Over the more than four-and-a-half billion years since the formation of the planet Earth, its climate has remained remarkably stable, and has apparently sustained life for about four billion of those years. Throughout that long period the oceans and the atmosphere have maintained an uneasy equilibrium; the sun has been a sufficiently steady source of heat so that the oceans have neither boiled their water away into space nor frozen down to the equator-fates that many other planets and satellites of the solar system have suffered.

Yet even in the recent past there have been dramatic shifts of climate. A mere 18,000 years ago most of Canada and northwestern Europe were covered by great ice sheets several kilometers thick in places (just as Greenland and the Antarctic are now). In fact, such ice ages and the kind of inter-glacial periods that we are in now have alternated approximately every 100,000 years for the last three million years or more.

If we turn the clock back more than 15 million years, however, we find an Earth with hardly any permanent ice at the poles and a generally much warmer climate. That was the condition that prevailed about 90 percent of the time for the past several hundred million years.

These historical facts illustrate the ever-changing character of our natural climate. Up to now the factors that have determined the climate have been the sun, gradual changes in the Earth's orbit around the sun, and the complex interactions within the planetary "climate system"-the atmosphere, the oceans, the land, the ice of the polar regions, and living things.

Within roughly the last 50 years a new factor has been added-the activities of man himself. A growing accumulation of evidence has persuaded most of the scientific community that human activity may be contributing to a substantial change in the Earth's climate on a global scale. In particular, large-scale consumption of fossil fuels (coal, petroleum and natural gas) is leading to an accumulation of carbon dioxide in the atmosphere, which if continued appears likely to increase the average surface temperature of the Earth by several degrees over the next 50 to 70 years. And the release into the atmosphere of other gases arising from human activity may add significantly to this overall "greenhouse effect."

There are, of course, many gaps in our understanding of how the climate system behaves, and hence many uncertainties in this prediction of the future. But few climatologists still doubt that there will be a gradual trend toward a warmer Earth in the decades ahead, assuming we continue to add enormous quantities of carbon dioxide to the atmosphere. By the same token, we can anticipate shifts in the current patterns of precipitation due to changes in atmospheric circulation, though the details of these shifts are still unclear.

Since this prospective global change is the result of human activities, we could in principle avert or at least defer it if we decided that the likely consequences were "unacceptable." Or we could accept their onset and take measures to mitigate the adverse effects of the change and to capitalize on its beneficial effects.

Four years ago an informative article in these pages undertook to examine what the broad impact of man-induced climate change might be, particularly in terms of food production and ecological systems.1 At that time the scientific community was more divided on the issue, but a great deal of research has now made some elements of the future picture more clear. There remains an important need for more organized and systematic analysis, especially in translating overall global trends into a more precise picture of what can be expected for the climate of specific regions of the world. But the trend itself is now so unmistakable that it is time to broaden the analysis and to unite the research and judgment not only of physical scientists but of a whole range of disciplines including history, geography, political science and economics.

To that end, the present authors-one a climatologist, the other a political scientist-published in early 1981 a book that sought to present an overview of the situation and the present state of knowledge concerning it.2 This article seeks to go further and to present, for what may be a wider audience, a more detailed picture of the possible shape of future climate change, followed by the implications of such change for the habitability of specific regions and nations of the earth in terms of three specific subjects: agricultural productivity, ecology, and human health and disease-which lead to a more speculative discussion of possible climate-influenced migration.

Since energy demands and practices lie at the root of the problem, it is then necessary to look especially at the interaction of these factors with progressive climate change, and the degree to which energy conservation and reliance on energy sources that do not produce carbon dioxide might mitigate or postpone the degree of climate change one would now foresee.

Finally, the article suggests some of the possible elements of a "rational" approach by our political, industrial, agricultural and scientific leaders. In the words of Immanuel Kant, "the whole interest of reason, speculative as well as practical, is centered in the three following questions: (1) What can I know? (2) What ought I to do? and (3) What may I hope?"


It used to be generally doubted that people, so insignificant in size compared to their planet, could have any real influence on the global environment. But for at least two generations it has been clear that human activity has significantly altered the climate of our large cities and the regions downwind, and that we have aided the spread of deserts and decimated large parts of the forests that used to cover vast areas of both the temperate zones and the tropics. Such actions have altered the heat and water balance on a regional scale. And now we are influencing the climate of the Earth on a truly global scale.

Burning fossil fuels converts carbon that has been locked in the Earth for tens of millions of years to carbon dioxide and water vapor. The result has been an increase of about 20 percent in the carbon dioxide content of the atmosphere since the start of the Industrial Revolution, most of it occurring in this century. Whereas the atmospheric fraction of carbon dioxide is estimated to have been approximately 280 parts per million prior to 1900, it is now reaching 340 parts per million, or 0.034 percent by volume.

Since 1900 the release of additional carbon dioxide from fossil fuel burning has risen inexorably. From 1900 to 1973 the average annual rate of increase was roughly four percent, but since that time the rate of increase has slackened to about 2.3 percent per year.3

Even if there should be no further increases, present levels of release are bound to deposit a large fraction of the new carbon dioxide in the atmosphere. In aggregate terms, the world was releasing 1.6 billion tons of carbon per year (in the form of carbon dioxide) in 1950, and 5.3 billion tons in 1980.4 The resulting increase in atmospheric carbon dioxide during that period has averaged one part per million per year. Comparing the additional amount that can be accounted for in the atmosphere to the amount released, the airborne fraction has averaged about 60 percent of the total.5

