Counting Cats in Zanzibar Rotating Header Image

Modeling sanity

The Hawaii Reporter joins the growing chorus of dissent towards the consensus view, with comments about the limits of the computer models in The Salesmans settled science -

as P.J. O’Rourke noted, there are a lot of people who would do anything to “save the planet”, except take a science course. Chief amongs these being Government ministers, journalists and environmantalists ?

Computer models do not consider variations of irradiance and magnetic fields of the sun

Computer models do not accurately model the role of clouds

Computer models do not simulate a possible negative feedback from water vapor

Computer models do not explain many features of the Earth’s observed climate.

Computer models cannot produce reliable predictions of regional climate change

We must conclude that climate models used by the Intergovernmental Panel on Climate Change (IPCC), do not accurately depict our chaotic, open-ended, climate system. They cannot make reliable predictions and should not be used to formulate government policy.

As Christopher Monckton recently admonished, “We must get the science right, or we will get the policy wrong”.


As Dennis Avery recently said, “Let’s have a real debate of the climate evidence. We’ve heard enough from the computers.”

Garbage in, garbage out, although garbage apparently validated by having passed through the bowels of a multi megaflop monster.


  1. NickM says:

    Look, it really is this simple. Computer models are useful tools but they are not a replacement for experiments. In order to prove anything you need experimental data, not computer data.

    For reasons that are probably historical computational physics is done essentially by theorists and that’s fine as long as it remains a theoretical calculation. It goes horribly wrong when it is seen as an ersatz experiment.

    OK, the AGW theories (yes, plural) should be able to predict (actually post-predict) stuff what happened. They should if they are to have much credibility be able to explain things like the medieval warm period, the Thames’ frost fairs, Roman Vineyards in Yorkshire and that’s just the UK.

    These climatic variations were clearly not the result of driving SUVs but they were dramatic nonetheless. I live on the edge of the Peak District and there are weird stones around on the hills. They’re actually detritus from glaciation (do any of these models explain that?) but the local folk-lore is that they are actually frozen people from the big freezes which turned the Thames to ice. That I even know that hundreds of years afterwards is testament to the impact these events had on our pre-industrial civilization.

    So, do your models explain this stuff. And if not why should I believe them?

    OK, example. In the C19th it was a minor scandal in celestial mechanics that the perihelion of Mercury advanced by 15 arc-seconds (not a lot) a century decribing a spirograph pattern rather than a regular ellipse around the Sun. Now some complete cluster-fuck tinkering with Newtonian Gravity (and I mean things like changing the r-squared from raised to 2 to 2.000000016 or somje such insanity. There are extremely compelling reasons why it should be 2 exactly. They’re called geometry. These folk were engaged in desperate tinkering and got nowhere because meddling and curve-fitting is always engineering and not science. It’s wrestling an air-bed. You shove it down one place and it comes-up somewhere else. It’s the scientific equivalent of patching a computer system.

    In 1915 a Young German fella who later became more famous for having mad hair and shagging Marilyn Monroe did a calculation with this crazy idea he had called General Relativity. And guess what happened? The advance of the Perihelion of Mercury just dropped into his lap. Albert Einstein later said he was beside himself with joy at this. He had the first hint that his entirely theoretical structure had some empircal merit and this was not because it predicted something new that was then discovered (that happened later) but because he had achieved a post-prediction. Mercury’s orbital oddity was down and it was done with an entirely new theory that wasn’t a botch-job that might explain Mercury but fuck-up the Jovian satellites or something. It wasn’t a patch. It was beautiful and brilliant. That is science. Science is not endlessly botching the job it always has to explain more than it sets out to.

    General Relativity also explained gravitational lensing and a few other things and it gave us our creation myth the Friedman-Lemaitre-Robertson-Walker Cosmology which is better known as the Big Bang. Real science is getting a proper result and not shot-gunning the problem until you have a solution which seems to fit until something else comes up. Of course, sometimes you have to shotgun because the problem is pressing. The first ever antibiotic (for Syphlis) was called Salversan 606 because it was compound #606 that they tried. They had bugger-all idea how it worked but I doubt the average sufferer cared and the share-holders made a packet.

    The difference is that the lads at Bayer (I think it was them) didn’t pretend that they’d done anything other than patched the problem. The broad scientific consensus on AGW is far less honest.

    (a) They deliberately conflate models with reality.
    (b) They are wrestling with air-beds and calling success when they’ve only pinned the bottom left-hand corner down.
    (c) I have never heard them ever claim that their theory post-predicts any of the many climatic anomolies we know of. This automatically brings us back to the air-bed. If you have no idea how the mini ice-age of the early modern period happened then why should I believe what you’re saying about life in 2050?
    (d) I have run computer models and unlike the “work it out with a pencil” stuff the capacity to create and not discover is phenomenal. OK, look at it this way. A modern computer game is a mathematical model. Is it a model of reality. Maybe but it doesn’t have to be and frequently isn’t which is why Halo 3 out-sold the “Advanced Washing-up Simulator” by like a billion to one. When I replace my computer I’ll be getting a fairly high-end machine with a Physics Engine in it. That it is a Physics Engine is true. That it has to be used to simulate real physics and not invented stuff is not the case. It is the case because flying star-fighters faster than the speed of light is much more fun than Ride-on Lawnmower Simulator 3. That wielding a Plasma Cannon is more fun than wielding a strimmer is a no-brainer yet they both rely on physical models. But physical models do not have to be realistic in the sense that they actually pertain to the real world. You just have to get a whipping on your 10 year old cousin’s Wii to know that one. There are two classes of folks trying to create models. One are scientists trying to model reality and the other are engineers trying to create realities which are more fun than the one we’re stuck with and somehow I thing there’s bleed-over. This is the reason why video games frequently involve the use of improbable weapons against unlikely creatures and very rarely a trip to Tesco. “Trip to Tesco: 3 for 2″ killed Sega. We can create models of other worlds and things that haven’t been invented (yet?) and this freedom to create can be abused. I mean if you’re convinced we’re gang-banging Gaia and you just modify a couple of constants (the value of which is debatable anyway) and you predict apocalypse then why not? It’s just the same as your pal who’s a games programmer getting around that pesky cosmic speed-limit because otherwise his moral sequel to Elite would play remarkably slowly and the average gamer does not have decades for the routine Sol-Barnard’s Star trading-hop. He or she wants to get there in seconds with a really cool hyper-space sequence.

  2. Sunfish says:

    It is the case because flying star-fighters faster than the speed of light is much more fun than Ride-on Lawnmower Simulator 3.

    When you miss the turn and hit the tomato patch just right, I’d have to say it’s a wash.

    That wielding a Plasma Cannon is more fun than wielding a strimmer is a no-brainer

    My friend, with your luck with powered gardening tools I don’t see them being all that far apart :-B

Leave a Reply

%d bloggers like this: