Moments of Revelation

Dr Iain Stewart holding a rock

Dr Iain Stewart holding a rock

Over the past few days I’ve worked my way through the three-part BBC series, Climate Wars, hosted by Dr Iain Stewart, a geology professor with a very cool Scottish accent. An excerpt from this series was featured in one of Peter Sinclair’s videos, which looked quite fascinating, and anything Peter refers to as “brilliant” is probably worth watching.

Worth watching indeed. I’d recommend anyone and everyone to watch this series. It’s basic enough for someone with little to no knowledge of this issue, yet presented in such a compelling way that the most experienced climate scientist wouldn’t get bored.

One of the film’s major strong points was simply the way it was organized. Dr Stewart traced the history of both the science and the politics around climate change, splitting it into three parts:

Part one: Scientists had known for decades that anthropogenic greenhouse gases could cause warming of the Earth, but now, following thirty years of aerosol-induced cooling, global warming was starting to show; almost every year was record-breaking. James Hansen was the first to “stick his neck out” – testifying to Congress that he believed anthropogenic climate change was underway. He later claimed that he had weighed the risks of being wrong and looking stupid, versus doing nothing and not telling the world about such a huge potential threat. Sort of like an early Greg Craven, I suppose. I found this part to be the least interesting of the three. It also began strangely – Stewart mentioned a letter to the US president, signed by top scientists, which warned of an impending ice age. I’d never heard about this before. Does anyone else know more about this letter?

Part two: The skeptics fought back as strongly as they could, questioning absolutely every scientific claim regarding global warming. I found this to be absolutely fascinating; it solidifed a lot of issues in my mind and helped to unify my knowledge on the topic. Stewart went through the research which showed that the Earth was warming as a result of human activities – and showed how all the yelling from skeptics helped to make the theory even stronger. He also “infiltrated the walls” of the Heartland Institute’s International Conference on Climate Change, which I found to be absolutely hilarious. They had a comedian making bad jokes about how New York could handle some global warming, Monckton and Singer making their usual accusations of fraud (Stewart remarked that “when these become the talking points, then I know that the scientific debate is really over”), and Patrick Michaels publicly admitting “Yes, the second half of the century did show some warming, and it was the result of human activities…..and now you all hate me for saying that…….” Dr Iain Stewart explained that, even though the controversy doesn’t really exist anymore in the scientific literature, the claims of skeptics still live on in the popular media and on the Internet. Instead of fighting a scientific battle, they’re now doing public relations.

Part three: Scientists knew that humans were causing global warming, but how bad would it be? After the brilliance of the second part, I wasn’t expecting to enjoy the last segment quite as much…….but I was proven very, very wrong. It both terrified and fascinated me. Terrified because it discussed the Younger Dryas, something I hadn’t really heard of before, where it warmed about 5 C in just a few years. So far beyond anything I thought was possible. When this research was released, the idea that the climate was steady and slow-moving could no longer be embraced.

And then it fascinated me because it was the first time that climate models seemed really, really cool.

The idea of modelling something – anything – on the computer is somewhat unremarkable to me. I am of the generation that literally grew up using computers; I vaguely remember playing astronaut addition games on Windows 3.1 when I was four. I have seen so many things digitalized; the prospect of modelling climate is obviously immense, but it doesn’t amaze me.

But then Dr Stewart made a “dishpan climate model” with a spinning bowl, water with some dye, an ice-cube Antarctica, and a Bunsen-burner Sun. He set it all up and before long… could actually see regular patterns in the water’s movements that looked like the prevailing winds. It was so, so amazing. Even more amazing than a complex model on the computer because it was real and tangible and you could touch it. Like a little Earth on the countertop. All of the complex processes of our climate eventually come back to these simple factors. (I want to make one myself. But I don’t have one of those spinny things.)

And then I started wondering what computer modelling would be like, and remembering how much I loved physics last year, how I liked to put four or five algebraic equations together and solve it all in one complicated step to reduce error. Manipulating variables and shifting things around. Like a little puzzle. I was remembering how much I love hard math problems, because you actually have to use your brain, try everything you can think of, stretch the limits of your logic…..and you feel such a sense of accomplishment when you finish that all the work is worth it.

