point in time are completely proven and understood by him.
    What a dolt. Until Rajaratnams assumptions are proven over time with experimentation and evidence, they remain assumptions making any conclusions from them pending at best.

    • I wonder if there are any actual statisticians in that group.
      Climate science has a long history of using “unique” statistical methods without actually bothering to understand statistics.

    • I loved this quote, because it reminded me of the ‘strong’ assumption of a positive, water vapor feedback that has yet to be found anywhere outside of a theoretical climate model. In an attempt to defend the models, Rajaratnam inadvertently brings up why all of the climate models are crap; the weakness of a strong assumption as the main component of a theory!

  12. I wonder whether their method is able to detect the warming since start of the temperature record, or if the method also thinks that there’s no change in that.

    • That was my thought.
      Particularly the 15 years leading up to Hansen’s testimony to the US Congress.
      If it can’t find that then they have officially debunked AGW as an issue requiring specific actions to be taken.
      Frankly, I’m surprised they didn’t look.

  13. But statistical analysis already had non-gaussian analysis tools. Why did the authors have to use a tool that was made up by one of the authors (Romano)? Twenty years ago, granted, but still one of the author’s own pet tools.

  14. I am no statistician, but something smells very wrong about the technique described. I eagerly await McIntyre’s input.

    • Amen to that TonyG,
      Let’s see what Steve McIntyre at Climate Audit says about this sub-sampling.
      All of these bespoke science results are so transparently timed (and created) to influence discussion before the Paris climate get together !

  15. So, have I got this right? The modern data taken over the last 20 years with modern instruments is all crap (so we adjust it) but the historical data is more accurate and reliable? Really?
    With regard to the whole approach, this is not serious science. Proper science, or indeed mathematics, wouldn’t start with the idea that “we know the solution so let’s adjust the methodology and data until we reach that desired answer”. Proper science would say “let’s look at the hard data and the risk/uncertainty factors inherent in it and see what it’s telling us with any degree of certainty”. Or is that just too simple?

    • An army of world leaders insist on lies so they can tax thin air.
      So any excuse, any tortured data to justify this is good in their eyes but the problem is, will world populations tolerate these immense taxes on nothing? I doubt it seriously.

  16. In the interest of taking this a face value, and in challenging my own assumptions and beliefs, I have some questions. The clearest evidence of the hiatus is the satellite record, which demonstrates some 17 to 18 years without a warming trend. So:
    1) Is the satellite data, UAH & RSS, really some sort of statistical analysis? I assumed each month’s temp anomaly was an average over that whole month, which I guess is technically a statistical tool, but hardly the type of analysis that one could argue is “inappropriate”. Am I missing something here?
    2) Given that around half of the satellite records shows a lack of warming, how is it possible to claim that this is an insufficient quantity of data points? Is there any legitimacy to this claim?
    3) With such a precise measurement system, what type of statistical analysis is really needed? Can’t we just, like, look at the observations and SEE what happened? Am I being naively ignorant here?
    Am I missing something obvious here?
    rip

    • From what I remember of my statistics, the only reason to ‘subsample’ a population, is the impossibility of sampling the entire population. The surface temperature record or the buoy temperature record by design and necessity, is already a subsample. These are subsamples of the population of temperatures everywhere on the planet at any given moment in time. Why, methinks, do they have to apply their super-special subsampling technique to that which is already a subsample? Subsampling implies missing information. So, to me it seems, they intentionally lose information to gain a trend.

    • ripshin September 17, 2015 at 8:18 am says:
      “With such a precise measurement system, what type of statistical analysis is really needed? Can’t we just, like, look at the observations and SEE what happened.”
      I agree, none is needed. I go with Ernest Rutherford who told us that “…If your experiment needs statistics you should have done a better experiment.”
      In our case the better experiment would be using satellites. Ground-based data are corrupted and falsified. Here is an example. In 2008 I was researching satellite data for my book “What Warming.” I accidentally discovered that there had been no warming in the eighties and nineties. It extended from 1979 to 1997, an 18 year stretch, just like the present hiatus. A graph of it is found as figure 15 in my book. But cross checking with ground-based sources I found that they were showing a phony “late twentieth century warming” in its place. That same phony warming is also shown by the Stanford worthies who authored the article. They actually don’t know that there was another hiatus before the current one, nor do they know how it was suppressed. I discovered also that a source for that phony warming was HadCRUT3 and put a warning about it into the preface of the book when it came out. Nothing happened. Later I discovered that that GISS and NCDC had been co-conspirators with HadCRUT in this cover-up. They had all used the same computer to adjust their output and the computer left its footprints in exactly the same places in their publicly available temperature curves. They are still there, since the nineties when the deed was done, They constitute sharp upward spikes that look like noise. Two of them sit directly on top of the super El Nino peak. I have periodically mentioned this but have been entirely ignored. This allegedly scientific organization has no discipline and no ethical guidelines and their so-called climate “scientists” ignore any complaints from the outside.

    • “With such a precise measurement system, what type of statistical analysis is really needed? Can’t we just, like, look at the observations and SEE what happened? Am I being naively ignorant here?”
      I am naively ignorant, so I too can’t see why any fancy statistics are needed. If the temp is measured the same way each time, and the raw figures show a flat line, what more do we need?

  17. So the hiatus as measured by surface thermometers and corroborated by satellite measurements is a figment of bad statistics by those doing the temp records. Well, indeed, the problem with 150yr trend has been the very egregious use of subjective data manipulation as the authors point out, jacking up recent temperatures, shoving down past temperatures and especially submerging the real record period of the 1930s/40s. This of course is to feed the ravens in Paris, but it is going to have unconsidered negative consequences. With the discipline of the satellite record pinning the present temperature levels, they will be forced to flatten the slope of warming period of the 1990s, particularly if they want to even irradicate a slowdown in temperatures. This will require the 1930s to be lifted halfway back by these methods and the IPCC’s lower bound in climate sensitivity to become the ‘best estimate’.
    To imagine what must be done too erase a significant slowdown during a period of rapid CO2 rise, think of a string layed on the temperature trace and fixed at the ‘present’ end. Now make your adjustments. It seems to me a terrible bargain for them to make for this last ditch effort for Paris – remove the urgency, constrain climate sensitivity to an unscary level and push off thermageddon by a century or more. Yes, these are desperate times for climate troughers. I suspect a fair contingent of those with vestiges of scruples will be unable to swallow this latest serving, especially when they see that with the new record, the jig is virtually up. There will be defections. Mark Steyn’s book “Disgrace” outed a fair number of scientists that may now have less to lose in voicing dissent and it will encourage some younger, frightened scientists to step out of line.
    This classic end game stuff – approaching the “Sauve qui peut” stage of a war. (“save himself who can”)

    • Re thermometer measurements: Part 3.1 of the paper concentrates on temperature trends 1998-2013. There was no statistically significant warming, but they did detect a warming trend.
      As of the moment (to August 2015), the trend in both GISS and NOAA/NCDC shows statistically significant warming (0.124 ±0.109 °C/decade (2σ) in GISS; 0.118 ±0.103 °C/decade (2σ) in NOAA).

    • I don’t think it’s a case of measured temperatures being a figment of bad statistics. To me, the critical graph is the top panel of figure 3. In essence, it seems to me that the “new improved” statistics makes the temperature in 1998 about 0.3degC lower than it really was, thus permitting a slope to reappear. I have no idea whether the statistics are correct or not, but you have to admit it’s a neat “trick”!
      Surely this approach must cast some doubt on the whole of the record? Since the authors accuse previous papers of using “naive statistical methods” and state that “only 15 years of data” is not enough for “classical statistics”, I wonder whether they would like to look at the complete record from 1850 onwards instead of just 1998-2013. I also wonder why they chose just that period, since Lord Monckton now puts it back to 1996.
      Finally, I note that the MSM has picked up on this very quickly. It fits perfectly with the agenda and we can expect it to be widely quoted. As long as it’s not disproven before Paris it will have done its job.

      • new and improved or not, 15 or so years is not enough data for any statistical test and is totally naive. The only useful analysis is to look at the actual, original measurements. The only valid conclusion is that the temperature data record is extremely noisy so any statistical model will have error ranges nearly equal to the overall change.

      • “Finally, I note that the MSM has picked up on this very quickly. It fits perfectly with the agenda and we can expect it to be widely quoted. As long as it’s not disproven before Paris it will have done its job.”
        It would seem odd that the MSM would say that new evidence shows that the pause didn’t exist, when they never admitted that there was a pause. It would make it apparent that they failed to give us the whole story.

  18. [snip – fake email address, a valid email address is required to comment here -see result -mod]
    splice@onet.pl – Result: Bad
    MX record about onet.pl exists.
    Connection succeeded to mx.poczta.onet.pl SMTP.
    220-mx.poczta.onet.pl ESMTP
    > HELO technotarget.com
    521 5.5.1 Protocol error
    > MAIL FROM:
    => RCPT TO:

      • They actually started in 1950 to reduce the slope of the increase to 1997. That way it better matches any increases since.
        eg from the paper…
        First, a standard regression of global temperature on time is fitted to both the 1998–2013 hiatus period and the period 1950–1997, with errors assumed to be independently and identically distributed (see Fig. 2 top left panel).

    • Wiser men know that only 15% of the globe has surface temperature data, which means that 85% of the data used to make that graph was made up.

      • Splice September 17, 2015 at 8:51 am
        Nope, it means only, that you know nothing about measuring temperature anomalies.
        ——–
        Did you read the article? They admit that sample size of the surface temperature data is too small.
        Perhaps you can explain to me how you can measure a temperature anomaly in the middle of the Pacific Ocean where there are no weather stations.

    • Of course they cherry pick the start date to the cold 70s not the warm 30-40s. BTW how are the computer models working.

    • Splice,
      1) Explain how 2003 peak was warmer than a strong El Nino in 1997/98 because it wasn’t on any world data sets even back in 2005. Nothing supports it apart from deliberate tampering of data made up by infilling regions with no observations. They can chose whatever they like to warm it up with this method and they have done.
      2) You have cherry picked by far the worst global non-data set there is and has lost all credibility. Only purpose to use it in a science paper is to highlight how awful it is.
      3) Any peak has a rise and flattens at the top. The non data graph describes actually that, so it’s no more that a flat part of the peak.
      4) Statistics can show all sorts of rubbish and the red line only makes it appear like it continues to rise because the leveled out area at the top is warmer than the rising part of the peak numerous years before it.
      5) It shows a pause at the top, not matter how you spin it. The warming rate has significantly decreased to an almost standstill.
      Even the most deliberate tampering of non-data any more towards warming, shows very little warming over the past 13 years.
      http://www.woodfortrees.org/plot/gistemp/from:2002/plot/gistemp/from:2002/trend

  19. Last line of the PR “…even though climate varies from year-to-year and decade-to-decade, global temperature has increased in the long term, and the recent period does not stand out as being abnormal.”
    Aren’t they supposed to be proving the recent period’s warming is abnormal?

