How complexity science can quickly detect climate record anomalies Santa Fe Institute

From Eurekalert

Public Release: 14-Dec-2018

The history of our climate is written in ice. Reading it is a matter of deciphering the complex signals pulled from tens of thousands of years of accumulated isotopes frozen miles below the surface of Antarctica.

When making sense of the massive amount of information packed into an ice core, scientists face a forensic challenge: how best to separate the useful information from the corrupt.

A new paper published in the journal Entropy shows how tools from information theory, a branch of complexity science, can address this challenge by quickly homing in on portions of the data that require further investigation.

“With this kind of data, we have limited opportunities to get it right,” says Joshua Garland, a mathematician at the Santa Fe Institute who works with 68,000 years of data from the West Antarctic Ice Sheet Divide ice Core. “Extracting the ice and processing the data takes hundreds of people, and tons of processing and analysis. Because of resource constraints, replicate cores are rare. ”

By the time Garland and his team got ahold of the data, more than 10 years had passed from the initial drilling of the ice core to the publishing of the dataset it contained. The two-mile ice core was extracted over five seasons from 2007-2012, by teams from the multiple universities funded by the National Science Foundation. From the field camp in West Antarctica, the core was packaged, then shipped to the National Science Foundation Ice Core Facility in Colorado, and finally to the University of Colorado. At the Stable Isotope Lab at the Institute of Arctic and Alpine Research, a state-of-the-art processing facility helped scientists pull water isotope records from the ice.

The result is a highly resolved, complex dataset. Compared to previous ice core data, which allowed for analysis every 5 centimeters, the WAIS Divide core permits analysis at millimeter resolution.

“One of the exciting thing about ice core research in the last decade is we’ve developed these lab systems to analyze the ice in high resolution,” says Tyler Jones, a paleoclimatologist at the University of Colorado Boulder. “Quite a while back we were limited in our ability to analyze climate because we couldn’t get enough data points, or if we could it would take too long. These new techniques have given us millions of data points, which is rather difficult to manage and interpret without some new advances in our [data] processing.”

In previous cores, Garland notes that decades, even centuries, were aggregated into a single point. The WAIS data, by contrast, sometimes gives more than forty data points per year. But as scientists move to analyze the data at shorter time scales, even small anomalies can be problematic.

“As fine-grained data becomes available, fine-grained analyses can be performed,” Garland notes. “But it also makes the analysis susceptible to fine-grained anomalies.”

To quickly identify which anomalies require further investigation, the team uses information theoretic techniques to measure how much complexity appears at each point in the time sequence. A sudden spike in the complexity could mean that there was either a major, unexpected climate event, like a super volcano, or that there was an issue in the data or the data processing pipeline.

“This kind of anomaly would be invisible without a highly detailed, fine-grained, point-by-point analysis of the data, which would take a human expert many months to perform,” says Elizabeth Bradley, a computer scientist at the University of Colorado Boulder and External Professor at the Santa Fe Institute. “Even though information theory can’t tell us the underlying cause of an anomaly, we can use these techniques to quickly flag the segments of the data set that should be investigated by paleoclimate experts.”

She compares the ice core dataset to a Google search that returns a million pages. “It’s not that you couldn’t go through those million pages,” Bradley says. “But imagine if you had a technique that could point you toward the ones that were potentially meaningful?” When analyzing large, real-world datasets, information theory can spot differences in the data that signal either a processing error or a significant climate event.

In their Entropy paper, the scientists detail how they used information theory to identify and repair a problematic stretch of data from the original ice core. Their investigation eventually prompted a resampling of the archival ice core — the longest resampling of a high-resolution ice core to date. When that portion of the ice was resampled and reprocessed, the team was able to resolve an anomalous spike in entropy from roughly 5,000 years ago.

“It’s vitally important to get this area right,” Garland notes, “because it contains climate information from the dawn of human civilization.”

“I think climate change is the most pressing problem ever to face humanity, and ice cores are undoubtedly the best record of Earth’s climate going back hundreds of thousands of years,” says Jones. “Information theory helps us sift through the data to make sure what we’re putting out into the world is the absolute best and most certain product we can.”

###

Advertisements

187 thoughts on “How complexity science can quickly detect climate record anomalies Santa Fe Institute

  1. It was fine & interesting …then Jones spoiled it with –

    “I think climate change is the most pressing problem ever to face humanity,”

    Proving she has difficulty thinking.

    • Too often scientists these days are extrapolating beyond their field and drawing grandiose conclusions. Reporters should be vigilant and nip that in the bud.

      • Interesting that you think journalists should do this. Is your expertise journalism? Or are you extrapolating outside your feild?

        • Mosher,
          As an end-consumer of the product of journalists, I think that he is qualified to comment on what kind of writing he wants to have. His field is a consumer of the product of journalists.

        • Journalists, like scientists, used to have code of conduct they needed to follow to publish anything. Today not so much.

          All you have to do is read the Climategate emails to understand what is going on behind the scenes in your field and look at all the fabricated stories on TV (Dan Rather) to understand today’s journalism.

          Like to be a carpenter all you need is a hammer, to be a journalist all you need is a computer and an Internet connection.

      • “Too often scientists these days are extrapolating beyond their field and drawing grandiose conclusions.”

        WOW – I’m not sure you realize how prevalent this is, even at the highest levels. I’ve dealt with top scientists at IUCN, NSF – even Sherwood Rowland – where they could only see their shoes though answers they were searching for – though thru a different lens, a different knowledge base (in this case socio-economic & land-use solutions) were right in front of them. FOR SOME REASON, THEY HAD NO ABILITY TO CREATE BRIDGES INTO OTHER METHODOLOGIES FOR SOLUTIONS – AND WERE VERY COMFORTABLE GOING THERE (made me laugh – if people only knew!).

      • Steven Mosher December 15, 2018 at 3:13 pm

        “Or are you extrapolating outside your feild?” Good question.
        By asking the question you answered it.
        journalists are not schooled in the areas they report. They rely on the professionalism of the individuals informing them..
        So no, not their job.

        • “The Dansgaard-Oeschger events described in Sect. 2 are climatologically
          important and scientifically interesting, but their mechanics and dynamics are
          not completely understood. During the early stages of the collaboration that
          produced this paper, the geoscientists on the team conjectured that these events
          would inject new information into the time series. As is clear from Fig. 4, however,
          that is not the case. That is, while DO events may reflect changes in the
          dynamics of the climate (cf., recent work on “critical slowing down” [7,14,17]),
          they are not associated with changes in the information production of that system.
          Rather, they appear to be just part of the normal operating procedure of
          the climate system.”