The other 40 percent of the added carbon dioxide must have been taken up mostly by the oceans. The oceans represent a sink for carbon dioxide that is some 60 times larger than the atmosphere; eventually almost all of the added carbon dioxide will end up in the oceans. However, the processes of removal from the atmosphere, limited by the rate of mixing of the large volume of deep ocean water with the surface mixed layer only a few hundred meters deep, are extremely slow and gradual, acting over a period of 500 years or more. Several studies have been made of the oceanic uptake and mixing processes and, while there remain a number of uncertainties, it seems likely that in the next 50 to 100 years the oceans will continue to take up a little less than half of the new carbon dioxide added each year.6

One must also take account of the influence on carbon dioxide levels of the Earth's biosphere. Photosynthesis in the plants and trees of the world is a major sink for carbon dioxide; until the last century there was a rough balance between the take-up of carbon dioxide by photosynthesis and its return to the atmosphere by decay or burning. However, the forests of the world, especially those in the tropics that contain 80 to 90 percent of the living biomass, have been cut down rapidly in recent years, with further deforestation in prospect. There now rages a fierce controversy over whether this deforestation may be reducing the significance of the biosphere as a sink for carbon dioxide, and whether it may actually mean that the decay or burning of all that wood constitutes a net source of "new" carbon dioxide entering the atmosphere.

One widely quoted estimate placed the biospheric source as equal to or even larger than the fossil fuel source.7 It now seems, however, that estimates of current global deforestation were based on too scanty data and were probably exaggerated-though the problem is indeed very serious in some developing countries for other economic and environmental reasons-and that regrowth of forests and the sequestering of carbon in the form of charcoal (which lasts for a very long time) has been underestimated.8 The matter is still not settled, but we would probably not be making a significant error if we assumed that the oceans were the main sink and that fossil fuel burning was the main source.

This is the past and present situation regarding atmospheric carbon dioxide. Let us now address the question of its future levels and the effects of a continued rise on climate.


Just before the turn of this century a prominent American geologist and president of the University of Wisconsin, T.C. Chamberlain, and an equally prominent Swedish chemist, S. Arrhenius, independently pointed out that carbon dioxide absorbs infrared radiation from the surface of the earth that would otherwise escape to space, and then reradiates some of this infrared energy back downward.9 They noted that this "greenhouse effect" should raise the temperature at the surface above what it would have been in the absence of the added carbon dioxide (though the analogy to a greenhouse is far from perfect), but no data then existed to quantify the magnitude of such an effect.

For reasons that are hard to understand, this startling hypothesis attracted relatively little attention until the 1960s, when a number of scientists began to develop quantitative theories to explain the way our climate system works. In 1967 Syukuro Manabe and Richard Wetherald of Princeton published an estimate of the change of average surface temperature that would take place with a doubling of carbon dioxide from the assumed pre-1900 level.10 In this calculation they correctly took into account the very important fact that a warmer atmosphere would hold more water vapor, and that water vapor is another good absorber of infrared radiation.

This estimate has been checked by progressively more sophisticated and complete theoretical models of the climate system developed by Manabe and his colleagues as well as by a number of other groups.11 The answer remains about the same: doubling of carbon dioxide levels in the atmosphere should produce an average temperature rise of the earth's surface of approximately 3° centigrade, with an estimated uncertainty of plus or minus 1.5°C.12

It must be emphasized that this estimate of a warming from increased carbon dioxide refers to the average for the globe; regional changes will undoubtedly be different. Both model results and experience with the real atmosphere indicate that the warming in the Arctic will be some three times larger than the average, and in the temperate zone of the Northern Hemisphere, where a large number of the developed countries lie, the change for a carbon dioxide doubling should be 4° to 6°C. In the Southern Hemisphere, because of its relatively larger ocean area and an Antarctic Continent that will initially be much less affected than the Arctic, the warming trend would be smaller than in the Northern Hemisphere. (Other regional differences will be discussed later, especially those concerning rainfall on a warmer Earth.)

These estimates have dealt only with the effect of carbon dioxide release. But carbon dioxide is not the only infrared-absorbing gas that we are adding to the atmosphere. Other products of fossil fuel combustion, notably hydrocarbons, carbon monoxide, and nitric oxide, react photochemically to form ozone and some methane, both of which add to the greenhouse effect. Still another very persistent and infrared-absorbing gas is used as the propellant in aerosol spray cans and as the working fluid in refrigerators-the chlorofluoromethanes (sold under the trade name "Freon"). There is also the added nitrous oxide produced from extensive use of nitrogen fertilizers. A recent estimate of the effect of all these other gases suggests that they have already contributed to the greenhouse effect and that their continued release could increase the temperature rise by half again as much as carbon dioxide alone.13

Given the increased concentration of carbon dioxide and other infrared-absorbing gases in the atmosphere that has already been observed, why have we not experienced a global warming by now? Theoretically the greenhouse effect should have caused a warming of about 0.5°C in this century, and yet there was a pronounced global cooling trend between 1940 and 1965. Inevitably this question has been raised, and the apparent contradiction has been invoked by some skeptics as a refutation of the whole concept of the greenhouse effect.

Two sets of studies have now furnished apparently conclusive answers. First, data on the retreat of Antarctic sea ice have demonstrated a clear long-term trend in that area that is consistent with the expectation of a Southern Hemisphere warming.14 And second, analyses of other natural influences on global mean surface temperature have revealed sporadic or cyclical factors whose cooling impact apparently more than offset the greenhouse effect in the years that showed the overall cooling trend.15 These natural influences on climate are primarily major volcanic eruptions, which are known to place enough particles in the stratosphere to attenuate solar radiation for several years at a time, and fluctuations in the total radiation from the sun.16

In short, as the fluctuation due to these natural factors (or "climatic noise") has been sorted out, a clear "signal" from increasing carbon dioxide is indeed revealed. While the data do not conclusively prove a warming effect from increased carbon dioxide, the temperature observations over the past 100 years are at least consistent with the notion of the greenhouse effect.