    • They made the very warm 1930’s go away so yes, it looks like even with the pause, it is getting warmer and warmer even though this is utterly false, thus the need to eliminate the 1930’s in various ways. Tricky dicky stuff.

  20. They have to have some ‘peer reviewed papers’ denying the pause in press so they can say ‘Recent work has shown that the hiatus was an error’ when they all get to Paris in time for Christmas.

  21. I done this study, like, and, like, taking into account all the then normal levels of community violence and such and disease and accidents from stuff happening and like as well as governments and such attacking their own people and stuff, well sampling the period 1939 to 1945 like, world war 2 never actually happened.
    Amazing isn’t it. Who’d a thunk it but statistics don’t lie.

  22. OK. So, based on the marbles analogy, their subsampling technique works by preserving the relationships of sampled points to each other, and they note that a given temperature point can be affected not just by the previous temperature point, but by the previous six to ten points.
    I await the review of this paper by someone with more statistical expertise, who can tell me if all they did was a variation on a long-term running average, effectively using the prior trend to “smooth” the hiatus out of existence as a short-term fluctuation.

  23. Go to woodfortrees.org and examine any of the global temperature indices – as opposed to hemispheric, land-only or sea-only ones. A consensus of them shows the pause starting in 2001. The satellite-measured ones of the lower troposphere (their UAH one is v.5.5 which is overwarming during the pause period) show the 1998 peak as a distinct El Nino spike, within the late part of the warming period rather than being the beginning of the pause period. Look at only RSS and HadCRUT3 – it looks like the pause started in 2001.

  24. . The Stanford group’s findings are the latest in a growing series of papers to cast doubt on the existence of a hiatus.
    What garbage science.
    The flurry of such flawed scientific studies are the alarmists typical daily out put until the Paris Conference . Expect to get one a day if not more . They are desperate I saw another such a flawed study where they claimed that if we burn all our fossil fuels currently in the ground, we will thaw the Antarctica. They are now doing studies going 10,000 years ahead. Making worst case scenarios to support their current flawed science that may never come about 10000 years ahead speaks to the nonsense that is happening in climate science these days

    • Lets not get caught in the fine points of what constitutes a pause , hiatus or a slowdown. Everyone has their own definition and there will be no consensus on this .The fundamental flaw that remains in the AGW science despite what they now say is that they predicted unprecedented warming until 2100 as co2 levels rise and this has not happened to date and is unlikely to happened in the future . If you monkey with the observable data at will, then all bets are off and the science is a sham.
      According to NOAA data, Annual temperature anomalies since 2005 or last 10 years for combined all GLOBAL LAND areas ( 149 million sq. km) have slight decline or flat trend at – 0.02 C/decade
      .
      • The pause is still real for global land with both land and satellite based measurements.
      o It is clear that GLOBAL as in ‘GLOBAL WARMING” is meaningless as the warming is not global wide as entire continents are actually cooling .
      o The trend of North American annual land temperature anomalies has been steadily cooling whether you go back to 1998,2000 or 2005 at -0.20 C /decade, -0.05 C /decade and -0.41 C /decade respectively according to NOAA
      o The trend of Northern Hemisphere annual land temperature anomalies has been slightly cooling or flat since 2005 at -0.05 C./decade
      o The trend of Southern Hemisphere annual land temperature anomalies has been slightly warming or really flat at + 0.06 C /decade. Africa is also slightly warming or flat at -0.07 C/decade.

    • Here is part of an unsolicited email I received today. This is one of the alarmists strategies for Paris:
      “This year, DeSmog is celebrating our 10th anniversary!
      With a decade of experience, our team is best positioned to investigate and report on the people who want to throw a monkey wrench in the critical climate negotiations happening in Paris this December.
      We need to raise $30,000 to ramp up our independent journalism efforts at the Paris climate summit and beyond. Will you help us clear the PR pollution?
      Even with over 7,000 published articles featuring hard-hitting commentary, in-depth analysis and our constantly expanding research database, it’s hard to believe ten years have passed since DeSmog began.
      DeSmogBlog was launched at the COP 11 global climate summit in Montreal in December 2005. A decade later, here we are on the eve of the critical COP 21 summit in Paris, where world leaders must reach an agreement to safeguard our future from global warming pollution.
      Can you make a generous donation now to help DeSmog raise $30,000 to support our unflinching reporting from Paris?
      Did you know that DeSmog was the first ‘new media’ blog to ever receive full press credentials at a United Nations climate summit? Our unique brand of journalism has shaken up these major events ever since, and we are preparing to hold truth to power again at COP 21 and beyond.
      Climate science denial hasn’t gone away, and we need to keep fighting the fossil fuel industry’s PR pollution. Together we can turn delay and distraction into accountability and action.
      Will you help support us for this important work? Just click here and chip in whatever you can — anything you can give will help make DeSmog an even more powerful and independent voice.
      Fossil fuel companies and the deniers and delayers they fund are hoping that we all sit back and ignore the epic threat of climate change. But their tobacco playbook tactics are no match for DeSmog’s team of independent journalists who clear their PR pollution every day.
      Click here to send your most generous donation and help us fight the fossil fuel industry and science deniers by supporting DeSmog’s 10th anniversary fundraiser.
      Thank you for your ongoing support,
      Brendan DeMelle, Executive Director, DeSmog
      P.S. We rely on individual donors like you. Please make a generous donation now to help us celebrate our 10th anniversary.
      * Donations are not tax deductible (but our hugs of gratitude are free)!”

  25. Without reading the full paper it sounds like they took the straightforward statistics that have been used to show a significant change in temperature trends (the hiatus) and used more rigorous statistical methods. In this case, more rigorous is not always better. For example, if you have normally distributed data you can use either a standard t-test, OR you can use a non-parametric test (rank-order, etc). The t-test might show a highly significant difference between your groups, while the non-parametric test might not show significance.
    This leaves the open question of whether using a different statistical approach is justified or necessary. If it is necessary, they are pretty much arguing that every previous Global Warming study that did not use their statistical approach is invalid.
    When scientists are dealing with data that shows only modest differences or has a low signal-to-noise ratio, but it on the edge of statistical significance they might shop around for a statistical test that gets them “over the finish line”. If they’re doing a t-test, they might use a one-tailed test instead of the standard two-tailed test. All of this is very questionable, and should raise a red flag in the minds of reviewers of these papers. But the fact remains that in science the level of significance is somewhat arbitrary, and some journals give scientists wide latitude in defining what they consider to be significant.
    There are numerous recent highly touted Global Warming studies that claim significance of p < 0.10, which is outside the mainstream.
    My guess is that this same group could carry out a rigorous statistical analysis of many of the foundational statistical findings of the global warming hypothesis and absolutely trash them. But in this case their goal was to obfuscate the statistics underlying the hiatus. They wanted to replace statistical proof with uncertainty, rather than demonstrate the robustness of the warming claims.

    • Ships use seawater internally to cool various ship systems. Seawater intake thermometers used by the US Navy are of 4 kinds, resistance, pneumatic, bimetallic and liquid level. All have a specified accuracy of plus or minus 2F. I understand that merchant ships use the same devices. These thermometers measure the temperature of seawater at the depth of the intake, roughly 20 to 40 feet below the surface for most ships. Their purpose is to provide a measure of seawater temperature so that the proper function of internal ship’s machinery can be assessed. They need be no more accurate than +/- 2F to do that. It seems odd that one would use such primitive devices sampling water at such depths to deduce air temperature changes near the surface of the oceans that are on the order of 0.1 of 0.2 C extending over decades. This makes no sense to me.

      • That’s +/- 2F on the day of manufacture. 10 years later, who knows?
        Beyond that, even if the temperature reported was accurate to .01F, there is still the problem of the depth of sampling varying as the ships ballasting changes. Plus the issue of possible heat contamination from the hull and interior structure of the ship.

      • As an former calibration tech for the US Navy, I can tell you that +/- 2F is the proper accuracy range for most ship borne temperature measuring equipment. The equipment is made to perform in the rigors of working ships, its not laboratory grade equipment or accuracy. The equipment is routinely calibrated to ensure its stated accuracy.
        However, during calibrations, not all equipment is found to be within calibration tolerances. That’s why they are calibrated in the first place. When this occurs an out of tolerance notice is generated and all usage of the items must be analyzed for effect on measurement. I have rarely if ever seen a disposition of any report that said it affected data. ‘Use As Is’ is the normal reply.
        But it gets better than that. When families of measuring equipment show repeated out of tolerance, recommendations for replacement are made. Now we enter the budgetary realm of operations. The ships have to sail and can’t wait for the bean counters. So entire family derations of accuracy are issued for what are affectionally known as dogs. Of course these are all reviewed and approved beforehand, but I seriously doubt that the complete pedigree of ship equipment accuracy is ever included in data use by third party researchers. In 30 years of cal lab work and administration, I’ve never seen such a request or delivered such a report. In practice, these derations can be as much as +/- 10% of full scale and no one blinks an eye. For a 0-200 degree F bimetallic gauge, that’s 20 degrees.

    • The data from the ships matched the output of the models. That’s all that is needed to prove the accuracy of the data.

  26. I LOVE this one.
    I am pretty sure that, using their ad hoc technique, “Our results clearly show that, in terms of the statistics of the long-term global temperature data, there never was a global warming” between 1970 and 2000 !
    🙂
    So indeed , no global warming, no hiatus …

  27. Whoever writes the rebuttal to this will hopefully mention that statistics is a common refuge for scientific ideas under fire — largely because the set of people who are actually actually fluent in statistics is truthfully only a subset of the scientific community. This approach will permit the scientists to make a last stand which most people will simply not understand. However, it offers not hope of gaining additional adherents.
    This approach was wildly successful at undermining Halton Arp, and it was also used to distract people from an apparently temporal anti-correlation between sunspots and solar neutrino production (a clear violation of the Standard Solar Model).
    For Arp, the statistical argument was problematic because it didn’t fully rebut his claims of a direct observation of bridges between objects of wildly different redshifts.
    For the neutrino anti-correlation, which was not supposed to occur because of an enormous amount of time that is hypothesized to exist between these two phenomena, it may turn out to matter that the Sun apparently has different “modes” for the solar wind where the dominant wind can originate from either its closed magnetic field lines or the open “coronal hole” field lines (???).
    Either way, statistical arguments would not be invoked unless the idea was “on the ropes”, to begin with. It’s a sign of the phase we are in with this.

    • This statistical analysis doesn’t change the data. All it does is provide cover for those who want to ignore it or pretend it doesn’t exist.
      You can still draw a negative trend line through the RSS data since 1997, and no global warming scientist or model predicted that.

      • Of course the surface data is consistent with the data models. The surface data was adjusted to match the data models.

      • Fairly consistent? They have historical data up until the dashed vertical line. The predictive capability of the models since then is virtually nil. Within 3 years, observed temperatures dropped outside the 5-95% confidence interval. They bounced back inside for a bit, then spent another ~3 years falling down outside the confidence interval.
        That’s “fairly consistent” with spectacular failure as far as I’m concerned. Once the observations drop out of the confidence interval, the models should be rejected, at least in real science. In made-up fantasy science, I guess anything goes.

  28. Here is the perfect example of statistics gone wild.
    https://wattsupwiththat.files.wordpress.com/2014/03/certaintychannel_ipcc_reality.png?w=720
    As the data steadily diverges from the models, as the model predictions get more and more wrong every year, the “confidence” that CO2 is the global control knob of temperatures goes up and up.
    I’d love to see Romano and the rest of this group apply rigorous statistics to the above relationship and show how absurd it truly is.

  29. It sounds to me like they break the time series into all possible 15 year periods, Yrs 1-15, 2-16, 3-17, etc, and then test for a difference between the current 15 year “pause” and the mean of all prior 15 year periods. I don’t see where they account for the trend. There are enough ups and downs in the temperature record that a any single 15 year period of flat temperatures will be indistinguishable, in a statistical sense, from the mean of all 15 year periods. I am probably oversimplifying their analysis, but I think I have described the central issue.

  30. I caught the “climate change” part of the Republican debate last night and the candidates who answered the question were unanimous in saying they weren’t going to destroy the economy over it. They didn’t even bother dealing with whether or not it exists as a danger. These scientific fr*uds are pissing in the wind anyway.

  31. We are told: This issue of whether global temps are rising is so simple that NO ONE should be questioning it.
    It is so obvious, we are told, that we can just look out the window and see all of these effects of higher global temps.
    Yet, the global temps lined up in a row do not reflect this at all, and taking sub-samples of the limited data then applying advanced statistical techniques, developed in recent months, is necessary to show the warming.
    Now, let’s just try to get an answer:
    is it blatantly obvious, or is it subtle, like how we could live in radon radiation poisoning and never know it, but it is profound-type subtle?
    Well, which is it? And if it is so subtle you cannot see it on the temperature charts, can I be excused for ever doubting you CAGW enthusiasts?

    • +1
      Even the IPPC have acknowledge the ‘pause’; the warmunists have produce over 60 contradictory explanations for it Geeze fighting denial with denial….

  32. I feel that a paper entitled “Debunking the hiatus” is more of a political than scientific statement..
    The problem is that using a block bootstrap (which I am not sure that I believe) is that the estimated increase in temperature is very small. Statistically significant effects may not be physically significant. The important issue is that the “pause” exists with a rate of change that is far below that predicted by virtually every climate model.

  33. Suppose there had been a sharp uptick in temperatures above the model forecasts rather than the hiatus over the past 20 years.
    Choose the likely explanation from the warmists from the following:
    1) This is just an anomaly. The temperature will soon drop down to the values predicted by our models.
    2) It’s worse than we thought!

  34. With regard to ships giving warmer temperatures than buoys: generally lots of oil gets burned on a ship.
    It really wouldn’t surprise me if some of that energy showed up in temperature measurements. I’d guess the reflectivity of the superstructure is probably different from open ocean as well.

    • Water temps taken by ships come from sensors in the engine cooling intakes. If the ship itself is warmer than the surrounding water, it cannot help but warm the water it is sampling. Engine rooms are notoriously hot places.

  35. I love it! Takes me back to the 7th grade and makes me feel young again.
    The Hiatus Debate
    “Is not!” “Is so.”
    “Is not!” “Is so.”
    “Not! Not!” “So! So!”
    “Not! Not! Not!” “So! So! So!”
    “Moron!” “I know you are, but what am I?”
    “I said MORON!”
    “I know you are, but what am I?”
    I believe the study behind this press release takes the “Is Not” side of the debate.

  36. “Using a novel statistical framework that was developed specifically for studying geophysical processes such as global temperature fluctuations…”
    Really? Have they no shame or even sense of irony at long last?
    In any case all their shouting, posturing, and ruining the economy of the West are wasted effort, even were man-caused global warming real, AND actually a problem.
    There are more than six billion people on the planet – one billion between North America, Europe, and Japan. There are two billion between China and India, both growing technologically and neither giving a toss about warming. That leaves another three billion in the developing world, who, while their CO2 contributions are less per capita, in total probably come close to the West in total. So, even if the West’s contribution to CO2 were 50% of the total AND it mattered, it would be a lost cause.

  37. The report states that they combined ship measurements with ocean buoy measurements to arrive at a compromise result. Ok., that’s fine, but I believe we have been using the RSS satellite feeds as the principal, longest running…. stable (read unchanged) form of temperature measurements available to arrive at our proof of the hiatus.

  38. First : None of these sorts of statistical studies touches on physics linking CO2 to these near noise level variations in estimated global temperature . Because no such quantitative theory exists .
    Second : What’s their estimated temperature rise over the century ? Even Karl’s estimate is only about 1.1c , not scary .

  39. Is that a photo of a bicycle rack for climate deniers?
    On BBC the former head of the Nobel committee now regrets awarding Obama a Peace Prize because it didn’t really have any effect. Confirms my suspicion that he wasn’t rewarded for what he did, but what they hoped he would do. Similar in politics to the IPCC climate team award. If we give you this trophy…

  40. Lies, lies and more lies.
    This type of behavior only happens with an agenda driven political government controlled so called non-scientists. Why don’t they just do away with any science and bring witchcraft back? They would argue that black is white if it hit them in the face.
    If the pause that happened being longer than the period when warming happened, also mean the warming period didn’t happen too? Well yes it does.
    Even if global temperatures risen greatly over the next decade, they would still be a period of around 18 years before it that had a pause. The warming temperatures after the 1970’s didn’t remove the cooling period before it, although over the decades they had done their best to try and remove it.
    Lying eyes evidence number 1.
    http://www.woodfortrees.org/plot/hadcrut3gl/from:1997.5/plot/hadcrut4gl/from:1997.5/plot/hadcrut3gl/from:1997.5/trend/plot/hadcrut4gl/from:1997.5/trend
    HADCRUT3 was not showing any warming for many years, so lets tamper with the tamperature until it does for the new HADCRUT4 version.
    Lying eyes evidence number 2.
    http://www.woodfortrees.org/plot/hadsst2gl/from:1997.5/trend/plot/hadsst2gl/from:1997.5/plot/hadsst3gl/from:1997.5/trend/plot/hadcrut3gl/from:1997.5
    The global ocean data set was not showing warming for many years, so lets tamper with the tamperature until it does again for the new HADSST3gl version.
    Lying eyes evidence number 3.
    http://www.woodfortrees.org/plot/rss-land/from:1997.5/plot/rss-land/from:1997.5/trend/plot/uah/from:1997.5/plot/uah/from:1997.5/trend/plot/uah-land/from:2005/trend
    RSS shows no warming and UAH although shows a little warming from far back, also shows a pause over the last decade.
    The GISS is just the biggest load of nonsense any one could wish for. How just after the La Nina finishing in 2001, global temperatures were supposed to have risen more than the strong El Nino in 1997/98 is one of the biggest lies in any not data any more sets in human history.

    • Your comparisons between HadCRUT3 and 4 and HadSST2 and 3 suffer a little from the fact that HadCRUT3 and HadSST2 haven’t been updated since June 2014! They’re obsolete sets. Discontinued. They were replaced by HadCRUT4 and HadSST3, so of course there are differences (otherwise there would have been no need for the upgrades).
      Both updates were accompanied by peer reviewed papers, which anyone is free to comment on via the normal peer review process, should they suspect foul play or incompetence.

      • They haven’t been updated since 2014 because they use the new version instead. It’s the difference in changing data sets that caused the warming, not that the global temperatures actually warmed. The same process was not used for previous decades in the current version so it contains a warm bias for the recent period.
        They are discontinued and illustrates that the change only occurred because global temperatures were not warming.

      • I’m sure HADCRUT 5 will appear as soon as there is any danger of HADCRUT4 flatlining.
        Business as usual.

  41. What a load of rubbish on ships’ sea temperatures. I’ve done it. At 3.30 in the morning an knackered junior engineer taking his log readings. He gets to the main circ inlet and because he would have climb down into the bilge to take a decent reading he leans over the rail and makes a guess. The alcohol thermometer in a brass case is calibrated two degrees per division, the scale is clamped to the thermometer glass and has probably moved, the thermometer pocket might or might not be filled with oil. He memorises the reading as it is a matter of pride that you can fill in some 30 odd readings in the log without writing them down. Not the most scientific process. It was not much better during the daytime.

  42. How can there be a pause in ‘warming’ when all the evidence for warming has been fabricated by ‘cooling’ the past?

  43. Wow. I mean, just… wow.
    I skimmed through it and perhaps I missed it, but I don’t think they actually said what warming rate or sensitivity to CO2 they calculated. In other words (again, unless I missed it) all they’ve done here is a very complicated and questionable data analysis where at they end they pat themselves on the back and say “see….it isn’t zero after all”.
    Well then, what the f*** is it?”
    They even have the audacity to ask what the post 1998 data would look like with 1998…. Yes, let’s just take data that doesn’t help our conclusion out of the picture entirely. OK, so fine, what does the data BEFORE 1998 look like without 1998. Well, it looks pretty flat and when compared to post 1998 with the 1998 spike…. well there’s a slight warming trend, but so close to zero that we could argue that without 1998 the pause goes back to 1987:
    http://www.woodfortrees.org/plot/rss/mean:3/from:1987/to:1997/plot/rss/mean:3/from:1999/to:2015

      • DWR54 September 17, 2015 at 10:11 am
        Your link is to a satellite data set. The paper specifically addresses the surface data:

        Well, that’s a very good point. In addition to dodging the question of how much warming has taken place, they studiously ignore the satellite data. Why do you suppose that is? Because it has broad coverage, granular data, and a sampling methodology that is consistent across the data set? Elements completely lacking in the surface data set?
        While we are at it, could you explain how it is that a warming trend can be present in the instrumental surface record but NOT in the satellite data? Are they measuring two different earths?

  44. Even if you take these results at face value, AGW is disproved.
    There is no catastrophic global warming.
    They are prepared to trade in no global warming for the past 18 years for a few hundredth of a degree global warming.
    If there is a natural background trend in temperature, it needs to be subtracted off before you do any statistical analysis to try and prove AGW.
    BTW when they changed the temperatures on the buoys, that did not change the actual temperature of the ocean – in other words, they did not say that there was anything wrong with the calibration of the thermometers. Hence the ocean temperatures do not honor the data at the ARGO buoys.

  45. It’s odd they didn’t look at the RSS lower troposphere satellite observations, or at the radiosonde balloon observations, both of which show no scientifically significant warming since 1996. In an endeavor fraught with scarcity of data and geographical coverage, you would think they would want to use the most comprehensive continuous sets of observations available, at least as a standard for comparison.

    • Keep in mind statistics only tells us about characteristics of the data(in this case some of the data is a calculation not a measurement).
      Statistical tests tell NOTHING about the next, as yet unmeasured, data point!
      They’d like you to forget that little fact and draw the obvious wrong conclusion.

  46. I’d be willing to bet big money the following translation applies:
    We assumed a long term autocorrelation model which took out the variability, and gave us a steady underlying rate of change such as we were looking to obtain, the only problems being:
    A) no assurance that the assumed autocorrelation is applicable to the process in question
    B) the result obtained has been a steady increase from before the time that CO2 concentration could have initiated it
    C) the underlying rate of change obtained is much smaller than that predicted by the climate models, and is not particularly worrying

  47. You have to love their admission that they use statistical methods “developed specifically for studying geophysical processes such as global temperature fluctuations”.
    I did not know that mathematics and trends analysis methods need to change based on the subject under study. I guess only in “Climate Science”
    It would be worth asking is there is any conceivable set of sample data where this new method shows decreasing temperature trends….. and also… Why the raw data and graphs of it cannot be used directly.

  48. “In fact, the rate of change in global surface temperature was not statistically distinguishable between the recent period and other periods earlier in the historical data.”
    This statement alone confirms that there is no unusual or extra global warming happening since 1998 and the need to control co2 levels is a waste of money . Yet the alarmists are saying the opposite . Just read what the President says or NOAA. These people cannot have it both ways . and be credible

    • I think what they mean by that is that the reported rise in global surface temperature from 1998 to 2013 was indistinguishable, statistically, from that reported between 1951 and 1998. That’s not to say that there was nothing unusual. Quite the opposite, in fact.

      • Climate has varied in the past and can be expected to do so in the future. Mankind has adapted
        to both cool and warm periods, and trade and economic growth over the past 300 years has greatly
        increased our ability to do so. In that context, forecasts of climate are of little value unless they are for a
        strong and persistent trend, and are accurate.
        The IPCC “forecasts” are for a strong and persistent trend, but they have been inaccurate in the
        short term. Moreover, there is no reason to expect them to be accurate in the longer term. The IPCC’s
        forecasting procedures violate all of the relevant Golden Rule of Forecasting guidelines. In particular,
        their procedures are biased to advocate for the hypothesis of dangerous manmade global warming.
        We found that there are no scientific forecasts that support the hypothesis that manmade global
        warming will occur. Instead, the best forecasts of temperatures on Earth for the 21st Century and
        beyond are derived from the hypothesis of persistence. Specifically, we forecast that global average
        temperatures will trend neither up nor down, but will remain within half-a-degree Celsius (one-degree
        Fahrenheit) of the 2013 average.
        This chapter provides good news. There is neither need to worry about climate change, nor
        reason to take action
        http://www.kestencgreen.com/G&A-Skyfall.pdf

      • DWR54,
        That only demonstrates that the ‘adjustment’ method used by GISS has been carefully constructed.
        With freedom to cool the past and warm the present and throw in a few ‘ad hoc’ adjustments there is a lot of scope.

  49. we realized that the assumptions of the classical statistical tools being used were not appropriate and thus could not give reliable answers
    Reliable meaning the answer we want.
    Now, pulling 15 rabbits out of my hat …

  50. “taking temporal dependencies into account” … this makes so little sense I don’t even know where to start … sound a lot like the California Dept. of Education and their finding that self-esteem is related to classroom performance, so we need programs to build self-esteem. It got them a bunch of pumped up punks with weapons demanding their “props” because a teacher told them they were special

    • The author is intentionally conflating the concepts of temperature measurement and temperature as a dependent variable in models. It’s pure slight-of-hand. Whether there is temporal dependence or not, I can’t see how it would have any bearing on the accuracy or precision of real measurements.

    • I was confused by what they actually meant by that as well. I believe this is a roundabout way of saying that the last 15 years of data is not sufficient enough time to say that global warming has stopped. Basically, that the rate of warming from 1998-2013 is likely enough to occur and is not statistically different than the prior ~15 yr periods (using the most biased data set of course).
      This is just more of the same junk science from the cult of AGW. They have rejected the null hypothesis (that there is a hiatus) and erroneously concluded without further analysis that the opposite must be true, that global warming continues unabated. The paper gives meaningless statistics and gives nothing but opinions on the interpretation of those statistics and then masquerades these opinions as scientific conclusions.
      This paper only contributed one insightful detail to the science, “This analysis also highlights the potential for improper statistical assumptions to yield improper scientific conclusions.”

      • RW you are being too kind with “I believe this is a roundabout way of saying that the last 15 years of data is not sufficient enough time to say that global warming has stopped.” If that were the case, the authors would have had no excuse for adjusting the data.
        They are suggesting that temperature measurements are temporally dependent (which is obviously true) but they fail to explain why that would invalidate the measurements or cause them to need adjusting. Simply ridiculous.

      • Well I think we can all agree that temperature and especially globally average temperature are temporally dependent. I prefer to call it hysteretic (adj of hysteresis) as that’s the perfect terminology to describe the climate system, not temporally dependent.
        I don’t think the authors adjusted the data for this analysis because they chose to use the already highly adjusted GIS data. Without being an expert in statistics it was simply my best guess that the statistical methods used somehow take hysteresis into account and is simply a way of saying that the 1998-2013 period is statistically no different that the 50 years prior.
        According CAGW hypothesis it isn’t supposed to be the same, it’s supposed to be getting worse. That’s probably why they used the existence of a hiatus starting in 1998 as the null hypothesis and tested the likeliness that it not real based on how it compares to the 50 years prior. If they had instead (and probably more correctly) they had chosen to use CAGW as the null hypothesis along with its predicted runaway warming — comparing temperature measurements from 1998-2013 to 1950-1998 and testing whether the factually observed slow down in warming was consistent with the null hypothesis — then they would have surely concluded that the null hypothesis was incorrect in that analysis as well.
        But I also think that in this statistical method they simply choose the p-values which are considered significant. Sounds like the perfect statistical method for having the data confess what you want to hear.

      • It seems to me that if “the 1998-2013 period is statistically no different that the 50 years prior” that the inverse would also be true…one could as easily say that since the current 15 year period shows no little or no temp change, that the previous 50 years is no different statistically than little or no change. I think you might have just said that in a different way…

      • My take is that the authors of the article are pulling a “bait and switch.” Under the pretext of examining whether there has been a leveling off of temperatures since 1998, what they really statistically analyze is the different question of whether the temperature data since 1998 is inconsistent with a long term, forced systemic trend. They assume the spatial and temporal stickiness of global annual mean temperatures rather than a Gaussian distribution, and then conclude that the temperature trend from 1998 could still occur within the superimposed long term upward trend. That’s why getting rid of the Gaussian distribution was important, because is makes it more likely that random movement about along-term trend will last longer and therefore minimizing the significance of a 17-18 year flat average.

  51. It looks like lots of words to say the same thing that Tamino said to deny the pause: Until temperatures return to their long term average, it is incorrect to say global warming has paused. I suspect their technique would give the same results if you just froze temperature at 1998 levels for 20 years or 30 years, or perhaps forever as in Tamino’s argument.

  52. Faulty ocean buoys
    “The Stanford group’s findings are the latest in a growing series of papers to cast doubt on the existence of a hiatus. Another study, led by Thomas Karl, the director of the National Centers for Environmental Information of the National Oceanic and Atmospheric Administration (NOAA) and published recently in the journal Science, found that many of the ocean buoys used to measure sea surface temperatures during the past couple of decades gave cooler readings than measurements gathered from ships. The NOAA group suggested that by correcting the buoy measurements, the hiatus signal disappears.”
    Many off the Buoys? Many? NOT all?? Did we not build and deploy this system because it was more accurate the the ship buckets?
    Is NOAA saying that the buoy system was a waste of effort and funds?
    Oh that’s right, the warmists are vacationing on “Fantasy Island” on our dime.
    michael

    • Get away! Not *all* of the buoys gave cooler readings of exactly -0.12C? How suspicious.
      Maybe they just used the *average* discrepancy (do you think?)

      • It’s odd that ARGO floats are accurate to milli-Kelvins but NOAA buoys have a built-in cool bias, and such a large one at that!

  53. Emblazon THIS upon the wall: “A look at the latter revealed that many of the statistical techniques climate scientists were employing were ones developed for other fields such as biology or medicine, and not ideal for studying geophysical processes. “The underlying assumptions of these analyses often weren’t justified,” Rajaratnam said.”
    So it seems the techniques used to establish the upwards trend were not valid to begin with, and so failed when whatever faux “pause” began. The 97% or whatever who believed in the pre-Rajaratnam stats were naive. The skeptical 30% (or so) who suggested the stats were wrong and misapplied, turned out to be, hmm, what’s the best word — “correct”? “Accurate”? “Justified in their skepticism”?
    Of course, it only takes one peer-reviewed paper to overthrown twenty years worth of prior work and set the course for the Paris talks in a few weeks time.

  54. Watching the climate obsessed defend climate science failure after failure is like watching a parody of religious fanatics trying to wish away evidence against a miracle.

  55. Gistemp 1979 – 2001 and 2001-2015 have almost same slope and no hiatus. RSS 1979-2001 has almost same slope as gistemp. This is a problem for climate models because RSS should warm faster than surface.
    RSS 2001-2015 has zero slope while gistemp continues to warm. This is a catastrophe for climate models. Also it shows the hiatus is real.

    • If the surface warms faster than the air aloft, then the atmosphere becomes unstable. Subsequent over turning then moves the excess heat to an altitude where it can be easily dumped to space, resolving the problem.

      • MarkW, thanks for explaining one component of weather.
        It does not resolve the missing troposphere hotspot problem.
        If you think it does, then you do not understand the problem.

  56. Right in-line with rule one of climate ‘science’ which is ‘when the models and reality differ in value ,it is reality which is in error ‘
    And the real joke is these people claim CAGW sceptics are ‘anti-science’

  57. “The finding highlights the importance of using appropriate statistical techniques and should improve confidence in climate model projections.”
    What qualifies these authors, over all other climate scientists, to be the ones to define which statistical techniques are most appropriate for dealing with temperature data? I suspect that the statistical technique that most improved confidence in climate model projections was the one they determined to be the most ‘appropriate’, and it just happened to be a technique invented by one of the authors.

