Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Introduction and Overview It has been known for over a century that trace gases in the atmosphere play a major role in controlling the Earth's heat balance and in maintaining surface temperatures at their current levels. Tyndall (1863) clearly described this "greenhouse" effect, pointing out that water vapor transmits a major fraction of the incident sunlight but strongly absorbs thermal radiation from the Earth. Later, Arrhenius (1896) and Chamberlin (1899) observed that CO2 also contributes to maintaining the heat balance and that changes in its abundance in the atmosphere could therefore affect the Earth's temperature. Indeed, Arrhenius (1896) estimated that a doubling of atmospheric CO2 concentrations would produce a global warming of about 6Â°C. Concern about human influences on CO2 concentrations and climate was voiced by Callendar(1938). Since then, our understanding of the physical processes governing climate has advanced markedly, but the inference that man-made changes in atmo- spheric composition can substantially affect climate has remained virtually unchanged. The stubborn refusal of the CO2 problem to "go away" is in itself significant. Increasing the sense of urgency is the firmly established observational evidence of steadily increasing CO2 concentrations in the global atmosphere. These agree within roughly a factor of 2 with estimates of CO2 emissions from growing fossil-fuel combustion, release of carbon by man from forest and soil reservoirs, and absorption of airborne CO2 in the oceans and the terrestrial biosphere. Moreover, we have acquired a heightened awareness of the sensitivity of our world society to changes in climate (e.g., World Meteorological Organization, 1979a). Thus, there is a clear need for as- 3
10 CARBON DIOXIDE AND CLIMATE: A SECOND ASSESSMENT sessment of the relationships between human activities, the composition of the atmosphere, global climate, and human welfare. The influence of changes in atmospheric composition on regional climate by a greenhouse effect is readily demonstrable from common experience. For example, we are all aware that in desert regions, temperatures drop rapidly after sunset, while in humid regions the day's heat lingers far into the night. In August, temperatures in Phoenix drop more than 30Â°F between afternoon and the following dawn; in Washington, the corresponding cooling is less than 20Â°F. Water vapor, like CO2, absorbs and re-emits heat radiation strongly, and a moist atmosphere acts as a thermal buffer between the Earth and space. Of course, short-lived local changes in atmospheric moisture or other constituents are not adequate analogs to the long-term global effects of increased CO2. (This greenhouse effect of water vapor is expected to play an important role in the total warming that might be caused by a rise in atmospheric CO2: the absorption of heat radiation by CO2 would increase evaporation and therefore the humidity, leading in turn to additional warming.) The possibility of climatic changes induced by human activities gained considerable prominence with the increasing postwar awareness of environ- mental problems. Revelle and Suess (1957) termed man-made injection of CO2 a "large-scale geophysical experiment," and the President's Science Advisory Committee (1965) report, Restoring the Quality of Our Environment, highlighted the CO2 problem and its potential consequences for the Antarctic Ice Cap. Climatic effects were more closely examined by a group of prominent U.S. scientists at the Study of Critical Environmental Problems (SCEP, 1970), held in preparation for the 1972 United Nations Conference on the Human Environment. This group strongly stated its concern over the potential effects of CO2 on climate and urged continuing study of the problem and monitoring of atmospheric CO2. SCEP prompted a more detailed review of the potential for inadvertent climate modification resulting from a wide range of human activities. The Study of Man's Impact on Climate (SMIC, 1971) involved leading scientists from all over the world and gave detailed recommendations on further research and monitoring efforts needed. The global carbon cycle, its possible alterations by man, and the consequent implications for airborne concentrations of CO2 were addressed at major international workshops (Stumm, 1977; Bolin et a/., 1979). Comprehensive reviews organized in the early 1970's to lay the foundations for national and world climate research programs cited the effect of increased CO2 on climate as a major research problem (U.S. Committee for the Global Atmospheric Research Program, 1975; Joint Organizing Committee, 1975). In 1977, a National Research Council (NRC) study (Geophysics Study Committee, 1977) concluded that "the climatic effects of carbon dioxide release may be the primary limiting factor on energy production from fossil