Let us now combine these theoretical calculations of global warming with estimates of the rate of increase of carbon dioxide that may occur over the next 20 to 100 years. Levels of carbon dioxide in the atmosphere are bound to increase under almost any circumstance; the issue of pace and timing is, however, obviously of great importance.

At this point, the uncertainties surrounding future energy demand and supply are such that any projection beyond the year 2000 can be challenged. A reasonable upper limit might be a continued (or resumed) increase in worldwide fossil fuel use at this century's historical rate of four percent per year, at least to the year 2000. Although this rate of increase would be higher than in recent years, it could conceivably take place if there were a major further increase in the use of coal (which releases slightly higher amounts of carbon dioxide per energy unit than oil or natural gas).

A reasonable lower limit might rest on some variation of Amory Lovins' "soft energy path," which assumes a progressive reduction in fossil fuel demand due to conservation and use of renewable energy sources.17 While it seems unreasonable to expect a sudden shift to such a soft energy path by all countries of the world, it is quite conceivable that there could be a steady decline in the rate of increase of fossil fuel use until at some time in the not-too-distant future the rate of increase would reach zero, and after that the use of fossil fuel would actually decrease with time. Accordingly, the lower limit is based on the present rate of increase dropping progressively to zero in roughly 50 years and becoming negative thereafter. (The assumptions involved in arriving at these upper and lower limits of future fossil fuel use have been discussed in an earlier paper by one of the authors.)18

Figure 1, opposite, shows the range of temperature increases that might result from carbon dioxide alone-in terms of global temperatures and temperatures in the Arctic. (The centered trend line at the right of the Figure represents a "best guess" that fossil fuel use would continue to increase at roughly two percent per year in this time frame.) The message is fairly clear: possibly before the year 2000 and in any case before 2020-well short of a doubling of atmospheric carbon dioxide-the world may be on the average warmer than at any time in the past 1,000 years or more-that is, 1°C warmer than now. This "first stage" of climate change would stem from a rise from the present 340 parts per million to an atmospheric carbon dioxide fraction of about 360 to 370 parts per million. (Figure 1 also contains an estimate of the probable effect of carbon dioxide in the present century: the solid line for 1900-1980 reflects observed experience in the Northern Hemisphere, while the dashed line underneath indicates what the actual temperature might have been in the absence of the additional carbon dioxide released in this period.)

While a 1°C average change does not appear at first glance to be very significant, consider first that this change will be amplified toward the poles, and at temperate and high latitudes in the Northern Hemisphere it may be 2° to 3°C. In the temperate zone such a temperature change would correspond to roughly a four degree difference of latitude, or the present mean temperature difference between such pairs of cities as Copenhagen and Paris or Boston and Washington. As the trend continues to a doubling of the pre-1900 level of carbon dioxide-in what might be called a "second stage" of climate change-Boston would approach the same mean temperature as Miami has at present, while the southern parts of the United States and Europe would become truly tropical. The world would be returning to the climate of a period at the dawn of recorded history, the Altithermal, some 4,500 to 8,000 years ago, when the world was definitely warmer and regional climates were very different from now.

Changes in solar activity in this interval could hasten the warming, according to a theory based on solar cycles;19 conversely, a period of intense volcanic activity (or a major nuclear war) could reduce the effect. And recall that the addition of other infrared-absorbing gases could add another 0.5°C in the first stage of change alone.

Figure 1 does omit one factor difficult to quantify. The oceans have a large thermal capacity, and a lag of about a decade is likely as the upper layers of the world's oceans gradually warm up.20 This oceanic lag is expected to be somewhat shorter in the tropics than at high latitudes; and the central parts of the major continents, far removed from the moderating effect of the oceans, should be less affected by the delay than places along the coasts. In any case, this factor can only introduce a modest delay in the warming.

This, then, indicates the time scale for the first stage of the climate change that is anticipated. There is, as noted, good evidence that the change is already taking place, but has been partly disguised by natural fluctuations, so that we are not yet generally aware of it. There will be little doubt about its reality when the change becomes larger than the fluctuations, and this should be by the turn of the century, perhaps sooner.


While it is the heat balance of the atmosphere that is directly affected by increasing carbon dioxide and its greenhouse effect, temperature is of course only one feature of weather and climate. Rainfall and evaporation, together determining soil moisture, are even more crucial, especially for growing food.

Though it is convenient to deal with average conditions, the variability of climate is a very real factor as well. Those who grow food, livestock and forests are accustomed to gambling on the good years predominating over the bad years. It seems likely that variability of the climate will decrease as the temperate latitudes approach a more "summerlike" condition, but climate theory does not give much guidance on this point.

Let us first look at a rough picture of the present world situation in terms of water-deficient and water-surplus zones. Such a picture is presented in Figure 2, opposite, based on the balance between water supply from precipitation and the water requirements of plant life, which hydrologists refer to as "potential evaporation."21 Arid and semiarid regions are those where annual precipitation is less than potential evaporation.

It is clear that patterns of natural vegetation and agriculture are greatly influenced by water balance, though soil and temperature also play important roles. The map shows that many developing countries are situated in water-deficient zones, which usually means that agriculture is limited to the rainy season or may be impossible without irrigation. The tropical areas of water surplus tend to be occupied by rain forests, and here cultivation is hindered by the excess water leaching the nutrients from soil that has been cleared of trees.

As to how that picture of the present situation might change, theoretical climate system models are not yet adequate to tell us, although they do give some useful hints. We can gain some additional hints by looking at reconstructions of the Altithermal Period when the climate was generally warmer.22 Then the Sahara was not a desert, but a kind of savanna able to support nomadic tribes with their cattle; North Africa and the Middle East were generally more favorable for agriculture; and the present Rajasthan Desert of northwest India was a region where several large cities prospered (cities now swallowed by sand). Still another set of useful hints comes from studies of particularly warm years or seasons during this century, for which good meteorological records of temperature and rainfall have been kept.23 From these records we can see how those anomalously warm times differed from the long-term average.

