Climate change: A road map

I want to follow up a previous posting with a road map for understanding climate change, something that can guide future postings and discussions. The case and supposed remedy for human-caused global warming have three components:

1. The temperatures: they seem to be going up.
2. The models: no one really knows why, but inferences from climate models seem to indicate that human activity is the cause.
3. The economics and politics: it's worse than anyone realizes: a bizarre combination of hypocrisy, ignorance, fear - and fear-mongering by the omnipresent news media and increasingly hysterical environmentalist movement, reacting off of each another in a death-dance.

To fully understand climate change and climate prediction requires touching on a lot of issues and drilling into some climate science and deep scientific issues. As we drill in, we'll discover that the supposedly solid case falls apart, piece by piece. We'll also encounter some far more likely explanations for climate change. My discussion here partially follows one of the definitive books on the subject, Essex and McKitrick's Taken by Storm: The Troubled Science, Policy and Politics of Global Warming.

First, consider the temperatures. Ignore the bogus "hockey stick" graph that purported to show a steady surface temperature from 1000 until 1980 - a definitely wrong claim now discredited. The case for global warming really rests on the last 150 years of temperature data from thermometers, largely from the northern hemisphere. On the surface, the case seems solid - the Earth's atmosphere probably has warmed a bit since the 1850s. But even here, there are some difficult questions.

What's an average of temperature? This statistical artifact is rarely examined in its own right, and yet it is problematic. An average, even a weighted average, of temperature has no physical significance. There is no one temperature for the Earth - it's a non-equilibrium system with an infinite field of temperatures, in three dimensions of space and one of time.*

And that's not even broaching the issue of measurement artifacts. The most important is the urban heat island effect, which is known to raise the measured temperature of a downtown or an airport as compared to the surrounding countryside by up to about a degree C. No one knows how to correct for this effect, except in an approximate way, even though temperature measurement networks have become more concentrated in the last century in urban areas.

What could the physical meaning of the average temperature be? When people talk about "global warming," what they really mean is "rising heat content" - so many more Joules or calories stored as heat in the atmosphere. The temperature (in absolute Kelvins) is supposed to be a proxy for this heat content, and that supposedly justifies the averaging: heat content of dry air is proportional to the volume of air times its absolute temperature.

The fatal difficulty for this simple equation is that the air is not dry; it contains a significant and variable amount of water vapor that can retain heat on its own. (Everyone knows the difference between muggy and dry heat.) Wet air's heat content is not a simple linear function of absolute temperature. Not only is the function non-linear; it suffers discontinuities when water evaporates, condenses, and precipitates. (Such discontinuous changes of the state of matter are called phase changes or phase transitions.) Evaporation makes the air capable of holding a lot more heat; condensation and precipitation take it out. Water vapor is by far the most important heat-trapping gas in the atmosphere, much more important than the next-leading contenders.

When the average temperature indexes widely trumpeted to indicate "global warming" are broken down into their basic component measurement series, more questions arise. Not only is there the urban heat island effect. The variability of the composite temperature index is dominated by large variations in the least reliable and most spotty measurements, ones taken in Eurasia and the southern hemisphere. The most reliable and extensive measurements - from Western and Central Europe, North America, and parts of East Asia - show less variation and, in fact, only slight warming. Global warming should be indicated by temperatures rising in most places, yet there is no consistent pattern. The Arctic is warming, for example, while the Antarctic is cooling.

Even so, the claimed temperature changes of the last 150 years, about 0.4 to 0.6 ^oC, are very much non-unprecedented in the history of the Earth's climate over the last 12,000 years, since the end of the last Ice Age, which have featured temperature changes of at least three to five times that amount. The Ice Age transition featured even larger changes, eight to 15 times as great.

Technically, we're still in an Ice Age, or one of its relatively ice-free periods called an interglacial. Until about three million years ago, the Earth was hotter, wetter, and ice-free. Since then, it's been cooling down and drying out. This development seems to have had an important impact on our primate ancestors in East Africa and probably was a controlling factor in the appearance of one of their offshoots, the genus Australopithecus - the immediate ancestor of genus Homo.

Second, consider the climate models. There are many - in pricinciple, there could be an infinite number of them. What's really going here is The One, Complete, But Unsolvable Theory of Earth's Climate - let's call it The Theory - that contains everything: the properties of the air, the oceans, the land, solar radiation, clouds, dust and other particles, etc. The climate models used by scientific groups to estimate the evolution of climate, like the models used to predict the weather, are impressive computer approximations to The Theory. (Climate models predict over months and years, while weather models predict over days and weeks.)

The critical fact about these models is that they are uncontrolled approximations and thus suffer from potentially fatal drawbacks, depending on how they're interpreted and used. In mathematical physics, controlled approximations are used all the time to estimate the behavior of systems too difficult to solve in their full complexity. What makes them controlled is that it's possible to determine upper bounds on the error made in the approximation. Moreover, many of these techniques offer iterative methods to make the approximations successively more accurate. You can then make the approximation as accurate as you want, limited only by your time, patience, and computer power.