          • Another tendency of yours Steve, when caught doing something stupid, you answer with a post that is completely non-responsive to the point being made.

          • “MarkW December 15, 2018 at 5:50 pm
            Another tendency of yours Steve, when caught doing something stupid, you answer with a post that is completely non-responsive to the point being made.”

            Point right on target, MarkW!

        • +10
          Tim Ball was right, although he appeared to limit it to climate scientists. Here on this site we often resort to it. It is so easy to do with some (like Griff) because we frequently repeat the same arguments which have not been rationally refuted, or they are using unsupported data. It is not just an expression of frustration that we can’t come up with refutation, but that the refutation is ignored or entirely misunderstood.
          Sometimes ad hominem attacks are cathartic – but not a winning strategy. And do we really feel better after using them?
          Mosh – did you note that your original post was an ad hominem attack?

      • Hardly a personal attack, Mosh. Jones made the comment which has nothing to do with the substance of the article, i.e. the value of high resolution, long period data records and the associated issues. There was no need to introduce such a partisan, politicised perspective. If the comment had been along the lines of ‘the climate change debate must have accurate, precise data sets from which to base the scientific discourse’ then that would be quite reasonable and an objective comment. What was said was apartisan take. It sounded like a ‘science communications’ sexing up of the topic for msm consumption.

        • quote was

          “proving she has difficulty thinking”

          A non personal attack would be : “She is wrong, here is why”

          This isnt that hard.

          Time for Raising the bar on comments at WUWT

      • “saveenergy December 15, 2018 at 2:18 pm”

        Provided a direct belief or religious quote made by the researcher. It is called “Conflicts of Interest”.

        Making your assumption the personal attack.

      • Interesting. in a piece in the West Australian 3 climate scientists involved in SR5 were talking about sceptics misrepresenting the science. And that when SR5 misrepresents Marcott. 😛

    • It IS the most pressing problem ever to face humanity — if you listen to the popular press and the politicians who respond to that press. It doesn’t mean that they are correct only that they have fooled themselves into believing they cannot be wrong.

    • Have you not heard the phrase ‘A sop to Cerberus?’

      That was the throwaway line that gets next years funding.

      She has less difficulty thinking than perhaps you give her credit for…

  2. Most pressing problem to face humanity? I think trying to unsubscribe from some magazine emails is more pressing.

  3. “I think climate change is the most pressing problem ever to face humanity, and ice cores are undoubtedly the best record of Earth’s climate going back hundreds of thousands of years,” says Jones.

    It sounds very much like, with most climate research, they first have their minds made up as to what they want to show and then tease the data to justify that preconceived notion. It would be refreshing if they said they would just let the data take them where it leads. But that may not lead to further grants.

    • Data is never pristine. Empirical fact. There are always data problems and data processing problems.

        • Although it’s conventional in scientific publications to treat “data” as a plural, it is not incorrect to treat it as a collective singular. This is well-settled in the finicky-grammarians’ world. (E.g., see Fowler’s century-old usage guide.)

      • However pretending that fancy statistical methods can take problematic data and make it pristine is not scientific. Heck, it’s not even rational.

        • Who is pretending that?

          Not me. Not these authors.

          had you read the paper, you would see that.

          Had you read the paper you’d see they were testing assumptions made by others who processed the data first

          ‘It is worth thinking about whether the preprocessing step outlined in the
          first paragraph of this section—which is the standard approach in this field if
          one wants an ice-core data set with even temporal sampling—could have disturbed
          the information mechanics of the data. The ramps introduced by linear
          interpolation introduce repeating, predictable patterns in the π of Sect. 3, which
          could skew the distribution of those permutations. For long enough interpolations,
          this should lower the overall WPE value, but the time scales of this effect
          are all but impossible to derive.

          To explore whether this WPE shrinkage was at work in our results, we carried
          out the following experiment…..

      • Mosher

        Definition of pristine

        1 : belonging to the earliest period or state : original the hypothetical pristine lunar atmosphere
        2a : not spoiled, corrupted, or polluted (as by civilization) : pure a pristine forest
        b : fresh and clean as or as if new used books in pristine condition

        Data may have problems, but the original data are pristine, whereas data that have been changed are not pristine.

        • If you guys want to go at it technically

          Data is direct measurement of observational outcomes under experiment.
          Interpretation and modelling is what you may do with data and it becomes a product, meaning the product either or both of those steps.

          Look at any scientific organization and they will call any modified data a “product” or “dataset product” to make it clear it isn’t direct measured data.

          In real science outside climate fantasy there is only data and products and you guys have conflated the two.

        • “Data may have problems, but the original data are pristine, whereas data that have been changed are not pristine.”

          Sorry all data is mediated unless you believe in errorless observation

          • All data is not mediated it is not permitted in science, products are data this mediated.

            I haven’t looked but I am betting that all major temperature, sea ice and sea level sets are called products if they are produced by a scientific organization because that is the correct term.

      • You cannot make a silk purse out of a sow’s ear. Reprocessing old flawed data just cannot reveal any new information. Try and try as you will, what they wrote down is all they wrote down, and all you and yours are doing is demonstrating your fundamental misunderstanding of the word “MEASUREMENT.”

        Tell all your friends.

        But, somehow, I feel that you will probably continue with your attempts, as you can fool some of the people all of the time….

    • …exactly what I was thinking

      “the scientists detail how they used information theory to identify and repair a problematic stretch of data from the original ice core.”

      …beat that data into submission…how do you “repair” something…you’ve never seen…unless you have a preconceived agenda of what you want it to be

      I’ll guarantee the highs will go missing

      • Conspiracy thinking.
        But here is a challenge. Make a prediction about what they adjust. A testable prediction.

        • Wouldn’t it be more effective to just look at what they have adjusted and note what directions and what amounts the changes were—if the data is publicly available. There should be copious notes on why the changes were made, so if there is a conspiracy, it will show up. So, is there a data set of adjusted and unadjusted temperatures with copious notes on why the changes?

          • No Sheri

            It is easier to throw rocks at a 2 year study without reading it

            “The Dansgaard-Oeschger events described in Sect. 2 are climatologically
            important and scientifically interesting, but their mechanics and dynamics are
            not completely understood. During the early stages of the collaboration that
            produced this paper, the geoscientists on the team conjectured that these events
            would inject new information into the time series. As is clear from Fig. 4, however,
            that is not the case. That is, while DO events may reflect changes in the
            dynamics of the climate (cf., recent work on “critical slowing down” [7,14,17]),
            they are not associated with changes in the information production of that system.
            Rather, they appear to be just part of the normal operating procedure of
            the climate system.”

          • Funny.

            quoting the paper is somehow a challenge to ” prove I am wrong”

            mr. anonymous will need a new game.

          • Once again, Steve proves that he doesn’t know the first thing about the scientific method.
            You can’t refute a criticism by merely repeating the argument being criticized.

          • All Mosher is saying is that the climate scientists couldn’t find any guilty evidence of CO2 in that study. There have been many studies as to the causes of pink noise (ubiquitous) as regards to climate and no definite conclusions have been proved. Extreme care must be taken when attempting to use Empirical Orthogonal Functions on chaotic non linear system time series data. Too many climate scientists do not have a good basis in statistics to properly analyze their data. They need to employ professional statisticians on their team. Michael Mann’s work being a prime example.

          • HotScot

            It’s your persistent MO.

            And there’s something wrong with being anonymous on the internet?

            For all I know you might not be Steven Mosher.

          • Mosher won’t even answer basic question about Berkley Earth which we have put to him and after he made various interesting claims. So I would not hold your breathe expecting any answers anytime soon.

      • Latitude,

        That is the problem with ice cores anyway. They tend to be an average of a ten to one hundred year period while the firn turns to ice layers. Their isotope measurements are not high resolution, to the peaks and troughs are always going to be missing. Ice isn’t much good for time periods less that a quarter century.

        • “The study reported here involves data from the 3405 m long West Antarctic
          Ice Sheet Divide core (WDC), which was gathered and analyzed by a team
          involving authors Jones and White [21,22]. This core, which covers a period of
          roughly 68 ka, is the highest-resolution and longest continuously measured record
          of its kind ever recovered from Antarctica. The high accumulation rate at the
          WAIS Divide—about 23 cm/yr in recent times—results in annual isotopic signals
          that persist for the last ≈ 16 thousand years”

          • I believe they are over-estimating resolution again. That is a constant problem in climate science. 23 cm gives a good long ways for the trapped air and dust to migrate within the layer and still be in the layer, but there is still 10 years or so before the firn is ice. This is a high resolution core – meaning 10 years at best.

            What does a 10 year running mean filter do to your data? It smears out peaks and valleys that are less than 10 years long.

        • “For the same time period, about 7,000 to 8,000 years before the present, two types of proxy estimates of CO2. The ice core data from the Taylor Dome, Antarctica, which are used to reconstruct the IPCC’s fficial historical record, feature an almost completely flat time trend and range, 260 to 264 ppmv (Indermuhle et al. 1999).
          On the other hand, fossil leaf stomata indices show CO2 concentrations ranging widely by more than 50 ppmv, between 270 and 326 ppmv ( Wagner et al. 2002).
          This difference strongly suggests that ice cores are not a proper matrix for reconstruction of the chemical composition of the ancient atmosphere.”

          Ice cores? Mosh’tly wrong!

      • “A very interesting feature here is the large jump in WPE between 5–8 ka. As it
        turns out, an older instrument was used to analyze the ice in this region. The
        WPE results clearly show that that instrument introduced noise into the data:
        i.e., every measurement contains completely new information, unrelated to the
        previous ones. As can be seen from examination of the red and grey traces in
        the figure, that noise was not visually apparent in the δD data itself, so the
        instrument issue was not detected immediately by the laboratory team. The
        fact that WPE brings out the disparity between the two instruments so clearly
        is a major advantage. (Indeed, that revelation has caused author White’s team
        to re-examine the data in the depth ranges where the blips occur in the WPE
        results, near 17, 26, and 30 ka.) Another interesting feature of Fig. 2 is the rise
        in δD WPE from 62–68 ka. This may be due to geothermal heat at the base
        of the ice sheet, which causes water isotopes to diffuse in that region, thereby
        injecting new information into the oldest section of the time series. This matter
        is discussed at more length below.”

      • That is in fact exactly the point of information theory techniques. To identify what you really don’t expect to see and look at it more closely.

        It’s a finer line then to be drawn between discarding it as anomalous or allowing it to refute a preconceived position.

  4. This work makes no sense to me. Ice cores by definition have limited time resolution. The time to firn closure varies with annual snow accumulation amounts, and hence depth. There is distortion, since ice flows.
    As one easy to understand (and readily googlable) example, Glacier Girl ditched on Greenland in 1942, exact lovation known from the pilot recoverynoperation. She was relocated in 1992 and recovered in 1993 at a location 2 miles away (ice flow) and 264 feet down (ice accumulation over 50 years.
    Just because something can be done (40 data points/year) doesn’t mean that it logically should be done.
    Or, to paraphase the great physics mind Ernst Rutherford, if you need complex statistics (or complex information theory) to get your experimental result, you should have done a different experiment.

    • In the shallower part of a core annual resolution is possible for the ice and dust, but of course not for enclosed gasses that can’t have better resolution than the time to closure.

      In the deeper parts the annual layers get very compressed and also smeared out by diffusion.

    • Personal incredulity is not evidence.
      When something does not make sense to you, the wise course and skeptical course is to suspend judgement.

          • Saying that a paper doesnt make sense to you is not evidence.

            Rud knows me. We just had lunch.

            Now it would appear to me that on a site devoted to evidence that one can point out what is evidence and what is not evidence, regardless of who makes the claim.

          • “Is that supposed to impress or something?”

            No all humans eat.

            But the supposition that I my purpose would be atagonizing Rud by pointing out a fact ( personal incedulity is not evidence) is rather funny. Do you also think Rud would be atagonized by healthy debate? Huh?

            You can ask him. go ahead.

          • Personal incredulity is evidence.

            The question is, of what?

            The second question is of course why one would choose to ignore or discredit it…

            There is, there can be, no truly objective view of the world.

            Despite protagonists on all sides claiming the One True Worldview.

            Science should be the means to correct the most erroneous. But it can never demonstrate the truth.

            Only the grosser lies.

            Personal incredulity is evidence that either the worldview is incorrect, or the data does not mean what you think it does.

          • Mosher
            Rud offered more than just personal incredulity. He offered a physical reason why the data should be suspect. You did not offer a counter argument as to why he would be wrong, except to point out you had lunch with him.

        • His point is correct.

          If you don’t understand study until you do, or suspend your input.

          Not sure why you guys have an issue with Mosher. As far as I can see his pokes are literally aimed at swatting the stupid on our side of the debate. This endeavor isn’t just rational, it’s literally one of the main skeptic critiques of the alarmist camp.

          • He didn’t say that he didn’t understand the study, he said their results make no sense.
            Until you can understand the difference perhaps you should suspend your input?

            PS: Others have answered the second paragraph of your post. And they did so before you posted.

          • Actually, he said the work makes no sense, not the results make no sense. thus its the study itself he has difficulty understanding not the outcomes. If you want to play semantics games try harder.

            As far as I can see the responses related to my second paragraph are nothing more than antagonistic tripe. Mosher is making valid critiques.

      • I was taking a course from the late Dr. Herbert Landar (Cal State LA, then) back around 1964. One day some students came in asking about one of the items on his collateral reading list (Benjamin Lee Whorf, it was). His reply was that his reading list was not confined to what he agreed with, but included work that was important, for various reasons. He admonished us that if what we were reading seemed to be so much nonsense, “…I beg of you, consider the possiblity that it may be so much nonsense.”

        • Benjamin Lee Whorf

          He proved that an autodidact, working full time in another field, could do work of great scientific worth. On the one hand it’s encouraging. On the other hand even just reading the wiki article shows the amount of work he did and it’s daunting.

          There are autodidacts posting on WUWT, and they deserve our respect and encouragement.

      • The way they use ‘information theory’ makes me think they aren’t particularly familiar with it. We’ve seen way too often that people dump a data set into Matlab and play around with modules until they get something that looks interesting.

        One of my heroes is Burt Rutan, an acclaimed engineer with vast experience analyzing data. I can’t find the quote but it’s something like, “If someone has to analyze the crap out of the data to get a favorable result, their work is almost certainly bogus.” link

        Just because you don’t understand something, doesn’t mean you should suspend judgment. If it doesn’t pass the smell test, you should become all the more skeptical.

        My knowledge of history told me that Dr. Mann’s hockey stick was wrong. I didn’t have to understand the statistics involved. The result was absurd.

        There’s also the old adage: BS baffles brains.

        Throughout my career, I noticed that the people with the most complicated explanations were those with a tenuous understanding of the subject at best. The senior engineers and scientists almost always communicated simply and clearly.

      • Steven Mosher December 15, 2018 at 3:22 pm
        Thank you for being civil. To all be civil.
        I am still looking through the paper.
        I may in the end disagree totally but I want to give the people who worked on this their due.

        michael

      • Yeah and neither is your personal credulity! The the wise course and skeptical course is to suspend judgement! And yet, here you are, all over this paper offering nothing of substance except your dance card!

    • “To create evenly spaced time-series data for δD, δ18O, and dxs, we first used
      the age model described at the end of Sect. 2 to convert depths to ages, and
      then re-mapped the data to a constant temporal spacing of 1/20th of a year
      using linear interpolation. The effective resolution of the data is 0.005 m. In the
      upper portions of the ice core, annual layer thicknesses are about 20 cm, so there
      are roughly 40 data points per year. At greater depths in the core, an annual
      layer may only be 4 cm thick, yielding eight data points per year. The accuracy
      involved in interpolating these unevenly spaced data to a uniform spacing
      of 1/20th year varies over the depth of the core; this matter, and its potential
      effects on the results, are discussed further at the end of this section. The specific
      age scale spacing of 1/20th per year was chosen because it preserves the structure
      and amplitude of the data—that is, there are no instances of significantly
      reduced amplitude in the signal, or losses in spectral power.”

    • Rud,
      Yes, that extra temporal resolution is likely to just be noise and that may be why they are finding ‘anomalies.’

    • The information content of the ice cores is limited by the factors Rud describes. The techniques in the study seem to promise to recover information which has been lost forever.

  5. I hesitated when I saw how the mathematician spoke of ‘tons of processing power’. Sounds like what we had back in the late sixties.

    • “Sounds like what we had back in the late sixties.”

      I worked on some of those machines; IBM 360 Model 30s and IBM 360 Model 40s. Some of those beasts (like the IBM 360 Model 91) had so many lights on the System Console that we used to joke that just pressing the “Lamp Test” button to light up all of the console lights so you check for burn-out lights would draw so much power that it would dim lights throughout the neighborhood.

  6. So that’s what polar ice is useful for other than a good G&T. Thousands of scientists employed to disappear up their own arseholes with infinite levels of data analysis.

    A common mistake that people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools.” – The Hitchhiker’s Guide to the Galaxy.

  7. The Total Perspective Vortex derives its picture of the whole Universe on the principle of extrapolated matter analyses. To explain — since every piece of matter in the Universe is in some way affected by every other piece of matter in the Universe, it is in theory possible to extrapolate the whole of creation — every sun, every planet, their orbits, their composition and their economic and social history from, say, one small piece of fairy cake. The man who invented the Total Perspective Vortex did so basically in order to annoy his wife. The Hitchhiker’s Guide to the Galaxy.

    • Yes tom. If your goal is understanding you always welcome more data and better methods.

      Some folks want to reject this before looking at it.

      Weird

      • From signal theory, that lost data is forever gone. If we “know” something about the underlying signal origin we can construct a theoretical reconstruction of the signal. If there is anything “unexpected” in that stretch of signal, we don’t reproduce it because we reproduce what we “expect”. This effort is no different and thus of no scientific value due to the circular reasoning involved. It works fairly well when the signal is something like human conversation, but compressed data signals are not worth trying on.

        • Actually not.

          “A very interesting feature here is the large jump in WPE between 5–8 ka. As it
          turns out, an older instrument was used to analyze the ice in this region. The
          WPE results clearly show that that instrument introduced noise into the data:
          i.e., every measurement contains completely new information, unrelated to the
          previous ones. As can be seen from examination of the red and grey traces in
          the figure, that noise was not visually apparent in the δD data itself, so the
          instrument issue was not detected immediately by the laboratory team. The
          fact that WPE brings out the disparity between the two instruments so clearly
          is a major advantage. (Indeed, that revelation has caused author White’s team
          to re-examine the data in the depth ranges where the blips occur in the WPE
          results, near 17, 26, and 30 ka.)”

          In general when commenting on a paper. It helps to read the actual paper RATHER THAN the MSM blurbs about the paper.

          or you can read the blurb and react to the Blurb using any set of standard skeptical
          responses. These responses ( raw data Rules!, they used models! ect ect) are not really thinking. They are just kneee jerk skeptical talking points.

          On your point. A change in the entropy of a signal bears investigation.
          All they did was calculate an entropy metric. When entropy increases that is a CLUE
          to look deeper.

