When I was in high-school in England in the 1930s, we learned that continents had been drifting according to the evidence collected by Wegener. It was a great mystery to understand how this happened, but not much doubt that it happened. So it came as a surprise to me later to learn that there had been a consensus against Wegener. If there was a consensus, it was among a small group of experts rather than among the broader public. I think that the situation today with global warming is similar. Among my friends, I do not find much of a consensus. Most of us are sceptical and do not pretend to be experts. My impression is that the experts are deluded because they have been studying the details of climate models for 30 years and they come to believe the models are real. After 30 years they lose the ability to think outside the models. And it is normal for experts in a narrow area to think alike and develop a settled dogma. The dogma is sometimes right and sometimes wrong. In astronomy this happens all the time, and it is great fun to see new observations that prove the old dogmas wrong.
Unfortunately things are different in climate science because the arguments have become heavily politicised. To say that the dogmas are wrong has become politically incorrect. As a result, the media generally exaggerate the degree of consensus and also exaggerate the importance of the questions.
It’s not a new interview, but if anything, it’s even more true now than then. The “consensus” has broken down considerably in the interim.
My impression is that the experts are deluded because they have been studying the details of climate models for 30 years and they come to believe the models are real. After 30 years they lose the ability to think outside the models.
My impression is they have made a living on doing this for 30 years. They fear admitting that they really don’t know or that the variations are within the margin of error. If they admitted those things, then their job would be done. Who would care if scientist could predict in 30 years that global temperatures will be somewhere between -.4 to .6 of what we experience today? It’s like saying I predict there will be a Superbowl winner; not the winner but a winner. That’s right, it only took me 49 years of careful study, but I think I can safely predict with superior precision that there will indeed be a winner.
Then again, part of their problem is having spent 30 years to only predict which way the trend might go in another 30 years. I would be impressed with the models of they could actually predict year to year when the variations would occur. Knowing which years would be colder or warmer could serve many purposes.
However, another reason I’m skeptical are the yearly predictions of worse hurricane season yet to come with the result being hardly any named storms. Further, a hurricane in terms of intensity is not measured by the dollar value of insurance claims. Telling me hurricanes get costlier as time goes by does not support rational arguments of increased climate change.
Dyson nailed the exact problem in the part Leland quoted. You don’t ‘study’ models, you build models to test your understanding of the real thing being studied. A model is always just an abstraction of the thing modeled. It’s parameters can never be exact and usually quite sensitive to small changes as well as leaving things out that will also affect outcomes.
A good model is predictive but even then doesn’t mean one day it will not be wildly wrong. Models simply are never the real thing. That’s true of simple models. The weather is not simple.
Dyson is my kind of wacko. He knew with absolute certainty (and he was right) that we could have sent a small city to saturn 4 decades ago… and everywhere else in our solar system by now. Since then even better designs have been offered (I like Zubrin’s torch ship.)
Imagine how different life would be today if some of his brilliance had been realized half a century ago?
You don’t ‘study’ models, you build models to test your understanding of the real thing being studied.
Actually, that is the traditional role of applied mathematics – to study the implications of a well-posed model. It can give you both interesting novel predictions of a model and interesting ways the model fails to work.
For example, a common model of single fluid flow are the Navier-Stokes equations. Various terms of equations relate to various things, like conservation laws and physic properties of fluids like diffusion and viscosity. It also has a famous way of breaking, when one sets viscosity and compressibility to zero, getting an “ideal Euler fluid”. At that point, one gets a fluid with potentially infinite energy hiding in arbitrarily small perturbations (a big deviation from the real world fluids of this sort!) which gets expressed as larger and larger scale vortices. The ideal model may be so broken that it doesn’t have a solution pasts finite time. We don’t know yet if that is true, hence, we have grounds for studying this particular model beyond its applications to the real world.
Karl, when do I get my home thorium reactor? I want one!!!