Is a climate model just a really large and complex collection of equations and puzzles that have to fit together in the right way? It would be pretty cool if it was. I knew that studying climate change required a lot of math, but this is the first time that I can see a clear path showing how an issue I care deeply about could coincide with aptitudes I enjoy.


38 thoughts on “Moments of Revelation

  1. “Is a climate model just a really large and complex collection of equations and puzzles that have to fit together in the right way?”

    I am not a climate modeler. But it must be. However, there’s an issue.

    It seems that what you like is problem-solving produced through mathematical insight. However, realistic models are going to be extremely complicated. Furthermore, computer models can assume an oracular quality, whereby simulation replaces understanding: You don’t understand the world, and you don’t understand the program either, but you trust that the structure of the program reflects the structure of the world.

    Of course, professional climate modelers are not just attending to oracles which they do not understand. The computer models are based on physical models; they wrote the code and understand how it works and why it produces the output it does; and the models do get tested in all sorts of ways. Nonetheless, it is inconceivable to me that the behavior of their models would always be conceptually transparent. Something as complicated as a spatial model of world climate is guaranteed to do unexpected things, even if you do eventually figure out the logic of what’s happening. (I hope the professionals who read this blog will speak up about this, and relate some anecdotes.)

    So while I’m sure it’s a rush to build a computer model of the world, and I wouldn’t want to discourage anyone from having that experience, the path to real and secure understanding must begin with “models” simple enough that you can solve them with pen and paper, and with facts that are similarly easy to perceive. Maybe there’s something matching this description in Raymond Pierrehumbert’s web textbook.

    I have always thought that, if I were to be a mathematical climatologist, I would want to focus on the ice-age cycles. There’s some beautiful data, it’s central to how climate sensitivity has been estimated (because the temperature change directly induced by the Milankovitch cycles is much less than the full change from glacial minimum to interglacial maximum, so something is amplifying it, and carbon dioxide and methane are the obvious contenders), and yet there are subtleties not understood, because the cycles are *not* just mechanically repetitive. Frequency analysis shows that there are rhythms in the cycles corresponding to the Milankovitch influences, but the exact cause and effect behind individual cycles is not known, though there is much work and much speculation.

    I guess the other thing I would emphasize is that the exact mechanism whereby the greenhouse gases amplify these temperature shifts, to the degree that they do, seems not to be known either. If you take the GHG time series from the ice core data, and assume particular warming potentials, you get a very good fit to the actual temperature curve. However, these curve-fitting warming potentials exceed the warming directly induced by the gases in question. That is not in itself a novelty. The standard value for climate sensitivity is 3 degrees, but only 1 degree of that is directly due to CO2; the rest comes from feedbacks like humidity.

    So to sum up, my personal climatological research agenda would be to produce physical models accounting for (i) the exact timing of the ice ages for the past several million years, and (ii) the exact nature of the feedbacks responsible for giving CO2 and CH4 the global warming potentials they are inferred to have. Obviously the existing literature addresses both these topics, but I’ve never reviewed it comprehensively.

    Please pardon the personal digression there. Maybe I’ll take it up at my own blog! :-)

  2. For the past few weeks, Belgian television was airing another program of Stewart which i found worthwile watching too : if i remember well it’s called “ring of fire” and Stewart is following the sides of the tectonic plates around the Pacific Ocean to see how people are coping with living in the approximity of geological dangers as volcanoes & earthquackes, high mountains, etc.
    Above that, he does an excellent job explaining how maybe the environment people are living in, has an impact on their entire culture and way of thinking. Geology determing psychology. Maybe it’s because i’m a geologist too, but it’s a subject i’ve always found very fascinating and one that is often forgotten by people when they are thinking about history.

    I didn’t know he had made a series on climate change, gotta urge Belgian television to air it !

  3. I’ve been meaning to watch it for a while; thanks for the endorsement.

    For the record, Mitchell’s right. When you say:

    Is a climate model just a really large and complex collection of equations and puzzles that have to fit together in the right way?

    You hit it on the head.

    In fact, Tamino built a very crude climate model step-by-step here. The rest is more or less adding in other regularities that are observed but aren’t captured by the general-form model there. Over time, the resolution gets better and better (i.e. we can better model regional effects, while the basic one in that link is global only) and the unknowns slowly get weeded out (things like feedbacks are still often left out, sadly).

    The physics behind Tamino’s basic model are covered in first and second year thermal physics classes, by and large, although they remain fundamental to more advanced work as well. The math won’t make much sense without calculus (which I assume you haven’t taken yet), but I can assure you that there’s nothing in there a first-year student couldn’t do after one semester of calculus. (Aside: The sigma*T^4 term is so common in thermal physics that a friend of mine once made a Halloween costume based on it – black bodysuit with sigma*T^4 on the chest with lots of red pipe cleaners bent into waves shooting off randomly. She went as blackbody radiation, yessee…)

    If you’re asking what climate model output looks like, Coby Beck’s got ya covered, too.

    By the way:

    And then I started wondering what computer modelling would be like, and remembering how much I loved physics last year, how I liked to put four or five algebraic equations together and solve it all in one complicated step to reduce error. Manipulating variables and shifting things around. Like a little puzzle. I was remembering how much I love hard math problems, because you actually have to use your brain, try everything you can think of, stretch the limits of your logic…..and you feel such a sense of accomplishment when you finish that all the work is worth it.

    That’s a good sign. Look for any opportunity to take calculus, though (it’s usually offered in pre-calculus form at high schools). The transition from algebra-only math to calculus (and more advanced algebra, such as vector and matrix operations at the most basic form) is literally like suddenly having a light come on, and you realize just how massive science and mathematics can be.

    • Yes, I’ve finished my high school pre-cal and am starting calculus in just a few weeks (sort of an accelerated program where you can get your first university credit while still in high school). I loved pre-cal, although I found it too easy after the first year. I’m hoping that calculus will be more challenging but still enjoyable.

  4. Oh, I forgot to mention: I’d seen some criticism of Stewart from a source I respect (Connolley – doubly so because it’s on the 70s cooling scare, which is an issue I completely trust him on), so take care about credibility.

    Note that Connolley also found mention of that 1972 letter; I haven’t found a link to it yet.

    • Yeah, I thought that the ice age claim sounded a little exaggerated. An agreement that it was cooling, yes. But a consensus about an actual “ice age”? Not so much…

  5. Several climate models have their code available including NASA’s GISS Model E.

    There’s also MagicC that you can run on a home computer.

  6. [Citation needed regarding “ice age consensus was a myth” – but finding such a citation is so easy that I’ll just fill it in for William.]

    I did email Stewart to point this out; his response was that he was currently busy; that others had also pointed out that he was wrong; and that he would get back to me. But he never did. So I’ve sent him a reminder now. My best guess would be that admitting error is too embarassing and that public silence will be the easier option.

  7. “Is a climate model just a really large and complex collection of equations and puzzles that have to fit together in the right way?”

    It’s a bit more than that, I’d think. The climate model has to be physically based. Absent of this physical grounding one can puzzle over collections of equations to derive any kind of nonsense. Anthony Watts illustrates this point on a regular basis on his blog, where he often find new mathematical ways to show urban heat island effects and correlations with sun spots, free of any requirement to justify his equations with actual physically understood mechanisms.

  8. Gavin Schmidt made some interesting comments on climate modeling during an Edge interview. Here are some snippets:

    it turns out that the assumptions we make in building the models (the slightly different decisions about what is important and what isn’t important) have an important effect on the sensitivity of very complex elements of the climate.

    It turns out that the average of these twenty models is a better model than any one of the twenty models. It better predicts the seasonal cycle of rainfall; it better predicts surface air temperatures; it better predicts cloudiness. This is odd because these aren’t random models. You can’t rely on the central limit theorem to demonstrate that their average must be the best predictor, because these are not twenty random samples of all possible climate models.

