Friday, October 7, 2011

Wanted: Climate Change Argument, Not Advocacy

Major Suspect in Global Warming Case Hides His Face.
Photo by the author.
     Global climate change seems to be of great interest to readers of RESPVBLICA. Not having studied meteorology or climate science, I cannot say much one way or the other about lots of what is presented on either side of the debate. Just when it seems clear that the rise in atmospheric carbon dioxide is probably responsible for rising global average temperature, another study purports to document a stronger link to sunspot activity. At first blush, either theory appears plausible to laymen like me. Moreover, knowing how people not trained in a specialty make basic errors—such as misunderstanding the burden of proof in a civil case—should make any specialist wary of hasty conclusions about matters outside his own expertise. This is especially so since we all tend to evaluate the global warming arguments as partisan positions. It is altogether too tempting to accept the arguments made by our side and reject the arguments made by the other side, whether or not we are really qualified to judge the science. But we laymen can do more than try to evaluate the science directly.



     We all have our own specialties. For a lawyer it is comparatively simple to recognize when someone is grinding an axe. That is to say, as a professional advocate, I can recognize advocacy. Much of the time, on both sides of the debate, I find advocacy instead of argument. Politics instead of science. Ad hominem attacks, false dilemmas, straw men, reductio ad absurdum—all the usual techniques of polemic. That’s perfectly normal, but it doesn’t clarify the issue beyond hinting rather strongly at bias. All the more so when advocacy becomes fraud, as has happened with the “hockey stick” incident Jillian Becker rightly noted over at The Atheist Conservative. Scientists should beware:  In the courtroom, once a witness has been shown to be untruthful about a small thing, the jury is all the more willing to conclude he is lying about the big thing. Once lost, credibility is very hard to recover.

     In the end, we are going to have to come to grips with the science. If one side has succeeded in obscuring the truth, all shame on them. Because this issue involves very high stakes indeed, we have to get it right. Toward that end, the more we can avoid partisan invective and actually explore the science, the better.  The thoughtful comments by RESPVBLICA readers Dan Pangburn and Harold Vandenburg are all in that spirit.  Gentlemen:  Well done.

8 comments:

  1. Consvltvs, I seem still unable to post. I have a response to Dan.
    Dan, thanks for the detailed reply, it is good to be able to discuss in a reasonable manner.
    Your equation fits a line to the data, so far so good, but for it to be meaningful and useful, the inputs have to be realistic. You have said

    "This model predicted the ongoing temperature decline trend. None of the 20 or so models that the IPCC uses do"

    In fact, the current temperatures are within the IPCC predictions, albeit at the low end of the range. The temperature data has been presented to independent statistitians without them knowing what it was, and they did not find that the trend had become negative i.e. there is at present no evidence for cooling. This does not rule out cooling, as that could be occuring but has not got through to the long term averages yet.

    So we are asked to believe that your model is superior to the 20 or so IPCC ones, even though both fit the data.

    Getting a line to fit data is not in itself proof that the equation represents reality, as you are well aware. I have some problems with accepting that your simple equation has better predictive ability those of the climatologists.

    A couple of specific problems I have.

    1) The sunspot integral is a proxy for "energy retained by the planet". Whilst sunspots do have a correlation, I doubt that the sunspot number actually incorporates all the factors that go into the energy retained. One glaring ommission is aerosols from volcanoes, for example Pinatubo. These are known to have global cooling effects. Similarly, man made aerosols, particularly sulphates, have a cooling effect, and in most models are responsible for the cooling after 1945. The absence from your model is a problem.

    2) ESST. I am still a bit unclear how this is derived. It is a smoothing of the SST data, and is a steady up and down with a period of 34 years. I am not convinced that such a global periodicity is sufficiently demonstrated - perhaps you could point me to the source. If you substituted some other thing with a 34 year period, say a planetary orbital period like Saturn, would you not obtain the same correlation? Now clearly the orbit of Saturn has no influence on climate, so for your model to be valid, there must be a mechanism for it to work, otherwise it is merely a correlation. Now the ocean stores massive amounts of energy, so there clearly is a potential mechanism, but does the energy balance out? The energy stored by the ocean in the cooling phase must be shown to be equivalent to the energy lost by the cooling surface, otherwise all you have is a correlation. These mechanisms must be elucidated before we can say we have a definive model.

