Here is part one of a response:
Global Warming: The Origin and Nature of the Alleged Scientific Consensus
Richard S. Lindzen
Richard S. Lindzen is the Alfred P. Sloan Professor of Meteorology at the Massachusetts Institute of Technology.
Most of the literate world today regards "global warming'' as both real and dangerous. Indeed, the diplomatic activity concerning warming might lead one to believe that it is the major crisis confronting mankind. The June 1992 Earth Summit in Rio de Janeiro, Brazil, focused on international agreements to deal with that threat, and the heads of state from dozens of countries attended. I must state at the outset, that, as a scientist, I can find no substantive basis for the warming scenarios being popularly described. Moreover, according to many studies I have read by economists, agronomists, and hydrologists, there would be little difficulty adapting to such warming if it were to occur. Such was also the conclusion of the recent National Research Council's report on adapting to global change. Many aspects of the catastrophic scenario have already been largely discounted by the scientific community. For example, fears of massive sea-level increases accompanied many of the early discussions of global warming, but those estimates have been steadily reduced by orders of magnitude, and now it is widely agreed that even the potential contribution of warming to sea-level rise would be swamped by other more important factors.
To show why I assert that there is no substantive basis for predictions of sizeable global warming due to observed increases in minor greenhouse gases such as carbon dioxide, methane, and chlorofluorocarbons, I shall briefly review the science associated with those predictions.
Summary of Scientific Issues
Before even considering "greenhouse theory,'' it may be helpful to begin with the issue that is almost always taken as a given--that carbon dioxide will inevitably increase to values double and even quadruple present values. Evidence from the analysis of ice cores and after 1958 from direct atmospheric sampling shows that the amount of carbon dioxide in the air has been increasing since 1800. Before 1800 the density was about 275 parts per million by volume. Today it is about 355 parts per million by volume. The increase is generally believed to be due to the combination of increased burning of fossil fuels and before 1905 to deforestation. The total source is estimated to have been increasing exponentially at least until 1973. From 1973 until 1990 the rate of increase has been much slower, however. About half the production of carbon dioxide has appeared in the atmosphere.
Predicting what will happen to carbon dioxide over the next century is a rather uncertain matter. By assuming a shift toward the increased use of coal, rapid advances in the third world's standard of living, large population increases, and a reduction in nuclear and other nonfossil fuels, one can generate an emissions scenario that will lead to a doubling of carbon dioxide by 2030--if one uses a particular model for the chemical response to carbon dioxide emissions. The Intergovernmental Panel on Climate Change Working Group I's model referred to that as the "business as usual'' scenario. As it turns out, the chemical model used was inconsistent with the past century's record; it would have predicted that we would already have about 400 parts per million by volume. An improved model developed at the Max Planck Institute in Hamburg shows that even the "business as usual'' scenario does not double carbon dioxide by the year 2100. It seems unlikely moreover that the indefinite future of energy belongs to coal. I also find it difficult to believe that technology will not lead to improved nuclear reactors within fifty years.
Nevertheless, we have already seen a significant increase in carbon dioxide that has been accompanied by increases in other minor greenhouse gases such as methane and chlorofluorocarbons. Indeed, in terms of greenhouse potential, we have had the equivalent of a 50 percent increase in carbon dioxide over the past century. The effects of those increases are certainly worth studying--quite independent of any uncertain future scenarios.
The Greenhouse Effect.
The crude idea in the common popular presentation of the greenhouse effect is that the atmosphere is transparent to sunlight (apart from the very significant reflectivity of both clouds and the surface), which heats the Earth's surface. The surface offsets that heating by radiating in the infrared. The infrared radiation increases with increasing surface temperature, and the temperature adjusts until balance is achieved. If the atmosphere were also transparent to infrared radiation, the infrared radiation produced by an average surface temperature of minus eighteen degrees centigrade would balance the incoming solar radiation (less that amount reflected back to space by clouds). The atmosphere is not transparent in the infrared, however. So the Earth must heat up somewhat more to deliver the same flux of infrared radiation to space. That is what is called the greenhouse effect.
The fact that the Earth's average surface temperature is fifteen degrees centigrade rather than minus eighteen degrees centigrade is attributed to that effect. The main absorbers of infrared in the atmosphere are water vapor and clouds. Even if all other greenhouse gases (such as carbon dioxide and methane) were to disappear, we would still be left with over 98 percent of the current greenhouse effect. Nevertheless, it is presumed that increases in carbon dioxide and other minor greenhouse gases will lead to significant increases in temperature. As we have seen, carbon dioxide is increasing. So are other minor greenhouse gases. A widely held but questionable contention is that those increases will continue along the path they have followed for the past century.
The simple picture of the greenhouse mechanism is seriously oversimplified. Many of us were taught in elementary school that heat is transported by radiation, convection, and conduction. The above representation only refers to radiative transfer. As it turns out, if there were only radiative heat transfer, the greenhouse effect would warm the Earth to about seventy-seven degrees centigrade rather than to fifteen degrees centigrade. In fact, the greenhouse effect is only about 25 percent of what it would be in a pure radiative situation. The reason for this is the presence of convection (heat transport by air motions), which bypasses much of the radiative absorption.