    The problem with climate prediction and projections going out to 2030 and 2050 is that we don’t anticipate that they can be tested in the way you can test a weather forecast. It takes about 20 years to evaluate because there is so much unforced variability in the system — the chaotic component of the climate system — that is not predictable beyond two weeks, even theoretically. This is something we can’t really get a handle on. We can only look at the climate problem once we have had a long enough time for that chaotic noise to be washed out, so that we can see that there is a full signal that is significantly larger than the inter-annual or the inter-decadal variability. This is a real problem because society wants answers from us and won’t wait 20 years.

    At the same time, all these models are plagued with uncertainty and have the problem of changing measurements. The data collecting is always improving and our understanding of different processes is always growing. But that doesn’t make things simpler; it makes things more complex. It means you need to add another element to your model. It means you have to measure everything again and run the model again.

  9. cj99, you conveniently left out the following paragraph (and more) just before your final quoted paragraph:

    We did this [climate modeling] 20 years ago and the predictions that we made then have been more or less validated, given both the imperfections we had at the time and the uncertainty in how we thought things would change in the future. So there is a track record that shows that these models are realistic.

    (emphasis mine)


  10. Yup. As frankbi points out, cj99 illustrates right here one of the main things you’ve always got to check for — the possibility that someone is giving you a superficially sensible response that uses sneaky, out-of-context quoting and fails to show omissions properly with an ellipsis. This is — at best — unacceptable and deserving of an F grade; it’s often done to mislead the reader. That was a good example.

  11. The quoting of Schmidt was to point to some of the issues of building computer models following the discussion above of what models are. I linked to the interview because that was where the snippets came from, and, following the guidelines of this blog in terms of referencing.

    For any computer model the key issues/questions are what is left out (because you can’t model everything) and what are the key assumptions/approximations? As Jo Weizenbaum argued 25 years ago:

    What is important in the present context is that models embody only the essential features of whatever it is they are intended to represent. … What aspects of reality are and what are not embodied in a model is entirely a function of the model builder’s purpose. But no matter what the purpose, a model, and here I am concerned with computer models of aspects of reality, must necessarily leave out almost everything that is actually present in the real thing. Whoever knows and appreciates this fact, and keeps it in mind while teaching students about the use of computers, has a chance to immunize his or her students against believing or making excessive claims for much of their computer work.
    (Weizenbaum, J. (1984). Computer Power and Human reason. From Judgement to Calculation. Harmondsworth, Middlesex: Penguin, p. xvii)

    Research, to me, is always about asking better questions and the Schmidt interview gave a good broad mapping of many of the key issues facing the modeling of climate.

    What I think is impressive is that the models produce coherent outcomes given the issues with measurements that Schmidt points to. What is equally intriguing is that it is the average of these models that does the best job. A recent publication explores this issue a little further and does some interesting work in trying to assess which models or set of models do the best job in terms of climate projections. An indication of the complexity of the problem is flagged by the authors (p. 4):

    In our analysis there is no evidence of future prediction skill delivered by past performance-based model selection. There seems to be little persistence in relative model skill, as illustrated by the percentage turnover in Figure 3. We speculate that the cause of this behavior is the non-stationarity of climate feedback strengths. Models that respond accurately in one period are likely to have the correct feedback strength at that time. However, the feedback strength and forcing is not stationary, favoring no particular model or groups of models consistently. For example, one could imagine that in certain time periods the sea-ice albedo feedback is more important favoring those models that simulate sea-ice well. In another period, El Nino may be the dominant mode, favoring those models that capture tropical climate better. On average all models have a significant signal to contribute.

    And, Hank, an ellipsis is used to “indicate an intentional omission of a word or a phrase from the original text”. I intentionally used the term snippets to indicate that I had selected some paragraphs that were of interest. Perhaps, in hindsight, the word selection would have been more apt.

  12. cj99:

    I intentionally used the term snippets to indicate that I had selected some paragraphs that were of inter/est. Perhaps, in hindsight, the word selection would have been more apt.

    Sorry, that excuse doesn’t fly. You didn’t even bother to separate out the paragraphs using “—” or “***” markers or anything of that sort. You’re just trying to mislead. [I’m not so sure that he’s trying to mislead, if he is he’s not doing a great job, as it seemed to me to just be an interesting musing about how climate models work. But I’m new at all this. -Kate]

    As Jo Weizenbaum argued 25 years ago: […] A recent publication explores this issue a little further […]

    Talking points and insinuations lifted from Anthony Watts’s denialist blog. Fail.

