Friday, 11 January 2008

Mathematical models and their climate change reliability

American Thinker's Jerome J. Schmitt has as much trouble with climate models as I do.

I've asked this question before - If someone built a financial model that was the result of massive back-fitting and had never been accurate in forecasting then would you entrust the nation's economy to its predictions?

Modelling climate is akin to modelling the human mind and trying to work out what someone is thinking in 2050. This stuff is complex.

I just don't understand American liberals and their attitude toward mathematical models. The left places an inordinate amount of faith in untested models predicting man-made warming of the global climate, while ignoring time-tested mathematical models in another important field important to all Americans.

Embracing Untested Models

Global warmists place unquestioning faith mathematical models of anthropogenic global warming, declaring their predictions to be settled science, as if the models completely and flawlessly took account of all significant variables in one of the most complex systems ever studied, the earth's atmosphere.

Yet experience teaches us that modeling atmospheric phenomena is an uncertain business, even in atmospheres much less complex than the planetary system. Far simpler artificial atmospheres or "climates" are routinely created and modeled in the semiconductor industry. Working in sealed vessels within which chemical vapor deposition (CVD) forms nanometer-scale thin solid films on silicon wafer surfaces, in order to produce integrated circuits and many other semiconductor products. Last year I wrote about the many difficulties these simple atmospheric models encounter:
Closed systems are also much easier to model as compared to systems open to the atmosphere (that should tell us something already). Computer models are used to inform the engineering team as the design the shape, temperature ramp, flow rates, etc, etc, (i.e. the thermodynamics) of the new reactor.

Nonetheless, despite the fact that 1) the chemical reactions are highly studied, 2) there exists extensive experience with similar reactors, much of it recorded in the open literature, 3) the input gases and materials are of high and known purity, and 4) the process is controlled with incredible precision, the predictions of the models are often wrong, requiring that the reactor be adjusted empirically to produce the desired product with quality and reliability.

The fact that these artificial "climates" are closed systems far simpler than the global climate, have the advantage of the experimental method, and are subject to precise controls, and yet are frequently wrong, should lend some humility to those who make grand predictions about the future of the earth's atmosphere.

So serious are the problems, sometimes, that it is not unheard of for an experimental reactor to be scrapped entirely in favor of starting from scratch in designing the process and equipment. Often a design adjustment predicted to improve performance actually does the opposite. This does not mean that process models are useless, for they undergird the engineer's understanding of what is happening in the process and help him or her make adjustments to fix the problem. But it means that they cannot be relied upon by themselves to predict results. These new adjustments and related information are then used to improve the models for future use in a step by step process tested time and again against experimental reality.
Although there is no record of accurate prediction for the totally unproven and skepticism-worthy models of the warmist climatologists, liberals are convinced that there is a "scientific consensus" over an impending Anthropogenic Global Warming Crisis that requires undertaking immediate extreme measures including increased carbon taxes as well as new "private sector" solutions such as carbon-trading and selling of carbon-credits.

Ignoring Time-Tested Models

Actuaries are mathematical professionals who build statistical models of human populations in order to plan for life-insurance, health-insurance and pension benefits. The accuracy of their models can have far reaching consequences. Errors might cost their private sector employers' millions and even billions of dollars. Fortunately, the actuarial profession's record in accurate prediction is quite good, particularly since their relatively simple statistical models are informed with reliable data collected from census figures as well as hospital and mortuary records. While predicting one individual's healthcare needs and time-of-death is impossible, when averaged across millions of people, such statistics can be quite reliable.

Today, a consensus of actuaries agrees that the Social Security system is in need of major overhaul otherwise it will experience debilitating financial shortfalls in a few decades as more and more "baby-boomers" retire with full benefits. In the face of this consensus however, liberals demur and minimize the importance of this crisis, undercutting President Bush's attempts to draw attention the problem as his prelude to proposing reforms. Unpopular measures such as cutting benefits, needs-based benefits, raising retirement ages or increasing payroll taxes are not to even discussed according to liberals. Furthermore, President Bush's proposals to increase investment returns on the Social Security trust fund are met with universal derision by liberals as unholy "privatization" unworthy of this New Deal institution, although one glance at the incredible returns being achieved with professional management of university endowments shows the enormous potential of this approach to solving the actuarial "crisis". Al Gore famously said the Social Security Trust Fund should be kept in a "lock-box", the financial equivalent of stuffing cash in a mattress for safe-keeping.

Do not attempt to understand the inconsistency over when liberals will heed a professional "consensus". It will only give you a headache.

(Nothing Follows)


Anonymous said...

If someone built a financial model that was the result of massive back-fitting and had never been accurate in forecasting then would you entrust the nation's economy to its predictions? - so many misunderstandings, falsehoods and bad thinking in this single sentence it's unbelievable.

1. Finances and the atmosphere are really quite different. If mankind has an imperfect understanding of economics, that makes absolutely no difference whatsoever to our understanding of climate.

2. "the result of massive back-fitting" - no, they are the result of physics. There's a lot more to it than just making a programme that will fit past data.

3. By massive back-fitting, presumably you mean that the model is able to fit past data. That is a good thing, not a bad thing.

4. "never been accurate in forecasting" - go and have a look at Hansen's 1988 projections under scenario B. Compare them to data since then. Evaluate. What do you find? I would wager a very large amount of money that you won't directly answer this question, but unfortunately for you those early predictions were rather good. It's been 20 years since then and I'm sure you know how much computing power has increased since then.

5. Models are not, as you seem to think, the sole reason scientists know that anthropogenic global warming is happening. You don't need models to understand the simple physics. More CO2=hotter atmosphere. That's been known for 100 years. Now, mountain glaciers are melting, the seas are warming, the Arctic is heating rapidly, Siberian permafrost is melting, hurricanes are becoming more powerful, and you bleat on about how models are rubbish. So ignore them - the facts of global warming are all around you now, and simple physics is enough to understand the basics of the situation, if you have a modicum of intelligence.

Jack Lacton said...


1. Finances are simpler than climate. That's one of the points.

2. No. They are the result of back-fitting. Yes, physics is applied but the example of the aerosol fudge - the physics of which is understood but the quantity is not, it was simply matched to observed cooling effect - proves my point.

3. Back-fitting means the adjustment of models to match observations. If the physics is understood then it's a very bad thing.

4. Hansen got a close answer with wrong emissions assumptions in scenario B.

5. More CO2=hotter atmosphere could well have been written "hotter atmosphere=more CO2", as you know. You've really killed any credibility you might have established with your shallow understanding of climate physics by highlighting the effects of warming. Glaciers advance and retreat all of the time. There seems to be a bit of a fight going on as to whether seas have warmed overall or not. Data collection methods have been a problem. Permafrost is a fragile thing and most of it is being lost due to local land use etc issues. Hurricanes are becoming less frequent and less powerful.

One interesting thing to note about the so-called observational proofs of global warming are that they nearly all seem to take place a long way from where the general population actually lives (so can't be empirically confirmed).

Anonymous said...

1. Consumers behave irrationally. Atmospheres don't. Apples and oranges do not taste the same. To try to compare the two is laughable.

2. Read the literature. The effects of Pinatubo were well reproduced and that was 17 years ago.

3. So basically, in your view, a model that doesn't match the data is bad, and a model that does is also bad?

4. Not true. Scenario B was rather close to what happened subsequently.

5. Hurricanes are becoming less frequent and less powerful. - I think you know that's not true. Dreaming up your own non-warming world to live in won't stop this one getting hotter, mate.

Darren said...

I recall a contemporary news article that said that Mt. Pinatubo released more greenhouse emissions in that one eruption than man has in his history as a species. I wonder if that's been backed up at all.

Still, this post is worth a *link*, and I provided one.

Oh, and Fudgie's not going to be convinced. I credit you for your valiant attempt, though.