          In this case they found an instrument problem.

          Helps to read the paper before forming a conclusion.

          This observation does not require a science degree. But those who have science degrees who commented before reading the paper, should be more diligent.

          A blurb written by MSM is basically an adjusted and filtered vesion of the actual science.

          read the paper first.

          its a good policy

          • Always fun when non science peeps start using entropy, I suspect I understand what you are trying to say but it is really badly worded.

            The problem is entropy in a signal is directly proportional to the amount of data but it is also directly proportional to noise and bandwidth. So high entropy data can be just noise. So for a good information signal you want a bit of entropy but not too much 🙂

      • In other words, if you don’t like the methods that Steve likes, you have proven that you aren’t a scientist.

          • I dunno Scot.

            That’s the title employers gave me. I don’t care much. In the end there are two kinds of people. people who rely on the authority of titles ( he’s a lawyer, he’s a doctor, he’s a marketing puke, he;s a scientist) and people who actually check the work and never check the title.

            So take Willis. I like willis because I can check his work. it would never occur to me to check his past. Doesnt matter. Same with Rud. It would never occur to me to check his science by looking at his law degree.

            In fact it would be hilarious to say Rud and Willis were wrong merely by the fact of their past.

          • Steven Mosher

            Establishing credentials is essential enough that governments do it before employing people.

            Your employers may have given you the title of scientists in which case they are either frauds or fools unless you have a science qualification none of us are aware of. Indeed, it’s incumbent on you to inform them you can’t accept the title of scientist as you would be misrepresenting yourself and them.

            Doesn’t Willis openly state he’s a self taught science enthusiast? So what would be the point of checking his background? And checking Rud’s credentials wouldn’t be important unless he were to represent you, then it becomes pretty essential.

            Nor do I wan’t to misrepresent your background, you’re a clever guy with an English qualification, it doesn’t mean you know anything about science. You cut and paste a lot, quote other peoples work but I don’t suppose you have ever undertaken credible, published science yourself.

            Your a marketeer Stephen, a noble pursuit in itself, but please don’t try to pass yourself off as knowledgeable about science, far less climate science, that just makes you less credible than me, and I have no qualifications, just two thirds of my life in marketing.

            I can smell BS a mile off.

        • mark

          I don’t recall ever saying anyone was or was not a scientist. That’s largely a social label and not an inherent property of folks.

          For myself I use the label employers give me or my co authors give me. not a big deal. what matters is does you method work and can others replicate your results

          • I was commenting on your tendency to dismiss anything that comes to a conclusion you don’t like, and to defend any method that you do like. Usually without understanding either method.

          • “I was commenting on your tendency to dismiss anything that comes to a conclusion you don’t like, and to defend any method that you do like. Usually without understanding either method.”

            Really, The job is to evaluate methods. let’s take nic Lewis on sensitivity

            As an example Mark, lets’s take his method in his papers with Judith Curry

            What do you think my opinion on this was?

            Guess?

            be careful you better read his papers start to finish.

      • When reading a statement such as “I think climate change is the most pressing problem ever to face humanity,” it is difficult not to believe the “scientific” interpretation of the data will find the global warming that is being pursued.

        There was a study showing that rapes and violent crime will dramatically increase as temperatures rise.

        Do these “scientist” ever tire of crying “wolf”?

        When I read or hear “climate models predict, hottest year ever, CO2 footprint, global warming, we must act now to save the planet, climate scientists predict, etc., etc., etc.”, whatever follows is not worth listening to or wasting time reading. “Climate Scientists” have lost all credibility, they wouldn’t know real science if it bit them in the rear end.

      • Yes tom, If your goal is understanding – the Flying Teapot – you will always welcome more data and better methods.

        Some folks want to reject – the Flying Spaghetti Monster – before looking at it.

        Weird 😉

  8. “A spike in entropy from 5,000 years ago.” What? they never even said what they were measuring other than “Isotopes.”

    This report contains little if any information…

    • Did they lift “the spike in entropy” directly from Star Wars? It sounds like that kind of science.

        • Steven, thanks. I found an amusing sentence in your reference: “All references and original papers seem quite new. Instead of calling it an emerging topic, it might better be considered embryonic.”

          • Every once in a while people attempt to apply information theoreci metrics in new fields

            it’s been around a long time. Here’s a hint, I was using Shannon entropy to evaluate
            stylistic shifts in texts back in the 80s.

        • It’s actually a rehash of some very old work circa the 70’s or 80’s, sorry I don’t remember the exact timing. It was picked up and used with EEG and it’s use was patented for anaesthesia monitoring in the early 2000’s.

          The C code for the technique was published and infact converted into VHDL code for use in FPGA there are dozens of variants kicking around on the net. Someone from that background obviously worked out it might work on ice core samples. It won’t magically fix your data but it does help with monitoring a signal.

          I guess the positive is finally some of the more advanced signal processing techniques are creeping over to climate science which is long overdue.

  9. The most remarkable thing about the WAIS Divide ice core is that drilling was deliberately broken off before reaching bedrock. Officially this was to protect the unique bacteria that might be living under the ice from the drilling fluid, but I have a strong suspicion that they were afraid that the basal ice or the sediments under it might be older than the last interglacial which would have killed the whole WAIS Collapse narrative. Also note that according to the party line there can’t even be any unique bacteria there, since the area is supposed to have been deglaciated just 100,000 years ago.

    • Suspicions are not evidence.
      Wait.
      Yes it was popper who said science was judged by whether tty had suspicions.

      • In his first post on this article, Steven took another poster to task for issuing insults rather than dealing with the arguments made.
        Once again Steven indicates that he has no intention of living by the standards that he demands of others.
        Typical warmist.

        • tty is wrong

          Suspicions are not evidence.

          if you think it is an insult to have your mistakes pointed out mark, then leave.

          Now if I called tty a name, or insulted him personally, you’d have a point.

          I am sure tty is intelligent and fair minded.

          he happens to wrong about this issue.

          • If you had merely whined “you’re wrong”, like you usually do, that would have been the end of it.
            The rest of your post was the insult.
            Sorry you aren’t man enough to admit your own hypocrisy.

          • Mosher
            If an “ntelligent and fair minded” professional has suspicions about the veracity of data or a claim, only a fool or egotistical person would dismiss the concerns out of hand without considering the stated reasons for the suspicions.

            Ridiculing tty is the same as insulting him personally. I would think that you would understand that. Do you not understand it, or are you purposely being disingenuous?

            If you think it is an insult to have your mistakes pointed out Steve, then leave.

          • “Ridiculing tty is the same as insulting him personally.”

            Not to worry. Being insulted by Steven Mosher is virtually a compliment.

            Incidentally I never claimed that my suspicion was evidence. However it seems to me that the excuse given for failing to obtain the scientifically most important part of the ice core after a five year effort is extraordinarily weak. Any number of ice cores have been drilled straight through to bedrock previously both in Greenland and Antarctica, with much of the scientific yield coming from the basal ice and the sediment under the ice.

            The only other case where this was deliberately avoided was when drilling over Lake Vostok. In this case the the wish not to contaminate a unique environment that has probably been isolated for tens of million years is understandable and reasonable, particularly since some of the penetrated ice was frozen water from Lake Vostok, so a great deal of data could be obtained without penetrating into the lake. However this would not apply when drilling just about anywhere else in Antarctica.

  10. “A sudden spike in the complexity could mean that there was either a major, unexpected climate event, like a super volcano, or that there was an issue in the data or the data processing pipeline.”

    Or it might be a melt layer. They are unusual in Antarctica but they do occur.

    Having done a fair amount of time series analysis I can state with some confidence that the best way to find anomalies in time series is to use the Mk I Eyeball. However our brains have an inveterate habit of finding patterns where there aren’t any, so the finds should always be verified by statistical analysis.

    • Our brains created the statistics. How can we know we didn’t write a pattern into the stats because we thought it should be there? (Not being contrary. It’s just that people often think we can somehow avoid prejudice in probability and circumstantial analysis. I don’t see that we can.)

      • We can if the math was not applied with an apriori bias in the construction of the method. If we do a straight dependent-independent variable analysis to test the null hypothesis of no relation we don’t know what the result will be. However, if we devise a method that will always pull the shape we want from red noise, then we can say the math was bogus.

        • ” if we devise a method that will always pull the shape we want from red noise”

          Incidentally a pretty good description of Bayesian statistics where the unavoidably biased choice of prior probability can do just that.

  11. No matter what the subject is, just remember to utter the magic words “Climate change” and our highly qualified politicians will twitch and send more money.

    M<JE

    • Steven Mosher

      People who worship raw data forget that all data comes from instruments. And no instrument is perfect.

      Nonsense. Data is derived from observations, instruments are convenient tools.

      • ‘Nonsense. Data is derived from observations, instruments are convenient tools.”

        do you think folks look at the ice cores with their eyes? and observe

        Nope.

        do you think we measure temperature with our skin?

        Nope we use instruments

        Do you think UAH observes the temperature in the troposphere?

        Nope.

        Methinks you have never worked in a lab or taken test data

      • Wrong wrong ,, wrong
        If an instrument has failed, at some point and you cannot locate the point of failure ALL data is suspect.

        michael

        • That is definitely how it is treated under normal science norms. If you want to exclude data based on a systemic failure you have to be able to identify the failure otherwise you are cherry picking data.

    • Once again, Steven demonstrates that he has no intention of arguing with honesty or integrity.
      Nobody worships”raw data”, what we do is object to the methods certain people utilize in a vain attempt to draw more meaning from the data than the data contains.

      If the data is corrupted by bad instruments, then you have to use it with increased error margins. It is not scientifically valid to “assume” that you know exactly what the “error” is, and remove that error without telling anyone.

      • Mark.

        Look around here and see how many people believe that observations with eyes is all science does.
        Explain to them that people use instruments, and they will argue.

        Further, once they identify a bad instrument they have choices:

        1. Reprocess with better instruments ( which they are doing)
        2. disgard
        3. Adjust

        Nobody is removing error without telling you

        • I have no issue with those options although I am taking some faith on 2, and for your options you have terms

          1.) new data
          2.) rejected data
          3.) product dataset

          Rejecting data is always problematic because it can be viewed as an attempt to filter problematic results. So long as there is a clear systemic fault that makes the data inconsistent then I am fine with rejecting it.

        • If they aren’t removing error, why aren’t there error bars on the graphs they produce?

          It’s not so much that they don’t tell us they are removing error, it’s that they don’t tell us how they removed the error.

          Regardless, you can’t remove error, you can only account for it.

    • Steven Mosher

      PS

      Your link doesn’t work.

      And if you want evidence of crappy instrument derived data just examine the data from the mid 19th Century on land and sea surface temperatures that you yourself quoted to 1/10 of a degree the other day. Utter nonsense.

    • “People who worship raw data forget that all data comes from instruments. And no instrument is perfect.”

      My thermometer works pretty good. I don’t need to manipulate the data it shows in any way, to tell what the temperature is outside.

      I can write that temperature reading down and in 50 years it will still be the same number and just as accurate as it was on day one. No amount of manipulation will make it more accurate..

      If I turned that data over to Climate Alarmists they would tell me I didn’t do it right, and that’s not the correct temperature for that day, so they have to modify the data to correct it.

      What a scam!.

    • People worship adjusted data, not raw data. That’s why AGW is know as a religion. If the raw data doesn’t fit the belief system, then it gets adjusted until it does or it gets rejected.

    • Since I’ve been reading for the last 60 years about one time airtight science being rejected and improved upon decades later, I don’t get excited by what is purported to be the last word. Good for them in making these advancements. But in a few decades a new group of scientists with even better technology will be saying “ Do you remember what that team did in 2018, yeah well, here is what we’ve learned since then”

      Bank on it.

      • ‘A First Step Toward Quantifying
        the Climate’s Information Production over
        the Last 68,000 Years”

        Note the hubris in their title

        • I didn’t take exception to them. It’s the idiots who will believe this is the last answer. They are the politicians and media and the warmist activists who treat every weather event and every study as evidence of AGW or whatever agenda they have.
          With a few exceptions, the actual worker bees, the on the ground scientists are just doing their best and an honest days work.

  12. The bottom line is there is variation, due to ice ages and many other reasons, change is constant, we already know that and more data does not change that fact.

  13. https://www.santafe.edu/news-center/news/predicting-unpredictability-information-theory-offers-new-way-read-ice-cores

    https://link.springer.com/chapter/10.1007/978-3-319-46349-0_30

    http://b-ok.cc/book/2803949/6ce377

    For guys who like Shannons work.

    “This paper is about one piece of that question: what the Shannon entropy
    rate of the water isotope signals in a specific Antarctic ice core tells us—about
    that data, about the past conditions at the core site, and about the overall
    climate. In an ice core, layers capture information about the local conditions at
    the time of deposition. A depth-wise series of measurements of some chemical
    or physical property of the ice, then, is effectively a time-series trace of those
    conditions. Water isotopes are a particularly useful property to study because
    they are good proxies for temperature and atmospheric circulation that result
    from variability in the hydrologic cycle. The time scale is unknown, though,
    and understanding the specific form of the relationship between the measured
    quantity and different aspects of the climate system requires forensic reasoning.

    ….

    The Shannon entropy rate is a potentially useful way to carry out forensic
    reasoning about the climate system. It measures the average rate at which new
    information—unrelated to anything in the past—is produced by the system that
    generated the time series. If that rate is very low, the current observation contains
    a lot of information about the past and the signal is perfectly predictable. If that
    rate is very high, all of the information in the observation is completely new: i.e.,
    the past tells you nothing about the future. Calculated over time-series data from
    ice cores, this quantity—described in Sect. 3—allows one to explore temporal
    correlations in the climate, which are critically important in understanding the
    underlying spatiotemporal mechanisms of this complex dynamical system. The
    results of these calculations, described in Sect. 4, are quite promising; they not
    only corroborate known facts, but also suggest new and sometimes surprising
    geoscience, and pave the way towards more-advanced interhemispheric entropy
    comparisons that could elucidate some of the deeper questions posed above about
    the larger climate system.

    …..

    “Modern ice cores, from which data sets like the one in Fig. 1 are derived, cover
    timespans of up to 800,000 years. These can reach over 3 Km in length and are
    typically analyzed on a scale of cm—and, for some properties, mm. Each sample
    may involve dozens of measurements: different kinds of ions and isotopes, dust
    levels, conductivity, and so on. Some of the more useful of these are the stable
    and radiogenic isotopes, the amount and type of dust (which are correlated to
    the energy and humidity of the atmosphere), and the conductivity. The dynamic
    ranges of these measurements can be huge: sulfate levels go up by a factor of
    1000 when a volcano erupts, for instance. Noise levels vary greatly across the
    different measurements, but those levels are not well established—and indeed
    are the subject of some important arguments about how to distinguish signal
    from noise. And of course the analysis equipment affects the data, sometimes
    without leaving any visually obvious trace in that data. That issue will return
    later in this paper.”

    ….

    “The study reported here involves data from the 3405 m long West Antarctic
    Ice Sheet Divide core (WDC), which was gathered and analyzed by a team
    involving authors Jones and White [21,22]. This core, which covers a period of
    roughly 68 ka, is the highest-resolution and longest continuously measured record
    of its kind ever recovered from Antarctica. The high accumulation rate at the
    WAIS Divide—about 23 cm/yr in recent times—results in annual isotopic signals
    that persist for the last ≈ 16 thousand years, ”

    ….

    “In this paper, we focus on the water isotope
    measurements in this record: specifically δD, the ratio of 2H (deuterium, D) to
    1H, and δ18O, the ratio of 18O to 16O. Their values are reported in mille (parts
    per thousand, or “per mil”), relative to a calibrated standard of the isotopic
    composition of fresh water [1], and are generally negative for glacier ice. A δD
    value of −250 mille, for instance, means that that water sample is depleted in
    deuterium by 250 parts per thousand, relative to that standard.

    ……

    All of those measurements are on a depth scale; to do any kind of time-series
    analysis, one must convert them to an age scale. This requires an “age model”
    for the core: a mapping of depth to age. Constructing this mapping requires a
    subtle, complicated combination of data analysis and scientific reasoning. Layers
    can be counted, for instance, but only to a maximum of 40–50 ka because the
    upper layers compress the ice underneath, thinning the layers to the point that
    they are unrecognizable. The measurements in the core play a key role in agemodel
    construction: the astronomically based “Milankovitch” theory of ice ages
    predicts how δ18O should vary through time, for instance. But ocean δ18O also
    depends on the total volume of land ice on Earth1, so this quantity is also a
    useful climate proxy. And near the base of the ice sheet, the ice often melts
    and/or deforms, making dating—or any kind of data analysis—very difficult.”

    • Steven Mosher

      You are a marketing bod, a salesman, who cuts and pastes.

      You pretend to understand science, you might even believe you understand science, but you don’t.

      • Thank you for your concerns Hot one.

        Lets assume that was true. ( my co authors thought otherwise– )

        But lets assume you are right.. just a marketing guy

        People should find it embarrasing that they are shown to be wrong by someone with no training who merely reads the actual papers.

        Thats too funny. Some marketing body can show that the other commeters are wrong.
        How?
        he reads the fricken paper.

        DOH!

        • Steven Mosher

          Thank you Steven, I’ll have the courtesy to refer to you by your adopted title on WUWT.

          Your co authors may have been misled by your employer awarding you the title of scientists and you not having the common decency to turn it down.

          And you should find it equally embarrassing to be challenged by someone with no meaningful qualifications whatsoever, that would be me (“Hot one” according to you).

          Reading, and cutting and pasting papers does not demonstrate an understanding of science, I mean, even I can do that.

          • “Your co authors may have been misled by your employer awarding you the title of scientists and you not having the common decency to turn it down.”

            Err, they gave me the title after proving myself by working with them.
            You seem to think that the co authors and employers are two different groups.

            And the second employer also looked at the publications and decided that the title
            fit.

            here is the thing, the Only person who cares about titles is you.

            Employers, Co authors, Other readers? they care about the work. They look at the work and choose to refer to me as they wish.

            you are free to use whatever title you want. or not.

          • Sorry to break it to you Stephen but if you don’t hold a degree in the science area you can’t really be called a scientist. You also can’t win the Nobel Prize for Sciences no matter what you discover, except if you name is Marconi but there is a special reason for how an Engineer came to win the Science prize.

          • I would add I am surprised that people in the Climate Science field haven’t objected to you being called a scientist or is that an internal title in your group?.

          • LdB said:

            “Sorry to break it to you Stephen but if you don’t hold a degree in the science area you can’t really be called a scientist.”

            scientist

            noun

            A person who is studying or has expert knowledge of one or more of the natural or physical sciences.

            But good luck with your attempt to redefine the word.

          • Sure and a doctor is someone who heals people, no degree required so I shall just hang a sign up and start treating people shall I.

            Just because layman use words in stupid ways doesn’t mean they can be used in that way in a discipline.

            Hence why I said I am surprised other climate scientists had not complained because he doesn’t have the qualification to be called a scientist.