  58. “While the Stanford group also concluded that there has not been a hiatus, one important distinction of their work is that they did so using both the older, uncorrected temperature measurements as well as the newer, corrected measurements from the NOAA group.”
    When they say “older, uncorrected temperature measurements,” are they referring to the ‘raw’ data? Did they use any charts in their paper? I’m surprised none were posted in this article. It would be nice to see a chart comparing the raw data with the corrected data, as well as one comparing the old statistical technique to the new.
    It would also be instructive to see the evidence for their claim that “the statistical confidence obtained using our framework is 100 times stronger than what was reported by the NOAA group.”

    • ‘100 times stronger’ refers to their new tooth paste that is better than anyone else’s tooth paste so hurry and buy your tube before the mobs buy it all out! 🙂

  59. I tried using a “novel statistical framework” to show my bank that I didn’t have an overdraft and that, in fact, they owed me money. Have a guess how impressed they were and what the outcome was.

  60. The boys are nothing if not predictable: if you don;t like the data, go back and change it.
    This won’t fly. The satellite data are the gold standard.

  61. Okay, they are clearly using the NASA GISS Global Surface Air Temperature Anomaly as measured by meteorological stations.
    http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A.gif
    Let’s not delve into the details of this temperature reconstruction for now.
    Their main point is, trend of 1997-2013 is not statistically different from trend of 1950-1997. Let’s accept that as well.
    Now, what’s the rate of warming according to GISTEMP during the 65 years from 1950 to 2014? It is 175 mK/decade. I wonder if applying the same method it can be shown to be statistically different from a zero trend or not…
    But that question is not asked in the paper.

    • Dumb question – were all these meteorology stations around in 1880? Is it just the mean value? Wouldn’t the number of earlier stations represent a less dense sample?

  62. DWR54:
    I write to congratulate you. In the 5 hours this thread has existed you have already contributed to it with 16 posts and every one of your posts is silly.
    That is a remarkable amount and rate of disruptive behaviour from an individual (assuming you are one person). Well done!
    Richard

  63. May they live in interesting times.
    Maybe, just maybe, cooling will be about to start.
    By 2030, it will be very interesting:
    “It’s not cooling, it’s inverse warming.”
    “Global warming has not stopped, it’s just a downward
    adjustment and temperatures are actually trending up.”
    ” Negative trend? There’s no negative trend: it’s a statistical
    artifact of faster warming.”
    “The thermometer is upside down.”
    “The Earth is going into a new ice age.”
    “Build more windflails.”
    “It’s not a minus sign, it’s a dash.”
    “That’s not ice in the Arctic, it’s surfactant froth.”
    “Polar bears? Aren’t they extinct from global warming?”
    “Temperature’s are not falling. You’re using an incorrect
    statistical technique.”
    “Be quiet and pay your carbon tax. Be thankful you’ve
    got one. What would the weather be like without them?”
    “Cooling is what warming does.”

    Yeah.
    Right.

  64. Rubbish!
    Cherry pick, convoluted nonsense using custom tailored statistic runs.

    “…Datasets used in analysis
    The NASA GISTEMP dataset uses the 1951-1980 average as the baseline period and estimates anomalies up to 1200 km from the nearest measurement station, allowing for broad spatial coverage. The NOAA data reconstructs land data for unobserved regions using a method called “empirical orthogonal teleconnections.”
    The HadCRUT4 data does not use any spatial infilling and thus has gaps in grid squares with very sparse (or no) data. The HadCRUT4 data therefore does not account for warming in the Arctic and Antarctic regions,
    leading to documented coverage bias (Cowtan and Way, 2014)…”

    1a) The analysis is not based on temperatures, only on anomalies, without error bars.
    1b) HadCRUT4 does not use infilling so is not considered valid…? Only GISTemp is thoroughly adjusted enough. Again, no error bar ranges for ‘adjusted temperatures’ before anomalies.
    1c) The chosen base period is 1950-1980. Odd that 1950 through 1975 is another ‘hiatus’ period in temperatures.

    “…where xt and ys are the 1950-1997 and 1998-2013 global mean temperature anomalies series respectively, and “t is random noise, … The claim is that the linear trend during the 1998-2013 hiatus period is lower than the trend during the previous period 1950-1997 …”

    2a) Strawman! Insisting that current hiatus statements are based upon a claim that Earth’s current hiatus trend is lower than 1950-1997’s trend is classic straw man smoke and mirrors.
    2b) Comparing trends from land series that undergo massive adjustments allegedly correcting temperature station changes, alterations and moves is a false approach without full defined error bars for all adjustments and changes. Without a clear defined and universally accepted rationale for adjustment, only the original temperature should be used.

    “…Changing the reference period from 1950-1997 to 1880-1997 only strengthens the null hypothesis of no difference between the hiatus period and before. This follows from the fact that the trend during 1880-1997 is more similar to the trend in the hiatus period. Thus the selected period 1950-1997 can be regarded as a lower bound on p-values for tests of difference in slopes. …”

    3a) As the period 1950-1975 is a hiatus period, so too is 1880-1916 a relative hiatus.

    “…Residual plots from a standard least squares fit and corresponding PACF and ACF plots are given below. These clearly illustrate the presence of serial correlation in the global temperature record, and thus the need to properly account for it. …”

    …”we either model the temporal dependence in the global temperature time series explicitly through a parametric autoregressive model, or account for it through the nonparametric circular block bootstrap, stationary block bootstrap, or subsampling…”

    …”The claim that the linear rate of change in global temperature has stalled can be restated as saying there is no linear trend in global temperature during the period 1998-2013. The corresponding statistical hypothesis can be stated as…”

    …”3.1.1 Method IA: No temporal dependence Under the assumption of independently and identically distributed errors, ordinary least squares is used to estimate the slope…”

    “…It is important to recognize that the observed temperature time series are potentially subject to errors due instrumental errors and other reasons. A more sophisticated formulation of the standard regression model could also be formulated. A key assumption that has been made in our analysis in this regard is that the observational errors can be absorbed into the residuals of the regression model…”

    “…using the iterative Cochrane-Orcutt procedure (Cochrane and Orcutt, 1949). A semiparametric block bootstrap is implemented in order to approximate …”

    And it goes on.
    4a) Assume false premise
    4b) Assign assumptions
    4c) Accept and utilize extremely malleable and historically modified data.
    4d) Mandate that infilling data is a requirement
    4e) Skip utilizing actual temperature slope changes over time, instead use a very vague concept of Global Temperature anomaly. Devise multi-step parameters.
    4f) Calculate Ordinary least-squares (OLS) regressions.
    4g) Further calculate OLS autocorrelation function (ACF) and partial autocorrelation function (PACF) in assigning a ‘Temporal dependence’ factors and correcting for them. An odd approach includes the rejection of the NULL hypothesis (simple dependence).

    All of this for simply calculating and comparing the slope for given sections of the temperature record? Something that any qualified meteorologist or student of weather can easily do?
    The more convoluted the posturing, the louder and more insistent are wild claims, the more likely a snake oil salesman is defrauding the people.

    • “…Datasets used in analysis
      The NASA GISTEMP dataset uses the 1951-1980 average as the baseline period and estimates anomalies up to 1200 km from the nearest measurement station, allowing for broad spatial coverage. The NOAA data reconstructs land data for unobserved regions using a method called “empirical orthogonal teleconnections.”
      The HadCRUT4 data does not use any spatial infilling and thus has gaps in grid squares with very sparse (or no) data. The HadCRUT4 data therefore does not account for warming in the Arctic and Antarctic regions,
      leading to documented coverage bias (Cowtan and Way, 2014).”
      1) HADRCUT4 may not account for some warming or cooling in the Arctic 80N+, but the area is so tiny relative to the size of the planet it hardly makes any difference.
      2) If you are so concerned with this, why not use DMI covering 80N+ that uses real observations (balloon, buoys, ship & plane readings etc) that organisations use for helping generate correct weather forecasts.
      3) Infilling data is making up nonsense because there is no way you can know what may be happening there. The weather on it’s own can easily distinguish between huge temperature changes. Infilling from land to ocean surface is the worst technical science rubbish that can be ever done. Temperatures on the water surface change significantly slower to those on land. Might as well stick a tail on a donkey with numbers on it and use that.
      4) Satellite data covers far more of Antarctica than GISS ever will and it shows no warming.
      Therefore the claim that Antarctica is warming when it is not with real observations, shows that infilling the grids with no observation has cause this result and difference. GISS shows far more warming than any other global data set because of infilling over Antarctica and Arctic coverage. This is despite satellites especially UAH having far more coverage of these regions than GISS does. The infilling also seems to be deliberately extreme on just odd occasions to get those record temperature peaks that none of the others do.
      HACRUT4 has it’s faults and is doing some of the GISS tricks, but it is far better observation data set than a made up one from GISS.

  65. Same old nonsensical argument. 15 years of rising temperatures is a catastrophic trend that requires immediate, drastic action from mankind. 15 years of without rising is statistically insignificant.
    The problem is not enough time has elapsed since the 1998 to 2013 period has passed, they only incrementally adjust the temperatures upward about once a year. In another 10 years, with 10 more years of adjusting upwards, the 1998 to 2013 period will show a significant rise. Remember in 2000 when James Hansen summarized the 20th century temperature trend as having no significant trend up or down? 10 years later, slower than the speed of grass growing but growing none the less, James Hansen’s temperature trends for the 80s and 90s grew into enough of an upward trend to make himself and many colleagues rich from telling us about a dangerous temperature trend that emerged from the data years and years after the measurements were taken.

  66. So, I’m still trying to find the proof that the warming since 188x is statistically significant, given the presence of AMO, PDO, and autocorrelation. Where is this proof? I can’t find it despite endless Google searches.
    When I run my own Monte Carlo analysis of trends on autocorrelated noise that has the same spectrum as GISS or Hadcrut4, I get a very tiny amount of warming above the 95% confidence interval only for the most adjusted temperature set – GISS, and lower than 95% confidence for Hacrut4.
    By the standards I”m using, the hiatus is also statistically insignificant. So I agree with the authors of the paper cited in this article. But so is the entire AGW idea…or at least the coefficient of C02 log2 function is ridiculously small….
    https://www.dropbox.com/s/9jt9l2nldo9sijg/gisstrendornoise.jpg?dl=0
    https://www.dropbox.com/s/9hjgka4r389j0h5/hadcrut44trendornoise.jpg?dl=0
    Peter
    Method: Generate noise of equivalent RMS and spectrum of length 8x that of the record in question. Run Monte Carlo simulation to find band of 95% of trends of the length of the record in question.
    Source: I also note this is rough draft work: https://www.dropbox.com/sh/6eweroc97i0dlk9/AAAhPRyxAb2XtJQp2MOpwqEEa?dl=0

    • I looked at the GISS dat using the n equivalent method Willis discused a while back for time series data. I had to adjust the equations for slope error a bit to get it to match typical errors generated with random time series data. So I am not sure the numbers are correct, but for the whole GISS record the slope was not significantly different from zero at alpha = 0.05. But for the first and second half of the data I did get slightly significant trends different from zero.