Introduction and Overview 11 fuels over the next few centuries" and recommended a well-coordinated research program to resolve the uncertainties in our understanding. A de novo analysis by a group of independent scientists (JASON, 1979) reinforced the general consensus on the problem's nature and magnitude, while a later study by the same group (JASON, 1980) contributed suggestions for research initiatives and monitoring strategies. At the request of the Office of Science and Technology Policy, reviews of the climate modeling aspects and economic-social implications were conducted (Climate Research Board, 1979, 1980). The national research program recommended in NRC reports took shape under the aegis of the National Climate Program and the Department of Energy (NOAA, 1980; U.S. Department of Energy, 1979). Internationally, the World Climate Conference (World Meteorological Organization, 1979a) highlighted the CO2 problem, and its study became a major objective of the World Climate Program. Comprehensive international studies in the context of energy-climate interactions were also conducted at the International Institute for Applied Systems Analysis (Williams. 1978) and the University of Muen- ster (Bach et al.. 1980). Under the auspices of the World Meteorological Organization (WMO), the International Council of Scientific Unions (icsu), and the United Nations Environment Program (UNEP), an expert group meeting in Villach, Austria, drafted an authoritative document (World Meteorological Organization. 1981). which is expected to form the basis for a coordinated international study effort in the context of the World Climate Program. The present study seeks to contribute to this continuing process of research, analysis, and assessment by re-examining the issues addressed by the 1979 Charney report in the light of the considerable additional research conducted since its preparation. As in the earlier report, the input to the deliberations was an assumption on the projected loading rate of atmospheric CO2. This involves not only a prediction of the anthropogenic worldwide production rates but also the role of the climate system (the atmosphere, oceans, cryosphere, and biosphere) in absorbing, transforming, storing, transporting, and interchanging carbonates and other trace constituents, which themselves interactâin short, the complex biogeochemical cycles. These processes are further complicated by the possibility that they, in turn, may be influenced by climatic change and variability. To simplify the working assumptions, we have accepted as the best current estimate of net CO2 loading rates that which was developed recently by the WMO/ICSU/UNEP group of experts (World Meteorological Organization, 1981): "... the atmospheric CO2 concentration in 2025 will be between 410 ppm and 490 ppm with a most likely value of 450 ppm." However, because of the large number of existing studies addressing the effects of doubled concentrations, we have employed a
12 CARBON DIOXIDE AND CLIMATE: A SECOND ASSESSMENT doubling as a convenient benchmark. In any event, the range of effects between present concentrations and a doubling covers no significant thresh- olds, and our conclusions are not critically dependent on the detailed time evolution of CO2 concentrations. It should be said at the outset that the problem of understanding the climate's response to a given scenario of CO2 change in the atmosphere is hardly distinguishable from the fundamental problem of understanding the natural variation and change of climate. For this reason, what is currently understood about the CO2 question is largely the result of progressive developments in climate modeling over the past quarter century. Climate models remain the most powerful means for dealing consistently and quantitatively with the complex system of highly interactive processes that determine climate. Further progress in sharpening our insights and estimates on the CO2 question, therefore, will continue to depend greatly on our ability to construct more faithful climate models. However, in this report we do not attempt to undertake a comprehensive review of the climate modeling problem itself, that is, such factors as the representation of the planetary boundary layer, the development of more accurate methods for computing radiative transfer, or an accounting for the influence of topographic features on the general circulation and regional climate. Nor do we attempt to catalog existing climate models and their relative performance. Such reviews are to be found elsewhere (e.g., Joint Organizing Committee, 1975, 1979). Our emphasis here is on those modeling problems that we perceive now most specifically to impede our ability to make sounder and more precise judgments on the likely response of the real climate to a given rate of CO2 increase. We emphasize radiatively connected processes because of the controversy in recent literature, and we specifically discuss a few papers that suggest a lower climate sensitivity. Quite clearly, the role of the oceans and the cryosphere seems paramount in influencing the nature of the climatic response, particularly in its evolu- tionary qualities. The possible interactive role that clouds may play in a shifting climate regime remains one of the unresolved mysteries. The question is how the cloud field would vary in concert with other changes induced by CO2 and thus alter the radiative transfer in the atmosphere. It is not yet even clear whether systematic cloud changes will provide a net amplification or an attenuation of the climate's sensitivity. Also in connection with the radiation field are the interactive effects of other trace gases and aerosols. Some new results indicate that, at least for the present, they cannot be ignored in addressing the CO2 question. Simplified climate models, which approximate only grossly horizontal transport processes, have become a popular tool for CO2 analysis. We
