Elsewhere"> container">

Cold Thoughts on Global Warming

Jonathan I. Katz

The climate has gotten warmer over the last (roughly) 200 years, since the end of the Little Ice Age c. 1300--1800. Although many of the details are controversial, and some of the data may be corrupted in a variety of ways, the preponderance of evidence strongly supports this conclusion. The results of the recent Berkeley Earth Surface Temperature project have confirmed previous results.

There is no "Emerging Consensus". The physics of "greenhouse gases" and their effect on climate has been qualitatively understood since the work of Tyndall (1862) and Arrhenius (1896). Arrhenius made an estimate of the sensitivity of climate to carbon dioxide that is not far from the results of some of the most sophisticated modern climate models. The conclusion that anthropogenic emissions of these gases will likely warm the climate has been generally accepted for more than a century. It is a consensus, but it is not emerging or new. It has been there all along. Only a panicky fear of the consequences is new.

Climate is a complicated system with many feedback loops, most of which are not understood. For example, about half of the emitted carbon is re-absorbed. No one knows where it goes, although probably into some combination of the ocean and biosphere.

We do not know how to predict the future climate. The computer codes do an excellent job of explaining the past history of climate. They are calibrated to do so. They all contain a variety of "empirical parameters" arising from the need to describe processes, such as the formation of clouds and precipitation, that cannot be calculated from basic physical laws. Sometimes these empirical parameters are called "fudge factors". They are tuned so the codes agree with present and past climate. This does not mean that the codes necessarily have any predictive value (as Paul Dirac said, and "Yogi" Berra made famous, it is hard to predict, especially the future). The codes disagree with each other about the future even when the same assumptions are made, with results for the warming (climate sensitivity) that range over a factor of about 2 1/2. Most of the codes must be wrong, not because anyone has made any identifiable mistake, but because poorly understood processes interact with each other, in the codes and in reality, in unpredictable ways. It is likely that all the codes fail as general predictive tools, even though a few must get the right answer (if only because the calculated answers span the entire range of plausible answers) to any specific question. It is likely that different codes get the right answers to different questions.

This is a general problem with complex "multi-physics" codes that include many different interacting processes, some of which are not quantitatively understood from first principles and must be parametrized, semi-empirically. Another example occurs in the National Ignition Campaign, the project to ignite a tiny capsule of thermonuclear fuel by a giant laser at the Lawrence Livermore National Laboratory. Capsules that were predicted to ignite with certain laser pulses did not ignite. Many hypotheses have been suggested (some by myself), but there is no generally accepted explanation of the discrepancy between prediction and experiment. With further work ignition may well be achieved, but this will not change the conclusion that complex multi-physics codes are fallible predictive tools, even when calibrated by many experiments (climate models can only be calibrated by one natural climate). Yet another example, from astrophysics, is the "core collapse supernova", a variety of exploding star. In the real world they explode, but not in the computer calculations, despite forty years of trying to capture all the physics.

We cannot predict climate, and there is no prospect of doing so in the forseeable future. No one knows how to reduce the critical uncertainties. For example: The Earth reflects 30% of its incident sunlight. Increasing or decreasing this by one percentage point would produce a cooling or warming equal in magnitude to the warming from pre-industrial times to the present, negating or doubling the effect of anthropogenic greenhouse gases. But no one knows why the Earth actually reflects 30%, not 29% or 31%, so we cannot predict if this fraction will increase or decrease (or neither). Without understanding, better computers don't help.

It is likely that the last 100--150 years of surface warming have been anthropogenic, the result of emission of greenhouse gases, chiefly carbon dioxide. When you throw a blanket over a heat source (like a warm body in a bed), it gets warmer underneath. That is common sense, and everyday experience, unless you kick off the blanket (and perhaps the sheet too). The analogue to kicking off the blanket would be an unanticipated negative feedback (for example, increasing cloudiness). There is no reason to expect such a strong negative feedback, but no reason, within our present understanding, why it cannot occur. Some positive feedbacks, such as emission of methane from thawed permafrost, very likely do occur, but are not quantified.

The coincidence of the recent warming with the increase of greenhouse gases suggests, but does not prove, a causal connection. In the distant past (over hundreds of thousands of years), increases in atmospheric carbon dioxide apparently followed, rather than preceded, warming, (and similarly for decreases and cooling). This is not directly applicable to our present concern with the effect of the anthropogenic injection of carbon dioxide because there was no known natural injection then. It does indicate some positive feedback (warming releasing carbon dioxide). Warming of 10 degrees C is correlated with an increase of carbon dioxide content of roughly 100 ppm (the modern warming of about 0.8 degrees C would then be expected to increase the carbon dioxide content by 8 ppm, a very small fraction of the anthropogenic increase over the last 150--200 years of 120 ppm, and even less if there are long delays in the feedback).

