Mathematics, Media Criticism, Political Commentary, Science And Society, Technology and Society

On Climate Change

12 Comments

…the science battle rages.

It would sure be nice if some of these self-identified climate “scientists” would learn some statistics. And stop calling people who understand statistics names like “denier” and “anti-science.”

[Monday-morning update]

Inside the latest global warming scandal:

This kind of thing is going on all over the world. It is one of the reasons why the satellite data (which, however, go back only to 1979) are so important: they have not been corrupted.

Yup.

[Bumped]

[Update a few minutes later]

Soviet-style disinformation dominates the climate “debate”:

Why is it that when a political figure makes a misstatement about a global warming-related issue, which happens many times every day, no government scientific agency or leading university scientist ever corrects them?

For example, all climate modelers correctly label their speculations of future world temperatures as “projections,” meaning that they have no validated forecast skill. Yet politicians, mass media, and the public treat the models as providing temperature forecasts or predictions. Because this misusage is never corrected, politicians cheerily continue to base expensive public policy on it.

Another example: carbon dioxide, as an essential factor in photosynthesis, is the elixir of planetary life, yet politicians dub it a “pollutant.” Similarly, badging the theoretical global warming problem as a “carbon” issue represents scientific illiteracy because it fails to distinguish the element “carbon” from the molecule “carbon dioxide,” and deliberately encourages the public to confuse a colorless, odorless, beneficial gas with soot. Again, climate-alarmist scientists say little or nothing to correct these mistakes.

Many in the public understand that Hendricks’ behavior is typical of politicians everywhere. But most people do not recognize that fraud is also being directly committed in support of this travesty by many of today’s self-appointed “leading climate scientists.” For when they are not directly massaging the data relied upon in their scientific writings, these scientists often report their findings in ways that are intended to deceive the reader into believing that dangerous global warming exists, or will shortly exist. The UN’s climate reports are the magnum opus of this style of operation.

Yes.

12 thoughts on “On Climate Change”

  1. ‘You don’t have to fully grasp Lewis’s analysis to realize the absurdity of liberals’ claims that climate skeptics are “anti-science.” ‘
    I don’t? He throws around some nice equations, and presto: Climate skeptics are right, and believers in AGW are the ones who are inept.
    Megan McArdle’s title for the ideal blog post: “Everything You Already Believe Is Completely Correct, and Here’s Some Math You Won’t Understand That Proves It.”
    I am saying this even though I have spent time now at climateaudit trying to clarify the issues in the post, and I expect that Nic Lewis is right and the Nature article ought to be withdrawn.
    I don’t really understand how non-math people are expected to make good decisions in this day and age.

      1. MikeR, what is the actual science being done here? For example, I don’t see any connection to real world data. For example:

        Additionally, Lewis claims that the values for climate feedback parameter α and ocean heat uptake efficiency κ are so uncertain as to render them useless. But the α and κ values we use were diagnosed previously using established methods, relying on strongly forced, idealized model simulations (Andrews et al. 2012, Kuhlbrodt & Gregory 2012; Vial et al. 2013, Flato et al. 2013). These approaches and simulations are defined such that α and κ can be viewed as being model properties. By contrast, Lewis used historical simulations in trying to diagnose α and κ.

        What is meant by “historical simulations” here? That sounds an awful lot like actual data to me.

          1. Ok, so what does that have to do with science? At a glance, I don’t see a connection to empirical data or even, for that matter, to behavior of known models. And this particular sentence from my original quote smells of circular logic:

            These approaches and simulations are defined such that α and κ can be viewed as being model properties.

            That only makes sense, if these parameters are stable enough to be considered as a model property. The criticism by Lewis was of the opinion that they were not. If true, then that “define” transforms the model into a new model which has eliminated a significant fluctuation or variation of the original model in a way that makes these parameters conform with the assumptions of the transformation.

    1. The part that has always boggled me is the propagation of errors for measuring surface temperature, and the complete unwillingness to perform actual calibrations and cross calibrations with different methods of measuring.

      Performing homogeneity adjustments on a pool of good data and bad data in preparation for searching for non-homgeneities is just not how I’d expect to find a climate signal.

      And I certainly wouldn’t be claiming a 0.1C error on the measurement of ‘surface temperature’ in a 100km x 100km box from a single (humidity adjusted, but otherwise perfectly) normal +/- 0.1C thermometer. I can’t make -that- assumption on a mixxed 1000 gallon tank with -much- simpler fluid and mass transport issues. I can -cheat-, and say “Well, I only -actually- need the inlet temperature, the ambient temperature, and knowledge of the mixing”. But a climactic shift is essentially directly analogous to moving the inlet without moving the thermometer. And changing the mixing. All of a sudden, my estimate of the tank temperature no longer comes within the calculable-but-COMPLETELY-WRONG error estimate.

      1. And I certainly wouldn’t be claiming a 0.1C error on the measurement of ‘surface temperature’ in a 100km x 100km box from a single (humidity adjusted, but otherwise perfectly) normal +/- 0.1C thermometer.

        Al, thank you so much for responding to my comment. I honestly could not have written my sentiment any better than you expressed. All of it is well stated, but the part quoted would be enough to make me worry about my own licensing and credentials.

        To some, I’m just a non-math skeptic who couldn’t possibly understand models for complex systems like the climate, and thus shouldn’t be in the role of making decisions. But hey, I heard some people in the name of Christianity during The Inquisition made similar statements to people who were skeptical of the accepted “science” of the time.

        1. I’ve been saying this for quite a few years now. The responses have been in one (or more) of these categories: crushing-a-random-strawman, “Yes, but that’s just a proxy, we just need the trend of the proxy”, “Why on Earth would you even attempt a formal calibration between different proxy/instruments?”, or something along the lines of dismissing the surface record entirely like “Yes, but the satellites show the same exact warming we’re all going to die!”

          The baseline is a key piece of evidence.
          The variability of the baseline is a key piece of evidence.
          And assuming any discontinuities at a specific station are always effectively “user error” seems daft.

          1. Above, Karl mentions circular logic. Like you, Al, I’ve heard references to satellite data as if the instruments on the satellites were magically calibrated with software that had no engineering assumptions and with hardware that had no fluctuation or variability. That’s interesting to me.

            I know the level of precision that goes into crafting instruments that are expected to operate without human intervention for a decade or more. But I also worked with some of the people who built or checked these items. Things like this happen. Further, we also kept logs on MTBF of space equipment, and there was variability in those failures, and often analysis of failure was based on accelerated testing environments that had questionable calibration.

            With all the calibration issues, variability in precision, and human error potential; it’s interesting to see people assume that satellite data is precise and accurate to provide that .1 degree variance in downstream modeling. Also note that the .1 degree variance has been reported for quite a long time now, so while we may have gotten better in building satellites, calibrating them, and launching them; the satellite data from a decade ago comes from satellites designed and built almost a decade before that.

  3. For me, the thing was that here was fluctuations that only appeared in the future predictions. If your models aren’t accounting for fluctuations in the data, then the past predictions should also be off because of those fluctuations, not just the future predictions.

Comments are closed.