Introduction and Overview 13 therefore endeavor to delineate their strengths and weaknesses in contrast to the physically more comprehensive, but far more expensive and time- consuming, three-dimensional, time-dependent general-circulation models of the atmosphere and ocean. The confidence that one can place on the results of climate simulations is calibrated by a model's ability to replicate a variety of known physical states. Since laboratory-based, experimental sources of validation are virtually nonexistent, we must depend on empirical determinations gleaned from the geophysical medium itself. Contemporary climate is one such obvious source, especially its geographical, seasonal, and interannual structure. Furthermore, past climates provide the only records of large excursions in climatic regimes and are a key test-bed for developing and testing climate theory. It is becoming increasingly clear that the major climatic changes of the scale of glacial fluctuations are at least in part produced by the temporal and latitudinal changes of solar radiation brought about by slight changes in the Earth's orbital parameters (obliquity, time of perihelion, eccentricity). This finding promises a major advance in our knowledge of the sensitivity of climate to a small but well-known external forcing change. With both external forcing and climatic changes identified, there is an opportunity to identify and to quantify the role of internal feedbacks. In particular, studies of past climates may be especially germane to the CO2 question because there is evidence that atmospheric CO2 itself may have undergone large variations in the past. Often, only limited properties of the climate system of the past can be ascertained. For example, in the case of the last ice age 18,000 years ago, it is primarily the polar-ice perimeter, the sea-surface temperature, and the continental vegetation regimes that can be reconstructed, to a limited extent seasonally, but with limitations both temporally and geographically. Because the state of the climate system is incompletely known, even at present, every attempt must be made to determine the state as completely and with as little ambiguity as possible. We venture here to document the methods of validation and the conclusions that one can draw about the fidelity with which models can at present simulate climate and its variations. Special emphasis is given to the oceans because of their central and only partially understood role. One of the validation objectives specific to the CO2 question is the early detection of actual CO2-induced climatic changes predicted by climate models. Here one wishes to identify a small signal that may be obscured by the background of natural variability. If this variability were understood, it would be internally generated by the climate model and would permit derivation of a signal-to-noise ratio. In this case, the model could be used not only to determine the optimum climatic indices to monitor but also when, where, and how to monitor. However, should the model not incorporate an important climatic process, such as the effects of ubiquitous trace gases or an ability
14 CARBON DIOXIDE AND CLIMATE: A SECOND ASSESSMENT to account for the effects of an extraordinary event such as a volcanic eruption, a monitoring strategy based on model simulations could well be misleading. These questions are discussed in this report. The end product that is needed to assess the likely impacts of climate change on human activity is a comprehensive picture, or "scenario," of the time-space structure and the amount of that change. The relevant climate indices will depend on the particular activity under consideration. Forexample, temperature, precipitation, soil moisture, and solar radiation reaching the ground will be germane to an assessment of the impacts of climate change on agriculture. In this report, we address the question of scenario development, with some indication of our present abilities and measures of uncertainty. Finally, despite the admitted existence of numerous uncertainties, the consensus on the nature and magnitude of the problem has remained remarkably constant throughout this long worldwide process of study and deliberation. Burning of fossil fuels releases to the atmosphere carbon that was extracted by ancient plants many millions of years ago. The most recent projections of future energy consumption suggest a slackening in the growth in energy consumption; nevertheless, even the most conservative estimates imply major CO2 injections. The details of the natural carbon cycle and the future disposition of injected CO2 are still unclear, but it seems certain that much man-made CO2 will remain in the atmosphere. Although questions have been raised about the magnitude of climatic effects, no one denies that changes in atmospheric CO2 concentration have the potential to influence the heat balance of the Earth and atmosphere. Finally, although possibly beneficial effects on biological photosynthetic productivity have been recognized, no one denies that an altered climate would to some extent influence how humanity secures its continuing welfare. This report addresses the uncertainties in but one element of this consensus: the effects of changed concentrations of airborne CO2 on global climate. Other questions will be similarly studied in the course of a comprehensive NRC study of the entire issue, which is to be conducted over the next several years.