These three sources have been used in constructing Figure 3 above.24 The process was somewhat subjective and also relied on some general reasoning based on concepts of how the large-scale circulation patterns would change with a global warming and a decrease in the temperature difference between the equator and the poles. (For example, it is fairly clear that the monsoon circulation that determines so much of the climate of Asia would be stronger and more regular, and that the storm tracks would on the average move poleward.) The resulting map should not be taken as a prediction of what will happen, but rather a scenario of the future climate describing what could happen.

The implications of this scenario are fascinating. It suggests that large areas of Africa, the Middle East and India, as well as a substantial portion of central China, might cease to be water-deficient areas, or at least be much less water-deficient. On the other hand, there would probably be much drier conditions in the central section of the North American continent, and drier conditions virtually throughout the central and northern areas of the Soviet Union, so that it would be harder to grow wheat, corn, barley and other major food crops in parts of what are now the "food basket" areas of North America and the Soviet Union.

In short, the climate scenario shown in Figure 3 is a useful way to illustrate some of the shifts of rainfall and soil moisture that may occur as the world gradually becomes warmer, but one must be very careful when drawing conclusions from it. After all, climate change is but one of a great many changes that will be taking place in the decades ahead. Poverty, loss of arable soil and tropical forests, condition of water supplies, distribution of natural resources, population growth, and so forth are all factors that must be considered when assessing impacts of climate change, as we shall see in the next section.

Finally, a discussion of the many aspects of a global warming would not be complete without mentioning the possibility of a future rise in sea level-which has been prominently featured by the media. Studies have suggested that the ice sheets of Greenland and the Antarctic have been shrinking slightly in the past 100 years as the Earth grew slightly warmer, contributing to a rise of five to ten centimeters in sea level. In the case of the West Antarctic ice sheet, there is a possibility that it could actually begin to partially disintegrate with the much greater prospective warming of the Polar regions and the neighboring areas.

While a straightforward melting of these enormous masses of ice would probably take many thousands of years, it has been pointed out that most of the West Antarctic ice sheet (and portions of the larger one of the East Antarctic) rests on bedrock well below sea level. Thus, warmer ocean water could work its way under the ice sheet, causing it to relinquish its frozen hold on the bedrock, and to begin sliding toward the ocean, gradually disintegrating and then melting much more rapidly through fuller exposure to the water. A rise of five to seven meters in sea level would result if the entire West Antarctic ice sheet disintegrated, since this ice sheet now rises several kilometers above sea level.25

This could make the migrations of the historic past seem almost insignificant by comparison. However, it must be emphasized that glaciologists do not agree on the timescale for such an event. Whether the West Antarctic ice sheet would disintegrate in a matter of centuries or millennia following a major warming is still being debated, but in any case few, if any, experts now claim that the time involved could be less than about 200 years.26

It is very likely that Arctic and Antarctic floating sea ice will become less extensive as the warming progresses, a trend that is already evident in the Southern Hemisphere.27 The Arctic Ocean is now mostly covered the entire year by floating ice, but early in the next century this ocean may become ice-free in summer.28 However, in contrast to the massive ice sheets that rest on land, a melting of sea ice would have no influence at all on sea level. It is nevertheless significant that, so far as can be determined from studies of Arctic Ocean bottom sediments, that ocean has never been completely free of its ice pack for some three million years-the change that we are talking about may thus be larger than any previous one in that entire period.


Carbon dioxide-induced climatic change will affect many aspects of society. Let us look at what seem likely to be three crucial effects-on food production, global and regional ecology, and human health, disease, and comfort, also referring briefly to the question of possible migrations of populations.

Agricultural Production

The climate changes suggested in Figures 1 and 3 are bound to have a substantial influence on agriculture. Broadly speaking, it would appear that hotter and drier conditions in key growing areas of the temperate zone would have a negative effect (or at least require drastic adjustments), while in the more tropical areas that may become less arid, the effect would be favorable. But such generalizations are too simple: much would depend on the pace of change, specific offsetting effects, opportunities for agricultural migration within a given region, and new technology to cushion the impact.

The primary cause of the global warming, increasing carbon dioxide, is of itself an asset to plants. Operators of greenhouses have known for years that raising the level of carbon dioxide in the air of a greenhouse will cause most plants to grow faster and larger (provided they are not limited by water or nutrients). Furthermore they use water more efficiently, which will favor agriculture in semiarid areas. In the range of 100 to 300 parts per million above the present carbon dioxide concentration (which we can anticipate in the next century), the potential for increased photosynthesis is about 0.5 percent for each one percent increase in carbon dioxide.29 Thus, even by the turn of this century the increase in average growing rate may exceed five percent from this cause alone.

There have been many studies of the effects of temperature during the growing season, the length of the growing season, and soil moisture. Of these three factors, soil moisture is by far the most critical in the semiarid regions of the subtropics, whereas temperature and length of growing season are more often determining factors in agricultural regions at temperate latitudes. As a rough rule of thumb, a 1°C increase in mean summertime temperature corresponds to a ten-day lengthening of the growing season, which can clearly make a large difference in Canada, Northern Europe and the Soviet Union.

As a longer growing season opened up new areas, might present patterns of agriculture simply migrate? One careful study estimates that the U.S. corn belt would shift 175 kilometers north-eastward for each degree (centigrade) of increase in temperature, if precipitation remained unchanged.30 This is due to the fact that a peculiar situation exists in our Midwest wherein the gradient of summertime temperature is northward and the gradient of rainfall is westward. In the Soviet Union and Europe both gradients are northward, so the corresponding shift would also be in that direction.

However, there are clear limits to such shifts of agricultural areas. First, rainfall and soil moisture patterns would often change as well-the predicted drying of the American Middle West, for example-so that a given pattern of agriculture could no longer be practiced even by moving poleward within a region.

Second, types of soil may impose a serious restriction on the migration of agricultural practices. Soil type is determined by climate and plant growth over periods of many centuries. In the humid tropics, for example, rainfall far exceeds evaporation, and basic minerals are slowly leached out of the soil, making it acidic and (when the growing vegetation layer is removed by deforestation or erosion) unproductive for sustained agriculture. In most of the Arctic, on the other hand, where evaporation exceeds precipitation, calcium carbonate and other basic minerals accumulate in the soil, making it alkaline. The best agricultural soils are those with just enough leaching to remove the excess basic salts but not enough to remove the essential minerals from the clay. An example is the prairie soil of the midwestern United States, among the best in the world.31

Thus, in Canada the scope for northward migration of grain production with warmer temperatures is limited to the Peace River Valley of northern Alberta and a few other scattered localities where good soils exist but the growing season is currently too short. Longer growing seasons in Alaska could greatly increase its agricultural potential. In the Soviet Union there may be opportunities to extend agriculture northward, particularly in the alluvial soils of the Yakutian plain along the Lena River-provided that rainfall or irrigation water is adequate. The same opportunities apparently do not exist in Scandinavia because of poor soils.32

In the present agricultural areas of the temperate Northern Hemisphere, agricultural technology can provide farmers with a flexibility in the choice of crops and agricultural methods that could go a long way toward mitigating the effects of a climate change and adapting to it.33 Development of water resources and water conservation measures can also soften the change to drier conditions, especially in view of the slowness with which the change will probably take place. A case in point is the massive program being undertaken in the Soviet Union to redirect the waters of the major rivers now flowing northward to the Arctic Ocean southward to the semiarid regions of central Asia-particularly in Kazakhstan and Uzbekistan.

Weighing all these conflicting factors, it appears likely that the net impact of climate change would still be negative in some important producing areas of the Soviet Union, China, Canada and the United States. At the very least, there could be substantial migration and a heavy burden of adjustment measures. Conversely, the climate changes suggested by Figure 3 would permit much more effective agriculture in the areas of Africa, the Middle East and parts of China that become moister. But again, one would confront the fact that present deserts or near-deserts would in some areas take a long time to become suitable for agriculture or grazing. The Sahara and the Rajasthan Desert of India, for example, would probably not revert for a long period of time to the more favorable soil conditions they apparently enjoyed in the Altithermal Period, and the transformation of these deserts could only be gradual.

From a global standpoint, future drier and warmer conditions could occur in areas where roughly 50 percent of the world's wheat and 55 percent of the world's maize (corn) was grown in 1978. On the other hand, approximately 86 percent of the world's rice and 50 percent of the world's barley could be affected by wetter soil conditions, according to our climate scenario. Under generally warmer and wetter conditions, rice yields may increase in the major rice-producing countries of China, India, Indonesia, Bangladesh, Thailand and Japan. Rice production there may be encouraged by both more rainfall and longer growing seasons, the latter permitting multiple cropping in the vast areas of the temperate latitudes where only one rice crop can now be grown in the course of a summer.34

Equally important to food production may be the changing climate's effects on the frequency and severity of pest outbreaks. Currently, losses in agriculture and forestry due to pests are extremely high: roughly 37 percent for agriculture and 25 percent for forestry. One authoritative study is gloomy about the future:

Insect pest populations will generally increase with an increase in temperature. Some insect pests, for example, produce 500 to 2000 offspring per female and go through a generation in 2 to 4 weeks. With a warmer and longer growing season these pests may pass through an additional 1 to 3 generations. The exponential increase of some insect pest populations under the new favorable environment could seriously increase insect losses and make their control more difficult.35

With a global warming we can also expect changes in the frequency and geographical distribution of plant disease epidemics.36 It is important to expand the studies of climatic effects on plant disease in order to reduce losses when the expected global surface warming occurs.

In short, the future effects of climate change could result in major readjustments in global patterns of food production. While we cannot predict with confidence just how these readjustments will take place until we have a clearer picture of the regional shifts of soil moisture sketched in Figure 3, that scenario illustrates the dimensions of the changes in store for the world-including a possible migration of the present "bread baskets" of North America and Eurasia. On balance, it appears that the world's total ability to grow food may not suffer in the long run, provided the appropriate agricultural technology is applied and planning for improved water resource management can keep ahead of the change. But the long-range challenge to agriculture must be recognized and faced now.

Global Ecology

Agriculture for food and fiber production, discussed in the previous section, differs from the less-managed ecosystems of the world, such as grasslands, savannas, forests, tundra, alpine lands and deserts. A major difference is that agricultural systems usually depend on a few specialized plant species, whereas a natural ecosystem, or "biome," represents a diversity of plants and animals that interact and live in a natural balance-that is, until the balance is disturbed by humans or by climate change.

Mankind has already significantly influenced many biomes. Forests are managed for their wood products (in some places mismanaged and devastated), rangelands are grazed by cattle and thus modified, and semi-deserts are formed where overgrazing or bad agricultural practices have caused vegetation and topsoil to disappear.37

There are few, if any, ecosystems remaining in the world that are free from human influence. Among the biomes still close to their natural states are the tropical forests that have not yet been exploited and the unpopulated tundra areas in the Arctic.

How drastically and immediately a shift in climate affects a particular biome depends in part on whether the change is fast or slow relative to the life span of the dominant species.38 From this standpoint, climate change due to increasing carbon dioxide would take place on an intermediate time scale, allowing some time for adjustment of the biome. But ultimately the new warmer climate regime will probably last for many centuries, compelling long-term ecosystem responses; the effects of long-lasting climate change are likely to be dramatic and permanent.39

An obvious historical example is the transition in northern Africa from a grassland that sustained wild animals, and perhaps herds of cattle, to what is now the Sahara Desert-a change that seems to have occurred rather rapidly around 5,000 years ago.40 In a somewhat more recent and less spectacular case, about 3,500 years ago the boundary between trees and the Arctic tundra in northcentral Canada abruptly migrated southward about 300 to 400 kilometers, probably in response to increasing cold and more frequent forest fires that accompanied a general drying of the region and destroyed spruce forests.41

Among the specific ecological situations likely to be affected by future climate change are the two biomes still close to their natural states. The untouched tropical forests will be affected chiefly by new rainfall patterns, while the Arctic tundra biome will be influenced by both precipitation and temperature.

Basically, forests require relatively high precipitation.42 They consume more water than other kinds of vegetation, and transfer it back to the air through their leaves by evapotranspiration. In semiarid areas, where water is precious (see Figure 2), trees are extremely susceptible to small decreases in precipitation. Furthermore, many kinds of trees are vulnerable to insect pests, and infestations are markedly influenced by temperature and precipitation. Hence, these factors suggest that climatic change, especially changes in precipitation, will have a significant effect on the distribution of areas favorable for specific kinds of forests.

The deforestation that is taking place in some parts of the world, mostly in certain developing countries of the tropics and subtropics, is cause for concern because of its short-term environmental effect on the countries involved; even more serious in the long run must be the loss of soil and the extinction of many natural species as entire forest biomes are destroyed. The search for new generic strains of plants for agriculture relies heavily on the reservoir of natural species, or global "gene pool," and once a part of it is lost it is irreplaceable. In arresting the present dangerous trends, a higher level of carbon dioxide in the atmosphere and more rainfall in the subtropics-both of which enhance forest growth and allow it to spread into areas that are now too arid-could at least be helpful.

For the frozen tundra of the Arctic Basin, a warming induced by increased carbon dioxide would also have a very large influence. This biome is characterized by permafrost, which accounts for its low-growing vegetation, lack of trees, and poor drainage in summer. If a warming trend occurred in the tundra, the permafrost would slowly retreat, allowing trees to grow further poleward, as occurred during the Altithermal Period in Canada.

Tropical forests and Arctic tundra are two cases where the impact of climate change seems fairly predictable now. In other situations the effects will become predictable only as the shape of regional climates becomes clearer. On an overall basis, while many biomes will be displaced or modified by climate changes in the next 50 to 100 years, it is well to note that steadily expanding human activities and pressures will probably have an ecological impact rivaling that of climate change in this period.

Human Health and Disease

Climate relates to disease in many ways, perhaps most directly in that precipitation, humidity and temperature critically affect breeding conditions, growth rates, and biological diversity of many insect and bacterial species, including parasites affecting humans. This is particularly true of the serious disease agents now disproportionately concentrated in the poor and developing countries of the world, countries that tend to be in the tropics or subtropics. For example, the hookworm disease that afflicts an estimated 500 million people in tropical and subtropical regions depends on a hookworm larva whose life cycle is directly related to temperature and soil moisture.

As with agricultural pests, climate change will affect disease agents in a variety of other ways as well, and the precise effects will be distinctive for each species.43 About all that can be said with assurance is that the climate changes depicted in Figures 1 and 3 will probably complicate substantially the task of bringing under control the major diseases that still ravage millions in the poorer countries of the world.

A more general question is how effectively human beings will perform in changed climatic conditions. This question may not be particularly serious in areas already tropical, since the changes in temperature will be less there and since the change from semiarid to water-sufficient conditions, predicted for many tropical and subtropical areas, may improve the conditions for life and activity. But in the present temperate zones, especially in the cooler areas that would experience a sharp warming trend-for example, Boston becoming like Washington, and eventually like Miami-there is a real possibility that labor productivity will be affected, not to mention leisure activities or, at a more basic if intangible level, those behavior characteristics that distinguish "northerners" from "southerners" in almost every temperate zone country, from China to Italy to the United States. Studies lend support to the common belief that very high or very low temperature and humidity can negatively influence the capacity for physical performance, personal motivation and social behavior.44

Viewed from a historical perspective, there is the celebrated thesis of Arnold Toynbee's A Study of History that societies tend to flourish when they are subjected to just the right degree of challenge from their environment, the interplay theory of "challenge and response." Whereas tropical and frigid climates have often stunted the growth of what he called true "civilizations," the temperate latitudes have generally been more favorable to ever new responses and challenges of the environment. If this were the case, then over a period of time (decades to centuries) we might see the most thriving societies appearing further from the equator.

There is no real way to assess what would happen to people who stayed where they were in the face of the kind of changes suggested by Figures 1 and 3. In the most advanced countries the change might be cushioned (as present differentials in living space temperature between U.S. regions are) by the extended use of air conditioning in the summer-while reduced rigors from winter weather might improve productivity. One of the few serious studies in this area concludes that Western workers accustomed to temperate climates experience a loss of output of two to four percent for every rise in temperature of one degree centigrade-but this is surely only a rough guess as to what would happen after adaptation over a long period of time.45

A related question is whether the kind and degree of climate change suggested in this article could trigger substantial population migrations, both within countries and from one country to another. One obvious motivation for such migrations would be to get away from areas that could no longer produce food as they do now, or to move to areas where climate change was favorable for this purpose, and some of these we have examined above. But another motivation for migration could be simply a desire to seek as nearly the same climate as possible to what one now has-a subjective feeling as to one's "comfort zone."

