Issues, Assumptions, and Values
Three questions frame the study of climate change: Is it happening? Can we stop it? How can we cope with it? Asking the third question may seem pessimistic. Nevertheless, if during the past two centuries mankind has committed the planet to a new climate, it must be answered. On the other hand, if the planet is not yet committed, balancing the cost of stopping climate change against its result still calls for an answer. And, if humanity is not changing the climate, we need to know how to cope with natural variations in weather and climate. So, against a background of present climatic differences from place to place and changing weather from day to day, the Adaptation Panel here tries to answer for Congress the final "How can humanity cope with climate change?"
Simply put, climate is the average state of the weather. At a deeper level, the climate of a locality is the synthesis of the day-to-day values of the meteorological elements that affect the locality. Synthesis implies more than simple averaging. Various methods represent climate, for example, both average and extreme values, frequencies of values within ranges, and frequencies of weather types. The main climatic elements are precipitation, temperature, humidity, sunshine, wind, and such phenomena as fog and frost. Climatic data are usually stated in terms of an individual month or season (McIntosh, 1972).
The critical matter is that climate is the accustomed seasons, daily cycles, variations, and ties among the factors of weather at a given place. The question is whether the nature and civilization that have evolved in the climate of a place will be greatly affected by or can easily adapt to the
future climate. Nature, here, means the natural or unmanaged living things outdoors. Humanity, of course, is part of nature, but we use the word nature to mean the unmanaged environment. In the past the study of the outcome of climate change for nature and humanity dealt largely with the impacts, or blows, themselves. Studies have dealt, for example, with the changes a new climate would cause in the plants of an ecosystem, the yield of a crop, or the safety of a seawall, assuming little or no adaptation. Here the panel integrates projections of impacts into its discussion of how to cope (see Chapter 34).
The most severe challenge in weighing impacts, however, is to compound the outcome of a changing climate with other changes that will occur during the coming decades. These changes range from technological innovations to social changes and from increased numbers of humans to ecological impacts. We can grasp the possible magnitude of some of the kinds of changes that might happen at the same time as a climate change by looking back eight decades. In 1910 the Ottoman, Austro-Hungarian, British, and Russian empires ruled much of the world. In America there were no income taxes, women could not vote, each person commanded 1.5 horsepower, and the major polluters were the 21 million horses (Nordhaus, 1990).
Although the crystal ball for seeing future life and technology is cloudy, the magnitude of the changes of the past 80 years teaches us that we must go beyond computation of the effect of warming of 1° to 5°C on, say, today's corn or coast. We must try to foresee the adjustments in nature and human behavior that will occur in response to changing environmental conditions amidst an army of technological, social, and economic changes. If we ignore these adjustments, which are here termed adaptations, we will write a "dumb people scenario." Imposing the climate near the middle of the next century on the activities of 1990 implicitly assumes that people will dumbly ignore any new environment and circumstances for 80 years and behave as they do today.
Three classes of adaptation by humans can be distinguished (Coppock, 1990). The first might be called adjustments. These are prompt, individual, uncoordinated, and largely spontaneous, such as the changes a farmer makes in crop varieties after a couple of cold years. A second class, which might be called premeditated adaptation, begins with anticipation and information and requires planning, coordinated action, and time. This class is typified by the building of a dam for irrigation. A third class might be called interventions. These actions, typically by governments, manipulate the circumstances of choices and are exemplified by the zoning of wetlands. Here, the word adaptation encompasses all these.
An illustration of a dumb scenario is a vision of the corn varieties and husbandry of 1990 in a changed climate in 2030 and then calculation of the impact of climate change as a change from the 1990 yields. A smarter
scenario is a vision of a crop changing for four decades in the future by methods and along a trend somewhat as in the past. Yields, too, would follow a trend, which would be a baseline or reference. With no adaptation to climate, future yields would deviate from that baseline, up if the climate change were favorable and down if it were harmful. Adaptation would change the yields, and, if the climate change were harmful, the adapted yields would lie between the rising baseline of no climate change and the trend of yields not adapted to the changing climate. In the smarter scenario, the impact of climate change would be the net sum in 2030 of the costs of adaptation and the difference between the baseline of yield in an unchanged climate and the yield of the adapted crop in the changed climate.
Ask a farmer who is 70 or 80 years old what is different now compared to when he was a child. He would likely find changing from horses to tractors, from dirt to paved roads and from open pollinated corn to hybrid corn plus the arrival of soybeans and pesticides swamped the assumed climatic warming of a half to whole degree during the 20th century.
Similar illustrations could be drawn from other activities. Figure 4.1 of Part One shows the general case.