    3) Feedbacks. These are the big unknowns. Your model has no feedback mechanisms. The change in cloud cover cannot respond to any changes in temperature, and is assumed to be totally defined by the sunspot number. This is unrealistic.

    In short, I do not find this equation convincing. It looks to be a fitting of the data to a series of proxies, which do not entirely describe the planetary climate. In contrast, the IPCC models attempt to include all known influences, and attempt to demonstate that the mechanisms (ocean currents etc.) are reasonable and fit with observations. They do not always succeed completely, which is why they are still being developed.
    Regards,
    Harold

    ReplyDelete
  2. I don't think there is anything about climate change that stops a reasonably intelligent non-scientist from following the debate. As it happens I am an industrial chemist with an environmental science degree, but I don't think I need to draw on that expertise to comment usefully.

    In a nutshell, global climate change is real and is caused by humans. This is basically obvious. If you had a sensitive enough way of measuring it, lighting a single fire would be enough to cause some level of climate change.

    The interesting question is how big the effect is and what we should do about it. More and more sophisticated models won't necessarily help with decision making. Even if a perfect one could be produced today, somebody like me could invent a technology that invalidated it tomorrow.

    It seems that the key is not science, but economics. If climate change is fairly mild and takes a while to happen, in all probability the free market will enable people to adapt to it. It might even end up being beneficial. Likewise, all that CO2 in the atmosphere is actually resource waiting to be exploited. And it is not remotely hard to think of ways to exploit it. The simplest is simply to plant trees. I have several more sophisticated ideas, but that is where I DO have an advantages as a scientist.

    Nigel Lawson and Bjorn Lomborg have both written very respectable books exploring these issues.

    The other aspect is not so much climate change in absolute terms, but whether or not climate change is causing greater variability in the climate. If the climate is becoming less predictable that could undermine economic activity a lot more than a quite a large rise in temperature.

    My personal prescription is that we can't do a huge amount about this problem, but that slowing down the process by cutting CO2 emissions is likely to help us cope with it. If we can keep the rate of change slow, we can probably handle it.

    ReplyDelete
  3. Here's my most concise summary for being a skeptic:

    First, what temperature data are you looking at? Is it satellite, ground, or water temperature? How do you aggregate that into a single "global temperature" with which to compare temperature trends over centuries? Where does the temperature data come from for earlier centuries?

    Next, how do you explain warming periods in the past when the world was not industrialized? If it's not by human activity, isn't it reasonable to think that current warming is also not because of human activity? We have not always had ice at the poles, so it's not all that unusual for it to melt. Why is the current state of the ice at the poles "the ideal" and why do we have to worry about current melting?

    Lastly, the science usually referred to is actually not science, but climate models. Every prediction is based on a computer model with certain assumptions developed by researchers. We are to assume that a computer model today can predict catastrophe 50 years from now, but no one can point to a single computer model that has been back-tested 50 years and accurately predicts today's "global temperature." It doesn't exist.

    Basing public policy, which would hurt the economies of the world, on computer models fed by assumptions (i.e. faith) from climate researchers who are dependent on governments and corporations for their funding seems unwise to me.

    ReplyDelete
  4. I sympathise with what you are saying Heathen Republican, but I think you might be going a bit further than I would. First paragraph - yes measuring the temperature of anything is a lot tougher than most people realise. Even getting the temperature of a beaker of water right isn't that straight forward in reality. But if you want to do anything you have to do some measurements.

    We don't know how sensitive the climate is to human activity. It could well be that pre-Industrial humans' forest clearances and eating large mammals might have been enough.

    As for climate models, they are weak evidence at the best of times and I think will be discredited completely before long. But it is easy to get carried away with them. If you want to know what the temperature in say Bognor Regis is going to be for your holiday next August, the best estimate is the temperature last August. For the purpose you need the estimate, that climate model is perfectly adequate. If you wanted to build a greenhouse in Bognor Regis you might want a bit more data. But even then you chose the model to solve the problem - there isn't a 'right' or 'wrong' model, and fiddling with details of a particular model is not being dishonest if you have a clear idea of what you are doing and why you are doing it. I don't think the very elaborate climate models actually help decision making very much. We can't make accurate predictions all we can do is get a feel for what is going on and make a gut decision.