  67. If you’re going to try to say someone else is using “faulty statistics” you should be using “tried and true” statistics not “…a novel statistical framework…”

  68. I always find a statement that the confidence is 100 times more as a good reflection on how we can rely on such information. Why not state that they are 99 times more confident or 101times . It’s all so meaningless and the fact that any one could take such fiction seriously is a bit disturbing.

      • Gloria Swansong September 17, 2015 at 1:49 pm
        Don’t dictators always get 97% of the vote. Or is it 99%?
        —————————
        97% of the dictators get 99% of the vote with a confidence level of 95%.

      • Gloria Swansong: “Don’t dictators always get 97% of the vote. Or is it 99%?”
        Some dictators have managed to get >100%.
        A bit like the climate “scientists” who reckon CO2 is responsible for >100% of the increase in the Earth’s temperature…

  69. sophocles,
    You have the right screen name. You wrote:
    By 2030, it will be very interesting:
    “It’s not cooling, it’s inverse warming.”
    “Global warming has not stopped, it’s just a downward
    adjustment and temperatures are actually trending up.”
    ” Negative trend? There’s no negative trend: it’s a statistical
    artifact of faster warming.”
    “The thermometer is upside down.”
    “The Earth is going into a new ice age.”
    “Build more windflails.”
    “It’s not a minus sign, it’s a dash.”
    “That’s not ice in the Arctic, it’s surfactant froth.”
    “Polar bears? Aren’t they extinct from global warming?”
    “Temperature’s are not falling. You’re using an incorrect
    statistical technique.”
    “Be quiet and pay your carbon tax. Be thankful you’ve
    got one. What would the weather be like without them?”
    “Cooling is what warming does.”

    I’ll add another ‘reason’: “Cooling is just a warming reciprocal”: one over Warming.
    And my fav: “Climate change is happening!

  70. The statistical method appears to assume that data points that are associated in time will be about the same.
    By saying the pause is short term and a data point in the context of climate change, it assumes that the data in the pause is associated with the warming that occurred before.
    So the pause can be seen as part of the hypothesised picture that the pause is an artifact that when looked at the broader context is part of warming.
    So the pause can be ignored as this statistical method shows.
    If my understanding is correct, this is a circular argument, as the conclusion is driven by the assumption that data points will be the same.
    It is used to disprove the Null Hypothesis, where it was postulated that where there is no difference between the global atmospheric surface temperature over 18 years then additional atmospheric CO2 was not a significant driver of temperature.
    The CO2 hypothesis was used as a call to stop CO2 production as temperature would rise dangerously.
    Some thoughts.
    If the hypothesis that temperature does not vary much because data pints are linked then there is no danger of a rapid rise in temperature, because there has not been any.
    So this paper, if believed, predicts slow temperature rise.
    Its weakness is that it uses, as pointed out above,small data sets to illustrate large data sets which are available.
    A statistical comment would be good.
    RagDuke comes to mind.

  71. I read the article by Rajaratnam et al. in “Climatic Research” and wrote a comment but discovered that they do not want comments. I am including the comment I wrote here. The Article is called
    “Debunking the climate hiatus” comment follows.
    From the abstract we read:
    “….We apply a rigorous, comprehensive statistical analysis of global temperature data that goes beyond simple linear models to account for temporal dependence and selection effects.
    …Our analysis reveals that there is no hiatus in the increase in the global mean temperature,
    …We find compelling evidence that recent claims of a “hiatus” in global warming lack sound scientific basis.”
    On the contrary, compelling evidence is that this paper lacks sound scientific basis. To start with, lets take the first sentence of the abstract above. Ernest Rutherford, of whom you undoubtedly learned in school, assesses it this way: “If your experiment needs statistics you should have done a better experiment.”
    Next, when you say “Our analysis reveals…” you are not correctly analyzing your observables. The section of global temperature included as part of the hiatus usually begins with the 1997 part of the super El Nino of 1998. It got started there when the observers were interested in including the maximum number of years into the observed hiatus and the Super El Nino fortuitously added to this. But this is incorrect because the super El Nino has a completely different origin from the rest of the data surrounding it.The correct start for counting the years of the present hiatus should be the year 2002, the ending date of the step warming of 1999. That step warming followed closely on the heels of the super El Nino of 1998 and is the only real warming since 1979. In three years it raised global mean temperature by a third of a degree Celsius. For comparison, the entire temperature rise for the twentieth century was only 0.8 degrees. This quick temperature rise should have attracted attention and it did attract Hansen’s attention. He noted that all of the first decade of the twenty-first century was warmer than the twentieth century was, except for one (1998), and declared this wonderful gift to be greenhouse warming. Unfortunately he had greenhouse warming on his brain and did not know how wrong he was. You can’t start greenhouse warming without injecting carbon dioxide into the atmosphere and we know this did not happen in 1999. And you can’t stop it without plucking all the carbon dioxide molecules out of the air which obviously did not happen in 2002. The correct hiatus temperature graph using NOAA’s ERSST v.4, startimg in 2002, does not have an upward slope and is a horizontal straight line. But it there are still no regular ENSO oscillation there. For the first seven years of the century there was no ENSO but in 2008 suddenly a La Nina cooling appeared. Probably the one Trenberth was cursing in Climategate emails. It is followed gy an El Nino warm peak in 2010 which gave us hope that ENSO had returned. I speak of its return because the super El Nino, not part of ENSO, had interrupted an ENSO wave train that was active in the eighties and nineties. Such super warm peaks are rare and usually happen on centennial time scales. It looks now like the temperature fluctuations following the 2010 El Nino have returned to the same doldrums we had in the first decade of this century. This does not promise much warming ahead. The prognosticators are putting their hopes on the return of another super El Nino but there is no chance of that and a regular El Nino is the best they can get, followed of course by a La Nina to balance it, . And now lets look at some technical data. The so-called “pause-buster” temperature data-set from NOAA is called ERSSTv.4. If you plot a linear curve with it starting in 1997 as most hiatus curves have done you do get slightly more warming than its version 3 showed. If you start in 2002 you get no warming as I mentioned. The reason for choosing 2002 is because that is the year when the construction of the hiatus platform by the step warming starting in 1999 was complete. If I understand correctly, the entire purpose of this paper is to prove the non-existence pf this hiatus. How would it strike you guys if I told you that this is not the first but the second hiatus you people have attacked? I bet you did not know that the first of these hiatuses existed in the the eighties and nineties. You are ignorant of it because it was successfully covered up by a phony warming called “late twentieth century warming.” I discovered it accidentally in satellite records while doing research for my book “What Warming?” in 2008. It turned out that there was no warming at all from 1979 to 1997. Comparing this to ground-based temperature curves I found that this temperature section was covered up by a phony warming. I also discovered that this phony warming had originated with HadCRUT3 and put a warning about it into the preface of the book when it came out. Nothing happened. I next discovered that GISS and NCDC had collaborated with HadCRUT in this conspiracy because of identical computer footprints in their data. These are high spikes near the ends of years. Two of them sit right on top of the super El Nino of 1998, easily recognizable by comparison with satellite curves. Luckily they still don’t control the satellites or we would never have known about any of this. If you seriously want to prove that there is no hiatus you have to make your proof include both hiatuses, nit just one. As to calibrating these hiatus data, luckily ENSO was active during the eighties and nineties and produced a wave train of five El Ninos with La Nina valleys in between in the middle of the hiatus. In a situation like this the global mean temperature is at the center point of a line connecting an El Nino peak with its neighboring La Nina valley. I put dots at these points wherever possible in this wave train and discovered that the dots formed a horizontal straight line 18 years long. Eighteen years of no warming, just like today. You will find its graph as figure 15 in my book. This calibrates the hiatus that is now covered up by fake warming. You can do it yourself by downloading the data from satellites. If you showed the true temperature instead of the fake one in official temperature curves you would have an eighteen year horizontal step in that smoothly rising temperature curve they foist upon us now. As to your last claim that ” ..We find compelling evidence that recent claims of a “hiatus” in global warming lack sound scientific basis.” — it is nothing more than just another pseudo-scientific boast intended to falsify the temperature record.

  72. “Stringing the marbles together preserves their relationships to one another, and that’s what our subsampling technique does,” Tsiang said. “If you ignore these dependencies, you can alter the strength of your conclusions or even arrive at the opposite conclusion.”

    No great assumptions they say? It seems to me they’re making the assumption that local effects extrapolate to global effects and that’s a fundamental and huge assumption.

  73. It somehow pictures the state of cimate science. Even with the homogenized temperature records, they have to use extremely advanced statistical methods to find what they want.
    If it was so visible, that they claim it is, why do you need all that statistics to see it.
    Without climate science, nobody would have noticed the globe getting warmer.

  74. I am curious how the cult of CAGW will respond to CGC.
    The corollary of the observational fact (and roughly a couple of dozen other observations and analysis results) that there has been no warming for 18 years, is the majority of the warming in the last 150 years was due to solar cycle changes, rather than the increase in atmospheric CO2. If that is true global warming is reversible.
    The solar observations (shrinking sunspot size, shorter sunspot lifetime, decreasing solar 2.8 Mhz flux, sudden drop in sunspot number) are changing quarter by quarter, continuing to support the assertion the sun cycle has been interrupted rather than is just slowing down.
    What has held back the cooling are solar wind bursts from persistent coronal holes and a complex mechanism that is related to how solar cycle changes cause large cyclic geomagnetic field changes.
    The solar wind bursts create a charge differential which in turn causes there to a movement of electrical charge from high latitude regions of the planet and the equator which cause warming.
    The coronal holes are now starting to move to high latitude regions of the sun where they no longer affect the earth and/or are starting to dissipate.
    If I understand the mechanisms, when the coronal holes are no longer producing solar wind bursts, there will be significant cooling, say 0.5C over a few years. We are going to experience the cooling phase of a Dansgaard-Oeschger cycle.
    .
    http://sait.oat.ts.astro.it/MmSAI/76/PDF/969.pdf