Estimating cannot be precise, but the panel realizes or foresees the following:
• The impacts of climate change must be sorted out from other effects caused by simultaneous changes in other factors.
• The baseline or reference of what would happen without climate change will trend up or down.
• The impact of climate change is the net of the cost of adaptation plus or minus the residual change from the baseline that occurs despite adaptation.
• The net impact of climate change can be negative, or, if the residual change is a help greater than the cost of adaptation, the net can be positive.
Studies of the impacts of climate change commonly begin with a climate scenario. Will it be 1°C or 5°C warmer? Will 10 percent more or less rain fall? And so forth. But these specifications are shaky at best and only compound the uncertainty already inherent in the complex response to climate.
To some extent, however, the studies of outcome can be independent of the climate scenarios. An activity like farming or forestry will be different in the future than it is today. If climate changes, part of that difference will be the impact of climate change. If the sensitivity to climate of the activity is constant, the impact is simply the climate change times the sensitivity. In that case a student of outcomes can estimate the sensitivity and then leave it
to others to calculate the impact of climate change by multiplying the constant sensitivity by any scenario proposed.
More formally, let the rate of change of activity A be dA/dt. Then let the rate of change dA/dt of A equal the sum of products of sensitivities to factors times the rates of change of the factors. Since our subject is climate, the first product is the sensitivity dA/d(climate) of A to climate times the rate of change d(climate)/dt. Bringing together all the other factors that affect A, we let the second and remaining product be the sensitivity dA/ d(other factors) of A to other things times the rate of change d(other factors)/dt. The relative importance of climate change to A depends on its sensitivity to both climate and other factors and on the rates of change of both climate and other factors. Finally, if the sensitivity to climate is unchanged by, say, adaptation or climate, the impact of climate change is simply the constant sensitivity dA/d(climate) times the change in climate, which is the integral of d(climate)/dt. In this last case, studies of sensitivity can be separated from scenarios.
Still, there is a limit to the separation of studies of sensitivity from assumptions of climate change. The sensitivity, dA/d(climate), depends on the value of climate. For example, the sensitivity of a tomato per degree is different from 5° to 15°C than at the threshold of 0°C, where it freezes. The sensitivity of a marsh to a sea rising 1 m is not likely to be just 10 times its sensitivity to a rise of 10 cm. The sensitivity of a 1-m wall to 50-and 150-cm rises in sea level is utterly different.
So the panel did not have to assume a precise climate scenario, but it did have to assume the sort and order of climate change. The scenarios that are supportable, the forecasts of climate change, and the warnings about the uncertainties of the forecasts were provided by the Effects Panel. Briefly, they are, in the absence of human efforts to mitigate emissions:
• Greenhouse gases will reach the equivalent of 600 ppm CO2 near the middle of the 21st century.
• Mathematical models project that this increase of greenhouse gases will warm the planet 1° to 5°C, on average, over the temperature of about 1990. This warming would be achieved if the planet comes to equilibrium with the 600 ppm.
• The warming actually realized during the past century appears to be 0.3° to 0.6°C. Simple logic suggests that lag will make the realized warming by the middle of the 21st century less than the warming at equilibrium with the 600 ppm.
• The projections are plagued by uncertainties about the accumulation of the gases and their absorption and the roles of oceans, clouds, and other environmental elements.
• Projections about the climate of a locality are doubly uncertain. Precipitation may be tens of percent more or less than now.
• A rise in sea level may accompany global warming, possibly in the range of 0 to 60 cm for the timing and temperature range listed above. These assumptions about the physical environment are consistent with those made by others (National Research Council, 1983; Smith and Tirpak, 1989; Schneider et al., 1990).
The assumed increase of greenhouse gases to 600 ppm near the middle of the 21st century implicitly assumes that population and economic activity will increase. The growth of material well-being is relevant, as it partly determines what adaptations are affordable. In its studies of policy responses, the Environmental Protection Agency (EPA) assumed a world growth rate of less than 1 percent per year for their slowly changing (low-emission) world scenario and over 2 percent per year for their rapidly changing (high-emission) world scenario (Lashof and Tirpak, 1991). The Intergovernmental Panel on Climate Change (IPCC) assumed that annual economic growth would be 2 to 3 percent in Organization for Economic Cooperation and Development (OECD) countries and 3 to 5 percent in Eastern Europe and developing countries during the coming decade and slower thereafter (Intergovernmental Panel on Climate Change, 1990). The Adaptation Panel assumed a positive growth of material well-being without specifying it precisely. To assess costs of impacts and adaptations, a further financial assumption on discount rates is required. The Adaptation Panel used discount rates of 3, 6, and 10 percent in its analysis.1
To gauge impacts and foresee adaptations, however, the panel also needed to go beyond a generality like "1° to 5°C warmer on average" to consider how the climate affecting a community might be transformed year by year. Alternatives are depicted in Figure 32.1. The top panel shows a frequency distribution and no trend for a climate steady in its average and variability. The second panel shows one sort of change: The frequency distribution is unchanged in its shape and spread but shifted left because the average falls. The fall could be either steady or sudden. The third panel shows another sort of change: The average is unchanged, but the variability is increasing. It could, of course, have been drawn with decreasing variability. The final panel shows both the average falling and the variability increasing.
The effects associated with a global cooling of roughly 5°C during the last ice age 18,000 years ago are known, and a repetition would crush the upper midwestern United States under ice. No similarly dire outcome has been found associated with a 5°C warming. None of the projected effects of the warming would make large areas uninhabitable. So while we complain about the uncertainty of warming by greenhouse gases, we should not forget that it would make any effects from cooling, which was a subject of some scientific, public, and congressional concern in the early 1970s, less likely.
Because computing the average temperature is hard enough, scenarios
usually give only the average that would be reached at equilibrium for an equivalent CO2 doubling. Nevertheless, some recent efforts to compute the variability in the new climate foresee a shift in the average temperature without a change in variability (Mearns et al., 1989, 1990; Rind et al., 1989;Smith and Tirpak, 1989). A projection of no great change in variability is not inconsistent with the statement by IPCC Working Group I that neither more nor fewer storms could be predicted for the future climate (Intergovernmental Panel on Climate Change, 1990).
Turning from temperature to precipitation, one encounters fully the need for regional and seasonal predictions of the frequencies of drought and flood. Yet predicting even the change in averages is beyond present skill.
A comparison among the frequency distributions of precipitation today, however, hints at how their interannual variability would usually change if the average changed. If the average falls, the absolute interannual variability, measured by the variance, would fall. The relative variability, measured by the ratio of the standard deviation to the average, however, would increase. Skewness would also increase.2 Since precipitation cannot be less than zero, future changes in its frequency distributions will generally resemble these differences among its present ones.
A final question concerns the path from the present to future climate. The panel concentrated on a gradual change like the upper trend in the second panel of Figure 32.1. As drawn, the trend might represent a change to less precipitation. If the trend were drawn up, it might represent a change to warmer temperatures. Annual variations around the trend will, of course, continue, and sometimes several years that are above or below the trend will mislead people. A gradual change is not a surprising effect of the gradual increase in greenhouse gases that forces the change, and a gradual change is generally consistent with the computations of models. The panel recognizes that the passage of some threshold or shifting ocean current might cause an abrupt, harmful change. We were unable to evaluate the likelihood of these events, and, while recognizing their potential for significant impacts, we have not analyzed their impacts or adaptations to them. The ability to adapt to every extreme outcome of climate change is, of course, in doubt. This report concentrates on the range of the changes that might occur within the next half century or so and that are stated in the assumptions above. Within that range, moderate changes seem more likely than radical ones. The IPCC stated that business as usual would likely raise the global mean temperature above the present one by about 1°C by 2025 and 3°C before the end of the next century (Intergovernmental Panel on Climate Change, 1990). Furthermore, examining the consequences of, say, a warming of 1°C makes sense even if the eventual warming is greater because the planet would first warm 1°C before warming more. The effects of abrupt and radical changes might well be examined in the future if the physical logic for them becomes convincing. For now, however, the panel concentrated on moderate changes.
In the end panel considered the sensitivities and adaptations to a gradual change in averages without a great change in variability to climates warmer by a few degrees, where precipitation is tens of percent more or less than at present and the sea is 20 to 30 cm higher by the middle of the next century.
Economic and Ethical Values
In much of the world, certainly in the United States, suppliers signal their willingness by prices and then buyers vote their wishes in markets.
Fundamentally, the economic view is centered on humanity. The activities that are worthwhile in this view are valued by people and expressed in market and political decisions. The economic view is not that nature is without value but that nature's value derives from human values about nature. If people do not care about the landscape, whales, and snails, as shown by their dollars and votes, these things will have no economic worth. On the other hand, in a society that loves the environment, high values and high prices would be placed on parks, wetlands, and open spaces, and these values would carry over to private transactions and legislative mandates.
In a preliminary economic analysis of the impact of climate change, the 1981 national income of $2.4 trillion was divided into 16 sectors. Their sensitivity ranged from the potentially severe impact on farms and forests to the negligible impact on mining and manufacturing. The calculated direct impact of climate change on the national income would be about 1 percent (Nordhaus, 1991). The estimate is only 1 percent because, among other things, the impact on sensitive agriculture may be positive or negative and the impact on the valuable manufacturing and service sectors is negligible. Although the absolute number of dollars may be large, the relative amount of 1 percent is a discouragement for large investments in climate change.
Although the national income accounts capture only the value of marketed goods and services, it is possible to incorporate goods like environmental quality that are not marketed or priced accurately, if at all (Coomber and Biswas, 1973). These environmental intangibles are generally outdoors and so exposed to climate. The incorporation of these nonmarket goods and services might increase the sensitivity of the economy to climate change.
Beyond intangibles that can be assigned a price are ethical values. An appraiser might devise a formula, for example, to give a price for a shade tree, adding it to the value of a home. Beyond that appraisal, however, lies the ethical question of whether it should be cut down, especially if birds are nesting in it.
If our present evolutionary impetus is an upward one, it is ecologically probable that ethics will eventually be extended to land. The present conservation movement may constitute the beginnings of such an extension. If and when it takes place, it may radically modify what now appear as insuperable economic obstacles to better land use (Leopold, 1933).
All these ways (i.e., economics of marketed goods and services, economics of nonmarket goods and services, and ethical arguments) of measuring human well-being and the quality of our stewardship of the planet need to be taken into account in assessing the seriousness of the impacts of climate change and the potential and appropriateness of adaptations. Taking them into account, of course, requires judgment as well as evidence and computation, and the panel has tried to take these into account fairly in judging findings and eventually recommendations.
1. As the exercise in Chapter 33, the section "Making Decisions in an Uncertain World," demonstrates, the assumption of a discount rate is important for adaptation. Rates are crucial for weighing mitigations, which are anticipatory, and rates are discussed at length in Part Three: Mitigation. The rates of 3, 6, and 10 percent are also found in Part One: Synthesis. For growth of incomes see Table 34.1.
2. Among 660 cases of 12 monthly distributions of precipitation at 55 stations, the variance increased as the 1.3 power of the mean. This specifies that both the coefficient of variation and the skewness of the frequency distribution vary as the -0.35 power of the mean (Waggoner, 1989).
Coomber, N. H., and A. K. Biswas. 1973. Evaluation of Environmental Intangibles. Bronxville, N.Y.: Genera Press.
Coppock, R. 1990. Definitions. Staff paper prepared for the Adaptation Panel of the Panel on Policy Implications of Greenhouse Warming, April 11, 1990.
Intergovernmental Panel on Climate Change. 1990. Climate Change: The IPCC Scientific Assessment, J. T. Houghton, G. J. Jenkins, and J. J. Ephraums, eds. New York: Cambridge University Press.
Lashof, D. A., and D. A. Tirpak. 1991. Policy Options for Stabilizing Global Climate. Washington, D.C.: U.S. Environmental Protection Agency.
Leopold, A. 1933. The conservation ethic. Journal of Forestry 31:634–643.
McIntosh, D. H. 1972. Meteorological Glossary. New York: Chemical Publishing Company.
Mearns, L. O., S. H. Schneider, S. L. Thompson, and L. R. McDaniel. 1989. Analysis of climate variability in general circulation models: Comparison with observations and changes in variability in 2XCO2 experiments. In The Potential Effects of Global Climate Change on the United States, Appendix I, J. B. Smith and D. A. Tirpak, eds. Washington, D.C.: U.S. Environmental Protection Agency.
Mearns, L. O., P. H. Gleick, and S. H. Schneider. 1990. Climate forecasting. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
National Research Council. 1983. Changing Climate. Washington, D.C.: National Academy Press.
Nordhaus, W. D. 1990. Adaptation: The role of government. Working paper prepared for the Adaptation Panel of the Panel on Policy Implications of Greenhouse Warming, February 12, 1990.
Nordhaus, W. D. 1991. To slow or not to slow: The economics of the greenhouse effect. Economic Journal 101(407):920–937.
Riebsame, W. E. 1988. Assessing the Social Implications of Climate Fluctuations: A Guide to Impact Studies. Nairobi: United Nations Environment Programme.
Rind, D., R. Goldberg, and R. Ruedy. 1989. Change in climate variability in the 21st century. Climatic Change 14:5–38.
Schneider, S. H., P. H. Gleick, and L. O. Mearns. 1990. Prospects for climate change. In Climate Change and U.S. Water Resources, P. E. Waggoner, ed. New York: John Wiley & Sons.
Smith, J. B., and D. A. Tirpak, eds. 1989. The Potential Effects of Global Climate Change on the United States. Washington, D.C.: U.S. Environmental Protection Agency.
Waggoner, P. E. 1989. Anticipating the frequency distribution of precipitation if climate change alters its mean. Agricultural and Forest Meteorology 47:321–337.