Historically, the former factor-usually in the form of famine-has triggered many massive migrations. Egypt in the third millennium B.C., Mycenae around 1230 B.C., central Germany in the period 1360-1450, Iceland in the Middle Ages, and (within the vivid family memories of many Americans) the Irish potato famine of 1845-51-all are examples of climate factors contributing to famine conditions that led to wholesale migration, loss of life, or both together.46 The degree and permanence of the climate changes suggested in Figures 1 and 3 may be greater than in any one of these cases; on the other hand, the adaptive capacity of societies today, at least in the temperate zones, is much greater than in early historical periods-one has only to note that whereas the Dust Bowl conditions of the 1930s drove hundreds of thousands of farmers from Kansas, Texas and Oklahoma, later serious droughts in the same regions have not caused nearly so much disruption.47

But when one combines food production and comfort factors, one still has the sense that large-scale migrations will be entirely possible, and that some of these may seek to cross present national boundaries, in a generally northward direction in the Northern Hemisphere. Canada coping with a possible American migration of several million people looks like one of the more easily resolved cases from a political standpoint alone. One can visualize other major movements of population as a result of the changed climate situation-migrations away from areas that become too dry and hot for food production, as well as repopulation of deserts and savannas previously crippled by drought and ecological deterioration. Such population shifts may be loaded with potential for political and social disruption, as evidenced by current migrations to bulging cities all around the world already overburdened by problems of unemployment, housing, transportation, sanitation and food distribution.

Nevertheless, the picture of population migrations triggered by continual climatic fluctuations or gradual changes should not be perceived as totally gloomy. Indeed, there are numerous bright instances of societies and civilizations adapting to droughts lasting more than several years and to shifting floodplain environments. The survival of the Tuareg herders and Hausa farmers in the semiarid Sahel in precolonial times, and the long course of Egyptian civilization, exemplify the dynamism of adaptive systems.48

Added together, carbon dioxide-induced climatic impacts may have major effects on human activities, but some aspects of the projected climatic changes could be beneficial. Put simply, future carbon dioxide-induced climatic impacts will cut both ways-to weal and to woe. There will inevitably be both "winners" and "losers."

Finally, a word is necessary on the implications of the possible melting of the West Antarctic ice sheet and Arctic sea ice. The latter, as we have seen, would have no direct influence on sea level, and, if the experts are right, anything like a full disintegration of the Antarctic ice sheet is at least 200 years off. But when and if it does come the effects would be devastating for the 30 percent of the world's population that now lives within about 50 kilometers of an ocean or sea.49


Turning now to the question of energy supply and demand in the context of future climate change, the first and rather obvious point is that it is the world's insatiable quest for ever-increasing energy supply that is the fundamental cause of the rise in carbon dioxide. At the same time the resulting climate change, when it occurs, will influence the demand for various forms of energy.

On the supply side, the world has been depending largely on fossil fuel for its energy-in 1975 only about eight percent of global primary energy was derived from other sources: nuclear, hydropower and solar sources combined.50 As we have noted, the rate of increase of fossil fuel consumption has slackened in the past eight years to a little over two percent per year-having averaged about four percent per year from about 1900 to 1973.

Most energy projections assume that the world will not return to the more rapid earlier growth rate. Figure 1 has shown reasonable "high" and "low" estimates, with a not unreasonable "best guess" being a continued increase of around two percent persisting well into the next century. The actual course that will be followed will depend on a number of factors that are very difficult to assess, including efforts to conserve energy and use it more efficiently, willingness to invest more heavily in nuclear and renewable energy alternatives, the rate of depletion of oil and gas and ability to substitute coal, and so forth. Whether the community of nations will be able to take any purposeful action to control and limit fossil fuel use to slow down or avert the climate change is also an important consideration discussed in the next section.

Energy supplies might be bolstered and fossil fuel use reduced if renewable energy resources were extensively developed; there are persuasive reasons for pushing such developments as fast as possible in any event. But some renewable sources may be more vulnerable to vagaries of climate than the present energy sources. Solar power is degraded on cloudy days, hydroelectric power suffers when there are droughts, and wind power depends on good winds. Furthermore, these are not independent weather factors. For example, the period of drought on the West Coast of the United States in 1976-77 and 1977-78 was also a period of generally weak winds, since both were determined by a more or less stationary "blocking pattern" in the pressure field. Clearly, climate variability and change should be taken into account when planning for such renewable energy resources.

Turning now to energy demand and the effects of a climate change on it, the outlook for the future is also difficult to assess. Energy projections generally assume that per capita energy demand will rise somewhat more in the developing countries than in the industrialized ones, and whether this will in fact take place will probably depend more on such factors as the economic squeeze on their financial capacities and the future price of fossil fuels than on factors associated with climate change.

In the developed world, however, the effect of climate change on energy demand could be substantial. Most of the industrialized countries are located in temperate latitudes, and the greatest single use of energy in these areas is for space heating and cooling. Although only 18 percent of the total U.S. annual energy demand is for space heating, in Canada the figure is 29 percent and in Denmark 50 percent.51 This demand for heating energy depends very markedly on the construction of the buildings, of course, and more attention is usually given to good insulation in colder climates.

Roughly speaking, demand for space heating is greatest in areas of the world with mean annual temperatures between 2°C and 18°C, and this includes most of the industrial areas of the world. At warmer temperatures the demand per capita is less, virtually ceasing at 18°C mean temperature; and so few people live poleward of the 2°C mean temperature isotherm that total energy demand for heating there is also negligible.52

In a carbon dioxide-induced climate warming, this band between 2°C and 18°C would generally shift poleward, and as we have noted, the average change in temperate areas would be an increase of 2°C in the first stage of a global warming (depicted in Figure 1), and perhaps 4°C in a second stage associated with a doubling of present atmospheric carbon-dioxide levels. How this might affect heating demand is at least suggested by recent U.S. experience in the opposite direction: in the unusually cold 1976-77 winter in the eastern United States a 1.8°C drop in seasonally averaged temperatures created 22 percent more heating requirements than in preceding years.