What is really going on is schematically illustrated in Figure 1. The surface of the Earth is cooled in large measure by air currents (in various forms including deep clouds) that carry heat upward and poleward. One consequence of this picture is that it is the greenhouse gases well above the Earth's surface that are of primary importance in determining the temperature of the Earth. That is especially important for water vapor, whose density decreases by about a factor of 1,000 between the surface and ten kilometers above the surface. Another consequence is that one cannot even calculate the temperature of the Earth without models that accurately reproduce the motions of the atmosphere. Indeed, present models have large errors here--on the order of 50 percent. Not surprisingly, those models are unable to calculate correctly either the present average temperature of the Earth or the temperature ranges from the equator to the poles. Rather, the models are adjusted or "tuned'' to get those quantities approximately right.
It is still of interest to ask what we would expect a doubling of carbon dioxide to do. A large number of calculations show that if this is all that happened, we might expect a warming of from .5 to 1.2 degrees centigrade. The general consensus is that such warming would present few, if any, problems. But even that prediction is subject to some uncertainty because of the complicated way the greenhouse effect operates. More important, the climate is a complex system where it is impossible for all other internal factors to remain constant. In present models those other factors amplify the effects of increasing carbon dioxide and lead to predictions of warming in the neighborhood of four to five degrees centigrade. Internal processes within the climate system that change in response to warming in such a manner as to amplify the response are known as positive feedbacks. Internal processes that diminish the response are known as negative feedbacks. The most important positive feedback in current models is due to water vapor. In all current models upper tropospheric (five to twelve kilometers) water vapor--the major greenhouse gas--increases as surface temperatures increase. Without that feedback, no current model would predict warming in excess of 1.7 degrees centigrade--regardless of any other factors. Unfortunately, the way current models handle factors such as clouds and water vapor is disturbingly arbitrary. In many instances the underlying physics is simply not known. In other instances there are identifiable errors. Even computational errors play a major role. Indeed, there is compelling evidence for all the known feedback factors to actually be negative. In that case, we would expect the warming response to carbon dioxide doubling alone to be diminished.
It is commonly suggested that society should not depend on negative feedbacks to spare us from a "greenhouse catastrophe.'' What is omitted from such suggestions is that current models depend heavily on undemonstrated positive feedback factors to predict high levels of warming. The effects of clouds have been receiving the closest scrutiny. That is not unreasonable. Cloud cover in models is poorly treated and inaccurately predicted. Yet clouds reflect about seventy-five watts per square meter. Given that a doubling of carbon dioxide would change the surface heat flux by only two watts per square meter, it is evident that a small change in cloud cover can strongly affect the response to carbon dioxide. The situation is complicated by the fact that clouds at high altitudes can also supplement the greenhouse effect. Indeed, the effects of clouds in reflecting light and in enhancing the greenhouse effect are roughly in balance. Their actual effect on climate depends both on the response of clouds to warming and on the possible imbalance of their cooling and heating effects.
Similarly, factors involving the contribution of snow cover to reflectivity serve, in current models, to amplify warming due to increasing carbon dioxide. What happens seems reasonable enough; warmer climates presumably are associated with less snow cover and less reflectivity--which, in turn, amplify the warming. Snow is associated with winter when incident sunlight is minimal, however. Moreover, clouds shield the Earth's surface from the sun and minimize the response to snow cover. Indeed, there is growing evidence that clouds accompany diminishing snow cover to such an extent as to make that feedback factor negative. If, however, one asks why current models predict that large warming will accompany increasing carbon dioxide, the answer is mostly due to the effect of the water vapor feedback. Current models all predict that warmer climates will be accompanied by increasing humidity at all levels. As already noted, such behavior is an artifact of the models since they have neither the physics nor the numerical accuracy to deal with water vapor. Recent studies of the physics of how deep clouds moisturize the atmosphere strongly suggest that this largest of the positive feedbacks is not only negative, but very large.
Not only are there major reasons to believe that models are exaggerating the response to increasing carbon dioxide, but, perhaps even more significantly, the models' predictions for the past century incorrectly describe the pattern of warming and greatly overestimate its magnitude. The global average temperature record for the past century or so is irregular and not without problems. It does, however, show an average increase in temperature of about .45 degree centigrade plus or minus .15 degree centigrade with most of the increase occurring before 1940, followed by some cooling through the early 1970s and a rapid (but modest) temperature increase in the late 1970s. As noted, we have already seen an increase in "equivalent'' carbon dioxide of 50 percent. Thus, on the basis of models that predict a four degree centigrade warming for a doubling of carbon dioxide we might expect to have seen a warming of two degrees centigrade already. If, however, we include the delay imposed by the oceans' heat capacity, we might expect a warming of about one degree centigrade--which is still twice what has been observed. Moreover, most of that warming occurred before the bulk of the minor greenhouse gases were added to the atmosphere. Figure 2 shows what might have been expected for models with differing sensitivities to a doubling of carbon dioxide. What we see is that the past record is most consistent with an equilibrium response to a doubling of about 1.3 degrees centigrade--assuming that all the observed warming was due to increasing carbon dioxide. There is nothing in the record that can be distinguished from the natural variability of the climate, however.
If one considers the tropics, that conclusion is even more disturbing. There is ample evidence that the average equatorial sea surface has remained within plus or minus one degree centigrade of its present temperature for billions of years, yet current models predict average warming of from two to four degrees centigrade even at the equator. It should be noted that for much of the Earth's history, the atmosphere had much more carbon dioxide than is currently anticipated for centuries to come. I could, in fact, go on at great length listing the evidence for small responses to a doubling of carbon dioxide; there are space constraints, however.
|