    Once again about global warming and solar activity
    Solar activity, together with human activity, is considered a possible factor for the global warming observed in the last century. However, in the last decades solar activity has remained more or less constant while surface air temperature has continued to increase, which is interpreted as an evidence that in this period human activity is the main factor for global warming. We show that the index commonly used for quantifying long-term changes in solar activity, the sunspot number, accounts for only one part of solar activity and using this index leads to the underestimation of the role of solar activity in the global warming in the recent decades. A more suitable index is the geomagnetic activity which reflects all solar activity, and it is highly correlated to global temperature variations in the whole period for which we have data.
    In Figure 6 the long-term variations in global temperature are compared to the long-term variations in geomagnetic activity as expressed by the ak-index (Nevanlinna and Kataja 2003). The correlation between the two quantities is 0.85 with p<0.01 for the whole period studied. It could therefore be concluded that both the decreasing correlation between sunspot number and geomagnetic activity, and the deviation of the global temperature long-term trend from solar activity as expressed by sunspot index are due to the increased number of high-speed streams of solar wind on the declining phase and in the minimum of sunspot cycle in the last decades.
    We will now compare the properties and geo effectiveness of the two types of solar drivers – High Speed Streams (HSSs) from coronal holes, and CMEs, additionally dividing the CMEs into two types – MCs and
    non-MC CMEs (which we will further denote as simply CMEs). Our study covers 11 years, from 1992 to 2002. In this period we have 92 MCs (Georgieva et al. 2005) and 128 CMEs from the list of Cane and Richardson (2003) from which all events identified as MCs have been removed and 126 CHs identified in the OMNI database (http://nssdc.gsfc.nasa.gov/omniweb).
    Figure 2 presents a comparison of the mean solar wind speed for the three types of solar drivers while
    Figure 3 shows the solar cycle variation of their speed. Figure 2 demonstrates that the speed of the solar wind originating from CHs is much higher than of the solar wind associated with CMEs and MCs. The yearly averaged speed of solar wind from CHs and MCs are comparable around sunspot maximum, and higher than
    the speed of CMEs, and everywhere outside sunspot maximum the fastest solar wind originates from CHs (Figure 3). Similarly, the average geo effectiveness of solar wind from CHs is highest outside sunspot maximum (Figure 4) while around sunspot maximum the most geo effective solar driver are MCs ….
    Therefore, when speaking about the influence of solar activity on the Earth, we cannot neglect the contribution of the solar wind originating from coronal holes. However, these open magnetic field regions are not connected in any way to sunspots, so their contribution is totally neglected when we use the sunspot number as a measure of solar activity. (William: Note to Leif. Totally neglected when we use the sunspot number as a measure of solar activity and we are hence missing a major mechanism as to how solar cycle changes effect planetary climate.)

  75. “we found that the statistical confidence obtained using our framework is 100 times stronger than what was reported by the NOAA group”
    Statistical confidence? Do you really need statistics to plot measured data over a timeline?
    Since the “subsamples” are not measured data, how can they claim these subsamples reflect actual temperatures?

    “In reality, however, the temperature could be influenced by not just the previous data points, but six or 10 points before,” Which on is it, six or ten? How would they know how they were “influenced”, given the subsamples are imaginary data?

  76. Statistics are generally used in climate science for hiding inconveniences and this topic is a good example, plus many others and SM/Willis has made a good job of exposing their worth in the past. Smoothing of data can be useful, but it should never be instead of the raw data as it can hide secrets.
    The reason for the pause and why in the near future the pause may be claimed to have finished.
    http://i772.photobucket.com/albums/yy8/SciMattG/RSS%20Global_v1997-01removal_zpszk83g0xi.png
    When we get a strong El Nino to show it’s effects in global temperatures in about 6 months time. Whether it will do a step up like it did previously is unknown, but maybe the signs are not this time with a lack of response so far with the satellites?
    http://www.woodfortrees.org/plot/rss-land/from:2015/plot/uah-land/from:2015

  77. What I love about roof-top solar is the 20-year payback. You have 20 year life expectancy solar panels sitting on a 15-year shingled roof that has to be replaced every 12-15 years. What could go wrong? And the installation of solar panels happens 5 years after the shingles are laid.

  78. I suspect that if you used 1000 year periods instead of 15 years and used the Greenland ice core data you could easily find that the planet is still cooling.
    In fact, I would also think that using 10 year periods might give a different answer as well. So, we can see right away they are cherry picking to get the answer they want. Not very scientific.

    • I also suspect one could apply this technique to a sine wave and claim the line is still going up even after it has started to go down.

  79. The … variety … of contradictory conclusions arising from such a “settled science” is enough to make one’s head spin.
    Perhaps tunnel vision focusing only on CO2 keeps them from getting dizzy?

  80. Oh, snap! Silly me, I thought we launched those buoys because the buckets dropped over the side of ships were sporadic and inaccurate. Now NOAA explains those pesky — and expensive! — buoys are somehow LESS accurate than readings [taken] by Shanghaid seafarers with buckets, rope and thermometers.
    Or is NOAA just pulling our legs again?

  81. “The finding highlights the importance of using appropriate statistical techniques…”
    OMFG!!! Seriously????

  82. Maybe the witch doctors (climate scientists) from Stamford could apply this same “precise level of statistics” to determine if the correlation coefficient between global temperature and CO2 is better than +/-0.3.
    “Precise level of statistics” – that’s a doozy. Isn’t statistics all about precision?

  83. A few things said in the article that struck me.
    !) …”many of the ocean buoys used to measure sea surface temperatures during the past couple of decades gave cooler readings than the measurements gathered from ships….by correcting the buoy measurements the hiatus signal disappears.” (This is short for what Karl did in his paper.)
    This follows the grand climate science tradition of using bad data to correct good data. The higher tainted temperatures of urban areas (caused by the Urban Heat Island effect) are routinely used to raise the pristine data obtained from rural areas. Much more could be said but why bother.
    2) “if one makes strong assumptions and they are not correct, the validity of the conclusion is called into question, Rajaratnam said.” (Rajaratnam claims he does not make such an assumption.)
    “For example if it is hot today, there’s a higher chance that it will be hot tomorrow because a heat wave is already in place.”
    What is Rajaratnam strong assumption? CO2 has warmed the earth in the past and since that CO2 is already in place (and even continuing to increase!) it is certain that it will be still hotter tomorrow. (The data must be wrong if it says otherwise.)
    Hmmm…I can state with 100% confidence that all heatwaves in the past have ended. Rajaratnam assumption is that his heatwave caused by CO2 not only never ends but gets continuously hotter. We do know that CO2’s small effectiveness for increasing global temperatures decrease as more CO2 is added to the atmosphere. Eventually it loses all practical new effect, if it has ever had any (this is ignoring all other mitigating effects which have a huge negative effect of their own). Rajaratnam’s heatwave also must end.
    I ignore the fact that we are coming out of the Little Ice Age and (though we can’t pinpoint exactly why) temperatures have been steadily rising. Of course, this will eventually stop also. In the long run warming comes and goes.
    3) [Previously] “if you wanted to determine the liklehood of getting 15 marbles of a certain color pattern you could repeatedly pull out 15 marbles at a time [out of a jar] and plot the average color on a graph” …etc.
    “In contrast the new strategy that Rajaratnom, Romano and Tsiang invented is akin to stringing the marbles together before placing them into the jar……. if you ignore these dependencies you can alter the strength of your conclusion and even arrive at the opposite conclusion.”
    Hmmm…To start let me just change one word of that last bit. — If you “create” these dependencies you can alter the strength of your conclusion and even arrive at the opposite conclusion.
    The claim is a “new” factor has to be taken into account. It is actually an old factor, the old claim that Increasing CO2 in the atmosphere means there must be increasing warming. It should be obvious to anyone that this new factor can only turn a flat trend into a rising trend. It is the equivalent of just steadily increasing the data numbers along a timeline. Tomorrow will be warmer, the day after that warmer still, etc.
    Beneath a “smokescreen of statistics” that is all that is going on here.
    Did they learn the concept from Gavin Schmidt? He assumes that since CO2 is constantly increasing the temperature of the earth in the past must have been cooler than the present so that temperature data has to be adjusted down — and from there all temperature data must show an upward trend — so raise the recent temperature data to show the upward trend.
    Eugene WR Gallun

    • I was confused by the rate of rise that was indicated by the graph, and then I noticed that they had used the trick that’s becoming all to common. Scientific information being presented in Fahrenheit rather than Celsius so the numbers look more alarming.
      These guys aren’t scientists, they’re marketing gurus.

  84. At last, I’ve found all the uninformed, psuedo-scientists all in one convenient site, feeding off of each other’s ignorance At least your quasi-religious anti-science beliefs are confined safely here and don’t threaten the general public
    [But SkS does threaten the public. 8<) .mod]

  85. Wait, so the older methods, while indicating warming trends, were correct, unquestionable, settled science, 97% consensus and all that. Now that they indicate no trend (and have not for over 18 years) they are obviously flawed and a new method must be created. Apparently the old way was somewhat shy of settled.
    What’s the over/under on how many statistical black boxes they had to try before they devised one that gave them their trend back? Oh yeah, this is absolutely unassailable. I see a Nobel here.

  86. According to Stanford:
    “Using a novel statistical framework that was developed specifically for studying geophysical processes such as global temperature fluctuations, Rajaratnam and his team of Stanford collaborators have shown that the hiatus never happened.”
    “Faulty ocean buoys
    …Another study, led by Thomas Karl, the director of the National Centers for Environmental Information of the National Oceanic and Atmospheric Administration (NOAA) and published recently in the journal Science, found that many of the ocean buoys used to measure sea surface temperatures during the past couple of decades gave cooler readings than measurements gathered from ships. The NOAA group suggested that by correcting the buoy measurements, the hiatus signal disappears.”
    So the pause can be made to disappear of one applies a “novel statistical framework”. Fancy that. And temperature samples taken from canvass buckets and engine room intakes are more accurate than the output from the finest temperature-measuring buoys money can buy. Who would have thought? And the mainstream press trumpets such claims worldwide without the least iota of incredulity. How absolutely unexpected.

  87. I take issue with the many commentators here who derisively dismiss the value of statistics. Stats is one of our most powerful armaments in the struggle to gain knowledge. It works. The fact that so many misuse it and get bogus, stupid, or dishonest results is just another symptom of our intellectual decline into a new dark age. But don’t dismiss statistics. So let’s use it right for once.
    The claim as I understand it is that the current flat line is simply an unusual statistical sample. If we had put our buoys in different places, or shifted a temp station or whatever, we would in all probability have seen a steadily rising line in accord with model expectations. Right? Because if not, then there really IS a pause.
    Well if that is true, then the next data point to be released is extremely likely to be very much higher than even the 1998 datapoint. And again for the one after, and so on. So after just a single additional datapoint, unless it rises hugely, right back on track with the climate models (not simply “hottest evaahhh”, but really, seriously hot – then their claim that this is a statistical aberration is exploded. And after two additional datapoints, write them off completely.
    But of course, if it really is “that hot” – how come we haven’t noticed? Maybe global warming isn’t all that bad after all.

    • Things might/would be different if ‘climate scientists’ ever used real statisticians. That was what I was taught to do. But, then the game would be up.

    • “…derisively dismiss the value of statistics. Stats is one of our most powerful armaments in the struggle to gain knowledge. It works…”

      Statistics is a tool, only a tool.
      Like any other tool, it is a poor workman that blames or credits their tool for the quality of work.
      It is people like Karl, Rajaratnam, Joseph Romano and grad student Michael Tsiang who utilize statistics to mask and disguise temperature trends that are obvious to any lay person with intelligence, eyesight or touch.
      The rot in climate science sullies the tool called statistics; let us hope that science and statistics are not dealt fatal blows by the climate team’s inappropriate use of statistics.
      “Who are you going to believe? Me or your lying eyes?”

  88. I am a member of a minority ( I am left handed) I am offended by this paper so therefore you must take it back!

  89. One can say it isnt really slowing down only if one first admits that it wasn’t speeding up in the first place.
    That is, if the ups and downs are just blips, but since they didnt accept this in the first place, they cant turn around and say it hasn’t slowed down or paused.

  90. So, IPCC’s AR5, which acknowledged in 2013 that there was no “statistically significant warming” 1998-2012 was wrong? That admission got the stamp of approval from all the high muckety-mucks and grand poobahs of the climate-government establishment, and it was wrong? All the scads of billions of dollars that was spent to get to AR5 got us the wrong answer?

  91. From the paper

    These two tables also analyze the sensitivity of the results to two important factors: first when the cut-off year is changed from 1998 to either 1999 or 2000; and second when the NOAA or HadCRUT4 datasets are used instead of the NASA-GISS dataset. As there are four hypotheses being tested, using a battery of rigorous test procedures, the number of hypothesis being tested are numerous. Hence the issue of multiple hypothesis testing surfaces. In particular, a certain number of these hypotheses are expected to be falsely rejected by chance alone, casting further doubt on any of the hiatus claims.

    This is hilarious. And deeply disturbing all at the same time.
    They didn’t alter their start date from 1950 to say 1970 which I can see would have altered their result a lot. They didn’t use RSS or UAH which is also likely to make a big difference. They used the phrase “battery of rigorous test procedures” which even on the surface is well overstated.
    I doubt McIntyre would bother with this paper. Willis will rip it to shreds anyway.

  92. The climate has no problems at all figuring out what it’s going to do — it is totally obvious to the climate. We are the ones having the problem forecasting what the climate is going to do.

  93. The abstract says: “Our analysis reveals that there is no hiatus in the increase in the global mean temperature, no statistically significant difference in trends, no stalling of the global mean temperature, and no change in year-to-year temperature increases.”
    So if the zero-trend “pause” is not statistically significant different from the warming trend, does it not follow that the warming trend is not statistically significant different from zero?
    The second-to-last paragraph in the paper concludes: “our results instead demonstrate that the evolution of global surface temperature over the past 1–2 decades is not abnormal or unexpected within the context of the long-term record of variability and change.”
    Exactly! If there’s little or no warming, then of course a “pause” is not abnormal!
    Hey, you could use this paper to prove that there is no warming – just change a few sentences in the conclusion.

  94. Since we were sold expensive renewable energy on the basis of the science was beyond question surely it is well overdue of a purge on climate science fraud on a scale at least equal to the mis-selling scandal of the banking and insurance industries.
    I estimate that my losses thanks to the banks is at most a third of my renewable energy subsidy costs.

  95. “the latest in a growing series of papers to cast doubt on the existence of a hiatus.”
    As if truth is subject to a paper count.

    • “science is not a democracy” … “with 97% of (valid) ballots (counted by me and my friends) we WIN THE FACT ELECTION!” (no recount needed, we promise)

  96. When the final denouement of the AGW farce has come, the authors will regret they ever wrote something so stupidly political correct. It will cost them their job, it ought to.

  97. I have just had a wonderful day at Harwell as part of a Cambridge University alumni guided visit. Golly, the scientists are impressive – professional, articulate, open-faced, knowledgeable, cutting edge. So different from the flaky ‘climate science’ lot. Programmes at Harwell are stunning – hi power laser, materials science, particle accelerators, medical programmes at molecular level, sun science, satellite instrumentation…… I will pick only one programme of hundreds – it’s the Sea and Land Surface Temperature Radiometer which will be launched on the ESA/EU GMES Sentinel – 3 satellite in late 2015. Reputedly accurate to 0.3Kelvin and with a discrimination fine enough to pick up Urban Heat Islands and Hot Spots! (it’s googlable for details) Blimey, will this not blow the warmist/alarmists out of the water? Satellite is on a polar orbit so does the whole planet daily. I asked the scientist whether he collaborated/communicated with U of Alabama – he didn’t fire up on that – ‘we collab w. many U’s etc’. (I’m a recent mature history graduate from Southampton U myself)

    • ‘Blimey, will this not blow the warmist/alarmists out of the water?”
      Not after the output has been Mannipulated.

  98. So by “stringing the marbles together” does this not mean that they are lowering the resolution to make the last 18 years (of single-marble variability) disappear?

  99. They’re right that there is no “pause”, but for the wrong reason. There is no global temperature, no mean global temperature, no average global temperature, therefore there is no pause. All of this is worse than debating how many angels can dance on the head of a pin.

  100. The technique appears to be statistically stronger at rejecting a hiatus because it is statistically weaker at detecting one. If you are going to assume a time longer time dependency for a sequence of years, why not use what we know, that given the multidecade ocean modes, 60 years is about one sample of the climate. The oceans integrate over decades to centuries. The warmer or colder temperatures upwelling from below may be influenced by events decades or centuries before.

  101. Looking at the supplemental data, their treatment of Hypothesis II confirms the hiatus _”Having said this, from Figure 2 there is a clear pattern in the distribution of 16 year linear trends over time: all 16 year trends starting at 1950 all the way to 1961 are lower than the trend during hiatus period, and all 16 year linear trends starting at years 1962 all the way to 1982 are higher than the trend during the hiatus period, with the exception of the 1979-1994 trend.”_ The current hiatus is lower than all prior 16 year periods until you get back to the previous hiatus, which was the mid century cool period thought to correspond to the current hypothesis, whether based upon the Atlantic circulation or the pacific trade winds and PDO. They only reject the hiatus it seems by averaging the whole 1950 to 1997 period together. Of course the problem with a hiatus was based related to expectations set by the 1980s and 90s and the model projections. The latest 16 year trend disappoints those expectations even by their data. There is some trend in the current hiatus under this hypothesis. One test of this the semantic usefulness of this method, is to ask how long the current 16 trend would have to continue before they would recognize it as a hiatus. The answer is they never would, because it would never deviate from that average trend that includes the mid century cooling, yet all of climate science would be scrambling for an explanation of the divergence from expectations.

  102. Using a novel statistical framework that was developed specifically for studying geophysical processes such as global temperature fluctuations, Rajaratnam and his team of Stanford collaborators have shown that the hiatus never happened.
    “Our results clearly show that, in terms of the statistics of the long-term global temperature data, there never was a hiatus, a pause or a slowdown in global warming,” said Noah Diffenbaugh, a climate scientist in the School of Earth, Energy & Environmental Sciences, and a co-author of the study.
    Faulty ocean buoys
    The Stanford group’s findings are the latest in a growing series of papers to cast doubt on the existence of a hiatus. Another study, led by Thomas Karl, the director of the National Centers for Environmental Information of the National Oceanic and Atmospheric Administration (NOAA) and published recently in the journal Science, found that many of the ocean buoys used to measure sea surface temperatures during the past couple of decades gave cooler readings than measurements gathered from ships. The NOAA group suggested that by correcting the buoy measurements, the hiatus signal disappears.
    While the Stanford group also concluded that there has not been a hiatus, one important distinction of their work is that they did so using both the older, uncorrected temperature measurements as well as the newer, corrected measurements from the NOAA group.
    “By using both datasets, nobody can claim that we made up a new statistical technique in order to get a certain result,” said Rajaratnam, who is also a fellow at the Stanford Woods Institute for the Environment.
    **********************************************************************************
    My bold. Am I missing something here? Within four paragraphs there appears to be a complete contradiction – is this a form of what psychologists call projection, but used on oneself?
    SteveT

  103. Reblogged this on Climatism and commented:
    “The (18+ year) Pause” has most definitely not happened if you measure global temperature via: in-filled, homogenised, UHI effected land based station data from NASA, NOAA, BoM, MET et al.
    But “The Pause” is most definitely still happening, if your data source is via more accurate satellite data, that crisscrosses the globe measuring the temperature of every square inch of the lower troposphere, the exact place where global warming is supposed to occur and be measured.
    Let’s not forget, satellite temperature data was all the rage in the 90’s when it *was* warming.
    Now it’s scoffed at.

  104. Firstly, they use GISS data which is based on making up half the temps (where there are no stations, like in most of Africa,) and adjusting the other half to death.
    The pause (in their analysis A) STILL shows up.
    Then they start treating everything which increases too fast or too slow as an error, e.g. the 1998 high, but NOT the 1998 low which followed it.
    So they basically smoothe the data out of the multidecadal oscillation, and so they don’t find that oscillation any more – since they smoothed it out.
    Significantly they say that the ecologist and the grad student had the “concept” first, and the statistician was coopted later to bend the statististics to the preconceived conclusion.
    Really, after such a job one HAS to flush.

  105. When I first began viewing the NOAA global temperature graphs there appeared to be a hiatus in temperature rise from 1880 until about 1910 and another from the early 1940’s until the early 1970’s. These were interspersed by two warming periods of about the same length as each of the two periods of hiatus. This struck me as revealing a definite possibility of a stepped pattern of global warming which should have led to the prediction of another hiatus from about 2000 until 2030. Thus if such a pattern does exist then it would appear that there will be warming from around 2030 until about 2060 followed by another hiatus from around 2060 until about 2090. Global warming would then occur during only 40 years in the current century.

    • “Thus if such a pattern does exist then it would appear that there will be warming from around 2030 until about 2060 followed by another hiatus from around 2060 until about 2090.”
      Yep.
      You are seeing a ~60 year cycle apparently correlated with the North Atlantic Oscillation superimposed over a part of the ~1000 year cycle responsible for the Minoan, Roman and Medieval Warm Periods and the concomitant cold cycles such as the Dark Ages and the Little Ice Age.
      At no point is there any evidence of anthropogenic contribution.

  106. “Statistically, however, this is a hard claim to test because the number of data points for the purported hiatus period is relatively small, and most classical statistical tools require large numbers of data points.”
    Which didn’t stop us grafting on those same pitiful data points to get a hockey stick in the first place and jumping to a pitiful conclusion as a result.

    • Wishful thinking on the part of the Met Office.
      The hiatus will end around 2030, thus producing almost incontrovertible evidence that the ~1970-2000 warming period was part of a natural ~60 year cycle, and absolutely nothing to do with anthropogenic CO2 emissions.
      Not that such evidence will make any difference to their alarmist prognostications, naturally.

  107. The abstract itself is largely a political statement. The “hiatus” which is to be statistically defined out of existence is never defined. Hard to reward the report card with no visible grades. –AGF

Comments are closed.