The IPCC says it is "90%" confident of a causal connection between anthropogenic greenhouse gas emission and the current warming. The number cannot be meaningful unless they have made a large number of predictions in which they claim a similar degree of confidence, and found that 90% of them are correct. They haven't done so; 90% is a spuriously quantitative way of saying "We think this is probably the case, but cannot be really sure". If you accept the 90% as quantitatively meaningful (you shouldn't), then there is a 10% chance the warming has some other cause. When statistical data actually exist, usually a 95% or 98% confidence is required for a conclusion to be taken seriously, and a remarkable fraction of such conclusions turn out to be wrong. In fact, it is almost proverbial that "half of all three-sigma [standard deviation] results are wrong" (if ideal statistical laws held, the fraction would be about 0.3%). Some things cannot be quantified, and attempts to do so only mislead. It is better to explain the physical reasoning to expect such a causal connection, and to stop there.

It is unlikely that the roughly 150 years of warming prior to the mid-19th Century were anthropogenic because emissions of man-made greenhouse gases were slight, and samples of air trapped in ice show that the atmospheric concentration of carbon dioxide did not rise significantly until the later 19th Century. Some warming must have been the consequences of natural processes. The more recent warming may be partly natural and partly anthropogenic. It is not possible to disentangle these two processes because we have no understanding at all of natural warming and cooling trends, beyond the fact (evident from the history of cold ice ages and warm interglacials) that both occur.

If global warming is anthropogenic, what should we do and why? Some talk about reducing carbon (dioxide) emissions. This is a fantasy. The developed countries aren't going to reduce their emissions much---10% or 20% over a decade is possible, but not the 50% or 80% some advocate. These would require giving up two or three of the following four: fossil fuel generation of electricity, motor vehicles, air travel and fossil fuel heating of buildings. It won't happen. Emissions from developing countries, already exceeding those of the US and Western Europe, will rise rapidly because so many sources (travel, electric power, infrastructure that requires carbon emission to build) are things people buy more of as they become more prosperous. People won't freeze in the dark for the sake of a scientific theory, even a correct scientific theory.

It is an inconvenient truth that carbon emissions are going to continue to rise, with a little help from politicians and hangers-on jetting to climate conclaves in Kyoto, Bali, Copenhagen and Doha. There is nothing we can do to change this.

What will be the consequences? The climate will get warmer. There is no evidence that storms or droughts will increase. In the past few decades severe storms may have become somewhat less frequent, and the Sahelian drought of the 1970's and 1980's has ended. There is no evidence these trends are related to global warming.

Global warming does not make the tropics, or warm seasons in temperate zones, much hotter. For good physical reasons, greenhouse gas warming occurs chiefly in arctic and sub-arctic regions and in temperate zones during winter; in other words, when the weather is cold. One consequence is the lengthening of growing seasons in temperate climates.

The last major episode of warmer climate, the Late Medieval Climatic Maximum (which had nothing to do with anthropogenic greenhouse gases---no one knows why it began, or why it gave way to the Little Ice Age later) may have been good for humanity, although because of the paucity of records we cannot be sure. The Vikings settled Iceland and Greenland, which was actually green, as it is becoming again today. Favorable conditions in northern Europe may have been accompanied by less favorable conditions elsewhere, in particular droughts in the Mideast (R. Ellenblum The Collapse of the Eastern Mediterranean: Climate Change and the Decline of the East 950--1072 Cambridge U. Press 2012).

Who is stoking the alarm about global warming? There is Al Gore, an over-the-hill politician who wants to remain in the public eye. His house uses 20 times as much electricity as the average American house and he flies private jets. Conservation is for the little people. He doesn't care about climate, or even about polar bears; these are just tools to get government subsidies for his investments in renewable energy, cashing in his connections as a Washington insider. Then there is Jim Hansen, would-be dictator who wants to throw in jail anyone who disagrees with him or burns coal. He may wish himself another Mussolini (or worse), but people just laugh at him. His extremism discredits sound science, some of which he may take credit for, and contributes to the widespread public skepticism about the reality of warming; with "friends" like him, climate science makes enemies. And finally John Holdren, who in his younger days was prophesying disaster from the ice age then just beginning (so he said). Jumping on bandwagons may be a good career move, but it is not good science. Fictitious crises are a demogogue's route to power.

Some of the more apocalyptic fears about global warming resemble a doomsday cult. Rather than God dooming mankind for its traditional sins (robbery, lust, murder, disbelief, etc.), Nature is said to doom mankind for the secular sin of carbon emission. Some (Greenpeace, and even more radical groups) think any human effect on nature to be sinful, and regard "Mother Earth" as a deity that is violated by any use of its resources for the sustenance, comfort or betterment of Mankind. Needless to say, this is opposite to the Biblical grant of the natural world to Man for his benefit.

Predictions of climate doom are no more rational than religious predictions of a Day of Judgement or Armageddon. Divine revelation is not open to rational argument, and its truth can only be judged by further revelation.

Global warming is real and much of it is probably anthropogenic. Nothing serious will be done about it, no matter how fevered the rhetoric and prophesy. It may be good for humanity by lengthening growing seasons; it may be bad for humanity if weather patterns, particularly rainfall, change for the worse. Man conquered the globe because he adapted, in prehistoric times, to climates from the Arctic to the Sahara. We will adapt to a warmer Earth.



Jonathan Katz
Monday December 31 2012