      • With all the appearances here, Mosher must be agitated about the pathetic record of the 49ers. But, there is hope for the Frisco fans. After going 2-14 in 1979 they followed Joe Montana to the promise land and won a bunch of Super Bowls.

  14. Climate lientists often make a show of being objective. But Garland gave herself away at the end. The agenda is, as always, to prop up Climatism.

  15. “For the WAIS Divide core, the construction of the age model required several
    person-years of effort. The top 31.2 ka of the core was dated by four different
    individuals and one HMM-based software tool [24]; this procedure entailed visual
    identification of annual fluctuations in several different chemical traces along
    thousands of meters of core, followed by cross-corroboration between different
    proxies and different daters [20,21]. From 31.2–67.8 ka, the age scale was based
    on stratigraphic matching to “gold standard” Greenland ice cores and cross
    referenced using uranium/thorium ratios from cores drilled from cave features
    [5]. This represents the state of the art for data analysis in this field.”

    For folks who dont know an HMM is a hidden markoff model.. same kinda model one can use for Speech recognition for example. Or if you were studying literature ( say english literature) you can use entropy measures to detect Stylistic Shifts. Some day you can ask me how I used Entropy to figure out it was Glieck who forged the heartland document.

    “The rate at which new information appears in a time series has been shown to be
    an effective method for signaling regime shifts: e.g., epileptic seizure detection
    in EEG signals [6], bifurcations in the transient logistic map [6], and recognizing
    voiced sounds in a noisy speech signal [3]. Estimating that quantity from an arbitrary,
    real-valued time series can be a real challenge, however. Most approaches
    to this problem use the Shannon entropy rate [15,19] and thus require categorical
    data: xi ∈ S for some finite or countably infinite alphabet S. This is an issue
    in the analysis of the type of high-resolution data produced by an ice-core lab
    because symbolization introduces bias and is fragile in the face of noise [4,13].
    Permutation entropy (PE) [3] is an elegant solution to this problem. It symbolizes
    the time series in a manner that follows the intrinsic behavior of the system
    under examination. This method is quite robust in the face of noise and does
    not require any knowledge of the underlying mechanisms of the system. Rather
    than calculating statistics on sequences of values, as is done when computing the
    Shannon entropy in the standard way, permutation entropy looks at the statistics
    of the orderings of sequences of values using ordinal analysis. Ordinal analysis of a
    time series is the process of mapping successive elements of a time series to valueordered
    permutations of the same size.”

    • Steven Mosher

      Some day you can ask me how I used Entropy to figure out it was Glieck who forged the heartland document.

      Now that’s something I would be interested in hearing from you Steven. Why not do an essay on it for WUWT. Far more credible than your claims of understanding climate science.

      Signed.

      Hot one.

      • “Now that’s something I would be interested in hearing from you Steven. Why not do an essay on it for WUWT. Far more credible than your claims of understanding climate science.”

        Why? so you can make more personal attacks? It would be foolish of me to spend any considerable time doing a post. In the past, yes, when commenters spent more time challenging the actual claims.

        maybe if the bar gets raised with regards to comments

    • “this procedure entailed visual identification of annual fluctuations in several different chemical traces along thousands of meters of core”

      Just as I said before. The Mk I Eyeball is still state-of-the-art when it comes to pattern recognition.

  16. “But imagine if you had a technique that could point you toward the ones that were potentially meaningful?”
    And who determines meaningful? Like the IPCC determined only CO2 should be studied?

  17. Mosher usually seems to be playing the role of devil’s advocate to me. I believe he is more interested in promoting discussion than trolling (sometimes hard to separate the two). Just my $.02 since so many posts on the subject of where his loyalties lie in with AGW.

  18. “When that portion of the ice was resampled and reprocessed, the team was able to resolve an anomalous spike in entropy from roughly 5,000 years ago.”

    By “entropy,” I assume they mean what cultists today call “climate weirding”?
    (Where things seemed to go haywire for a period in the weather.)

    That “weirding” probably happens at all times scales, depending on the resolution of the data.
    Extremes Happen. 5,000 year tail end probability events happening about every 5,000 years.
    1,000 year tail end probability events happening about every 1,000 years, etc.

    So who is the culprit who was driving SUVs and burning coal 5,000 years ago?

  19. Climate “science” meets Complexity “science”.
    Complexity science is much like google. Humans make a bunch of algorithms that tease out small things from millions to billions of points.
    Didn’t google start something recently where they use complexity “science” to tell us which news is legitimate and which is fake?

  20. “It’s vitally important to get this area right,” Garland notes, “because it contains climate information from the dawn of human civilization.”

    Wut? There is double ought zero connection. Humans didn’t even get to Antarctica for millennia.

    From the West Antarctic ice shelf. One point in 500,000,000 square kilometers. It’s not vitally important at all. It’s not even important at all. It’s interesting, and curious, but not important.

    To the extent that the ice cores are a decipherable proxy for weather, they tell us about one point on a vast planet. Useless.

    BWTM: the temperature was below 32 degrees forever. Else not ice. Since most of the earth is NOT covered with ice, what is a non-typical ice covered area supposed to tell us?

  21. Interesting concept,so information theory modelling of data may be useful in detecting signals the eyeball misses.
    Time will tell.
    That thin slices of Antarctic ice can give detailed information of the worlds? climate in days past?
    Just what is the signal to noise ratio?
    Awful lot of speculation in that.
    Mark Twain would love modern science.

  22. Interesting how every site needs a village idiot, WUWT certainly had theirs. You tell who they are they are the ones rambling nonsensical with everyone who cares to engage

  23. Insight from the economic modelling given by Mervyn King in his book “The end of alchemy”:

    Optimizing over a false model is in many instances worse than the use of a coping strategy that works in your particular environment. Rather than attempt complex statistical calculations, it is better to make investment decisions using a choice of heuristics that reflects a sensible narrative

    Rather than investing in climate models and limiting CO2, we should be prepared to a variety of weather incidents

    Forest management is a good coping strategy for wild fires.

  24. Too funny!
    They’re still keeping up the pretence of being interested in palaeo-climate.
    But only denyers deny that the world was created in 1850.

  25. “I think climate change is the most pressing problem ever to face humanity,”

    Why disagree? So long as politicians continue to believe that the earth is warming, and will continue to warm disastrously, as a result of humans using fossil fuels, the economic cost of their idiotic attempts to drastically reduce CO2 (and CH4) emissions could eventually lead to civil unrest on a global scale.

    But perhaps this is not what she meant and the phrase was only included to ensure publication and maximum publicity.

Comments are closed.