    All scientists are dependent on their funding sources - I am one so I know that very well. But so is every other specialist. I don't think climate scientists are either any more or any less prone to this. And I think it is an economic question more than a scientific one, and economics is just as prone to fashions, vested interest and unjustifiable assumptions as any other branch of knowledge.

    But the planet could still be warming up and it still might be very sensible to do something about it. On balance I think this is probably the case.

    ReplyDelete
  5. Harold,
    “…fits a line to the data…” is misleading. A line fit (such as a regression analysis as used by statisticians) would have no predictive ability. My equation is based on the physical phenomena. The coefficient ‘a’ is just an offset, much like the derived constant (the one that is not multiplied by the independent variable) in a linear regression, it merely shifts vertically the location of the curve on the chart without changing its shape. The influence of CO2 is either discovered (by finding the maximum R2) or selected by the value of ‘d’. The influence of CO2 is eliminated by setting ‘d’ to zero. Note that assuming that CO2 has no influence has no significant effect on the coefficient of determination (R2). The Effective Sea Surface Temperature oscillation range is given by ‘c’ and is between 0.34 and 0.39. That leaves only ‘b’ the factor that indicates the influence of the sunspot number (a proxy for energy retained by the planet), to be determined. The result is an equation that calculates temperatures for over 115 years with a coefficient of determination of 0.88. You may or may not know that that is better than anyone else has done.

    The equation explains 88% of the temperatures. That means that only 12% is available to account for the effect of aerosols, feedbacks and everything else whether it is known or not. Aerosols/volcanoes have a comparatively brief transient effect that is spread out over time. Other factors simply don’t have much effect on agt.

    ESST is discussed further under ‘Measurement Artifacts’ starting on page 3 of the pdf made public 9/24/11 at http://climaterealists.com/index.php?tid=145&linkbox=true. The 32 year duration (64 year period) of the trends was estimated by observation of the temperature trajectory and also by a study of the PDO. The magnitude of the ESST was initially estimated to be 0.45 (assumed no influence from CO2) as described starting on page 14 of the pdf made public 4/10/10. This was later refined to 0.343 if CO2 was assumed to have influence and 0.385 if CO2 was assumed to have no influence.

    “…does the energy balance out?” Yes. The first law of thermodynamics is imposed which accounts for all energy to and from the planet. Any change in the energy content of the planet appears as change to agt. There is no net energy change associated with ESST. The influence of CO2 is not significant.

    ReplyDelete
  6. http://www.washingtonpost.com/opinions/the-scientific-finding-that-settles-the-climate-change-debate/2011/03/01/gIQAd6QfDM_story.html

    This week's "big news" is fairly compelling. A long-time skeptic and critic of climate change science, funded in part by Koch brothers money, takes the largest data set ever assembled and concludes global warming is real.

    Although he does not address the subsidiary issue concerning the human contribution -- is it human caused global warming -- the study completely demolishes the notion that climate scientists did bad science. Murray entered with that preconception, and found that previous studies came to the correct conclusions. We cannot claim the science is bad, or that the climate scientists have skewed the results.

    Logic would seem to indicate, if global warming is real, even if it is a natural oscillation in a long-term climate cycle and not 100% human caused, there are human inputs to the cycle which probably should be mitigated. The likelihood is, our contribution to the warming effect is significant. But we do not need to quantify that exactly to know it's important not to make it any worse.

    My own take is that even if global warming were non-existent, poisoning our atmosphere as we do is an extremely bad idea -- my athsma tells me so. So I have long considered the "global warming" debate to be an unfortunate distraction from the larger issue of cleanliness on the planet.

    ReplyDelete
  7. Excuse me, the reformed skeptic physicist's name is Muller not Murray. Sheesh.

    ReplyDelete
  8. GTChristie,
    Muller showed that the temperature measurements were right. He did not address whether the Consensus’ science was bad. It is bad. Muller’s paper was misleading as shown more recently at http://news.heartland.org/newspaper-article/2011/10/31/muller-backtracks-temperature-claims

    The pdf made public 9/24/11 at http://climaterealists.com/index.php?tid=145&linkbox=true shows the temperature measurements for the last decade and they are flat. It also shows an equation based on the physical phenomena involved, with inputs of only sunspot numbers and ppmv CO2, that calculates the average global temperatures (agt) since 1895 with 88.4% accuracy (87.9% if CO2 is assumed to have no influence). CO2 is not pollution.

    ReplyDelete