In an uncontrolled approximation (which is what all climate models are), there is no way to tell if you're converging on the right answer. In general, if you attempt to make an uncontrolled approximation more accurate, you will converge to nothing - or anything. The methods are instead subject to modeler bias: modelers tend to iterate the uncontrolled method until they get the answer they want or expect. If the answer is already known from controlled laboratory experiments, this might be useful - although even here, it just reproduces what you already know, and by a questionable method. If the "answer" is laid down by non-scientific preconceptions, what you've got is a serious case of scientific corruption or self-deception. Overall, there has been not nearly enough systematic exploration of how modeling approximations affect the climate model results - learning what's robust and what are merely artifacts of the methods.

Lurking in the background is a deeper and more serious problem for long-term climate prediction: the Earth's climate is chaotic - in the "butterfly effect" sense of dynamical chaos. We'll learn in detail what that means in later postings. Suffice it to say that chaos makes long-term climate prediction impossible with present methods. New methods might be invented to circumvent this difficulty, but it's not a direction climate research has taken, because it has been subjected to such serious conceptual distortion by the "global warming" hysteria.

Chaos arises from a certain kind of non-linearity in the world, where effects are often not proportional to causes. The specific physics that makes climate hard to model comes in two basic types. One is that the atmosphere exhibits turbulence, the form of chaos familiar in fluid dynamics. (What we call "weather" is this turbulence in the lower atmosphere.) The other is the already-mentioned discontinuities in the phase of water and their effect on atmospheric heat content.

You only need to compare the situation with climate forecasting to that of weather forecasting. Back in the 1950s, forecasters naively thought all you needed was just more computing power to make accurate forecasts into the far future. From the 1960s on, thanks to Edward Lorenz and others, the phenomenon of chaos came to be understood as an essential property of weather. The best forecasts today use the available firehose of satellite and other data; even so, they're only good out to about two weeks. "Climate" is supposed to be a "long term" or "average steady state" of weather. What chaos is telling us is that there ain't no such thing. Most discussions of climate modeling are, rather amazingly, stuck in a pre-chaos naiveté.

Last, consider the economics and politics of climate change. Given the level of industrial civilization and the rise of the new economic powers of China, India, and Brazil in the next century, large reductions in CO2 emissions in the next century are impossible. Those countries didn't sign the 1997 Kyoto Treaty in any case. The accord is a case study in hypocrisy. All European countries agreed to it, yet have failed to implement the required CO2 emission reductions. The US initialed it, but the treaty was then rejected in the Congress in a nearly unanimous vote. European governments continue to talk about the treaty in grave tones, while their economies fall further and further behind their scheduled emission reduction targets. In the US, the growth of CO2 emission has slowed, but don't expect the US to get any credit. The Kyoto Treaty is a perfect object of media chatter: all talk, all pretense, no reality. The fact that there's no reality makes it especially gripping television.

The smash-up of the manufactured "global warming" "consensus" is coming. Expect everyone responsible - the professional hysterics, the news media, the self-deluded scientists - to duck responsibility for it. The smash-up will constitute another body blow to the toxic news media that have done so much damage to our society, our politics, and, now, our science. I hope that everyone - voters and politicians alike - will take a smart pill and view what's in their daily "news" with ever-greater skepticism. Even now, much of it is propaganda and lazily rewritten press releases. The opposition to the hysteria was, until recently, practically non-existent, hobbled by a largely hostile media and politicians devoted to "doing something" about an imaginary threat. There's more public opposition now, but it is all uphill.

The imaginary "global warming" crisis has also diverted attention from the reality of the environment in advanced societies: it's getting better every year, and most of the progress from modern regulation happened early on, in the 70s and 80s. There are still some areas for improvement (like the regulation of "light" trucks and SUVs), but the battle for a better environment here and in other wealthy countries is largely won. The arguments today, when not over marginal issues, are increasingly unreal hysteria having little to do with reality and everything to do with fundraising and a quasi-religious demonization of rational and scientific thought. The religious atmosphere of the environmentalist movement - its preaching of Original Sin, its search for authorities and scapegoats, and its ominous need for sacrifices to placate the supposedly angry gods - deserves its own discussion I'll leave to capable others.

The important thing about pollution and conservation is that the frontier now lies in the so-called Third World, both in countries on the ascent (China, India, Brazil) and in much poorer countries. They all need, not a shutdown of development, but a leapfrog that gets them into better, less polluting technology sooner, so they can avoid the mistakes that Europe and the United States made in their histories. (Visitors to Shanghai and São Pão will understand.) "Global warming" is a wasteful diversion from this, much more important task.
---
* A recent pathbreaking paper explains why temperature averages are physically meaningless - the paper is summarized here.