    * * *

    To others:

    It seems that climate cranks like to trot out the Weizenbaum quote, out of any useful context, in support of their crankery. I can’t find that quote yet, but I’m guessing that Weizenbaum was talking about abstraction, not approximation.


  13. Yes, computer models have their problems… so have non-computer models. Especially the “model” that says that “surely us little humans cannot possibly change something as big as the climate”.

    All models fall short in complexity and detail compared to the real world. This “model” has none. Furthermore, it makes no useful predictions on paleoclimate behaviour, on response to volcanic explosions; the only useful prediction it makes, for the last few decades, cannot be falsified because — uncertainty bounds? What uncertainty bounds?

    So, when someone again claims that “the models are no good”, you can at least point out that even the lousiest ones beat this “incredulity model” hands down…

  14. Trouble with folks like cj99 posting at places like this is: people follow the links and don’t blindly take what is implied/stated/inferred as true/right.

    That works at partisan sites as such…er…”snippets” reaffirm confirmation bias for those who need it.

    Martin wrote:

    So, when someone again claims that “the models are no good”, you can at least point out that even the lousiest ones beat this “incredulity model” hands down…

    That’s a good line.

    And we also don’t get a ‘control run’ or a ‘re-run’ in the reality model of the very large earth model.



  15. Climate Wars was a very good series. I think I watched it last year??

    I doubt the Youtube version is legal??
    My TV licence is paid for that, i’m surprised the BBC hasn’t had it taken down.

  16. [Gord cites “Falsification Of The Atmospheric CO2 Greenhouse Effects Within The Frame Of Physics” from the International Journal of Modern Physics B to explain why the greenhouse effect does not exist.

    This document seems sketchy, and I highly suspect that it is a common talking point among WUWT et al; however, I can’t figure out whether or not there are adequate grounds to deem it “discredited”. Could someone with more experience in peer review, background knowledge of this study, access to ISI Web of Science, etc give their two cents before we allow Gord to accuse Arrhenius and Fourier of fraud? -Kate]

  17. Citation needed – that heat energy can flow from cold to hot as per Trenberth’s Energy Budget Diagram.
    “Second Law of Thermodynamics: It is not possible for heat to flow from a colder body to a warmer body without any work having been done to accomplish this flow. Energy will not flow spontaneously from a low temperature object to a higher temperature object.”

    [Gord, you don’t need to cite physics laws, those are common knowledge (the full comment policy is clearly in the sidebar). What you do have to cite, with legitimate peer-reviewed sources, are claims that climate change is natural/nonexistent/a global conspiracy. Those are certainly not pieces of common knowledge. -Kate]

  18. [Gord, I see what you’re doing – I know you like to find loopholes in the comment policy, but I’m not stupid. -Kate]

  19. Cosmic Ray Decreases Affect Atmospheric Aerosols And Clouds

    “The effect of the solar explosions on the Earth’s cloudiness is huge,” Henrik Svensmark comments. “A loss of clouds of 4 or 5 per cent may not sound very much, but it briefly increases the sunlight rea-ching the oceans by about 2 watt per square metre, and that’s equivalent to all the global warming dur-ing the 20th Century.”

    Journal reference:
    1.Svensmark et al. Cosmic ray decreases affect atmospheric aerosols and clouds. Geophysical Research Letters, 2009; 36 (15): L15101 DOI: 10.1029/2009GL038429

    [This certainly shows that there is a link, but nowhere does the article say that this decrease of 4-5% is actually happening. If it was, the stratosphere wouldn’t be cooling. -Kate]

  20. Strongly recommend watching and listening to this excellent presentation by Prof John Abrahams that pulls apart Moncktons misrepresentation of science:

    Abrahams goes through many of Moncktons presentation slides and cuts them to pieces.

    Haven’t seen it yet but it looks interesting….I heard about it from Peter Sinclair. Will sit down and watch it when I have some time! -Kate

Leave a Reply to Dano Cancel reply

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

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

Facebook photo

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

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.