Climate change may also affect energy demand by creating additional needs to pump water for irrigation. In areas where Figure 3 suggests declines in rainfall, there would probably be not only a greater need for energy to handle irrigation projects already installed, but also a substantial expansion in such projects. Again there is a recent U.S. example from the 1976-77 experience, when the rainy season in California was much drier than usual and water allotments to agriculture were reduced by 60 percent. (In a normal year 80 percent of the developed fresh water supply in California is used for crops.) Consequently, farmers were forced to replace the deficit by pumping up more ground water for irrigation, and the additional energy required for this operation was about one billion kilowatt hours in California alone.53

And we might note that in 1976-77 the cold in the eastern part of the United States and the drought in the western part were caused by the same large-scale general circulation: a persistent ridge of high pressure in the west and a trough of low pressure in the east. This kind of shift of the semipermanent ridges and troughs could occur again; we can expect other shifts to occur during long-term climate changes that would have major influences on regional climate.

The above examples of climatic impacts on U.S. energy demand were all due to extreme, short-lived anomalies that caused temporary hardships and unexpected costs; there are equally good examples from other countries. Although we have some experience in coping with such temporary climate variations, dealing with a more gradual, long-term change is likely to be quite different. Given enough time to prepare, agriculture and industry probably can adapt by modifying their practices. Furthermore, rising costs of energy will tend to reduce demand; people will be more willing to conserve. In other words, economic and social forces as well as climate will shape future energy demand and supply.

To summarize, a global warming will ease the requirements for space heating in temperate latitudes in winter but increase requirements for cooling, shifting the demand from the direct use of fuel to electricity. Patterns of energy use will shift but not move uniformly poleward; there will be marked regional differences in the temperature and precipitation changes due to related changes in atmospheric circulation. When our climate scenarios have been refined somewhat, the net effects on energy demand could be estimated, since the relationships between energy use and seasonal "degree days" have been reasonably well established. However, the move to conservation and use of renewable energy resources, especially solar heating, would have to be taken into account, as well as future attitudes regarding nuclear energy.

The direction that future energy policy will take in each country seems impossible to predict. The choice between the "soft energy path" and expansion of centralized (mostly nuclear) high-technology energy sources has been hotly debated for several years.54 In the end the choice in most countries will probably not be based on the systems analyses of the technical experts, but rather dictated by shorter-term goals and constraints and economic realities. In the debate so far, the issue of climate change from fossil fuel use has been given passing attention and lip service, but it has hardly ever been advanced as a crucial "selling point." This issue could become more central when (and if) climate change has been perceived as an inescapable reality and its potential impacts are better understood.


Clearly, the strategies or policies that we may adopt to deal with the carbon dioxide problem depend in large part on the way we perceive it as well as on the adequacy of the national and international decision-making processes to address it. Even if the future could be predicted and precise estimates of future carbon dioxide-induced damage could be made, efforts to control fossil fuel use and carbon dioxide emissions on a worldwide basis would face extremely tough political opposition. Both consumers and powerful vested interests in the fossil fuel production and use industry would oppose any moves to curtail the availability of this conventional and convenient source of energy. Given the great uncertainties in our scientific predictions of future temperature and precipitation changes, any concerted worldwide agreement to limit fossil fuel use seems out of the question for some time to come.55

And the same negative conclusion almost certainly applies to any early attempt to act on a number of proposals that have been advanced for removing carbon dioxide from the stack gases of fossil fuel-fired power plants and sequestering these billions of tons in the deep ocean or depleted oil and gas fields. While there is no doubt that this is feasible from a purely technical standpoint, the most efficient means presently available for scrubbing the carbon dioxide out of stack gases requires nearly 50 percent of the combustion energy of the fuel, and that is only the first step.56 Even a regional project of this sort-a dumping facility off the coast of Spain has been suggested, drawing on power plants of Western Europe-would be a mammoth undertaking, and the economic hurdles and political obstacles seem virtually insurmountable, at least under present conditions.

Thus, the action focus in the short and medium term should be on strategies for arriving at a clearer picture of the future, and for mitigating or delaying the adverse effects of prospective climate changes and taking advantage of their favorable aspects. Such strategies should aim to be effective over a wide range of eventualities, since the problem areas often cannot be pinpointed as yet. Moreover, we must be realistic about which courses will be acceptable and feasible.

What, then, ought we to do? Three types of strategies seem to make clear sense:

- Strategies that lead to improved choices: examples are the organization of environmental monitoring and warning systems designed for early detection and attribution of carbon dioxide-induced climate change, provision of improved climate data and the knowledge of their application (especially to developing countries).

- Strategies that help to slow the increase of carbon dioxide: examples are energy conservation, adoption of renewable (non-fossil fuel) energy sources, increased use of nuclear energy, and reforestation (which protects soil and also takes carbon dioxide out of the atmosphere, but probably could not remove enough to counteract the current production rate).

- Strategies that increase resilience to climate change: examples of such measures are the application of agricultural technology, protection of arable soil, improvement of water management, and maintenance of adequate global food reserves.

It should come as no surprise that these measures seem like "conventional wisdom"-they are good ideas, all of which have already been acted on to a greater or lesser degree. They will help us to cope with the inevitable short-term vagaries of the climate as well as a longer-term climate change, and will make our food and energy systems more reliable. Hence, these measures should be adopted in any case. The issue of carbon dioxide-induced climate change may serve as a stimulus to employ them sooner.

The drafters of the World Climate Programme, adopted by the World Meteorological Organization in 1979, recognized the need for such actions, and the component programs for World Data and Applications are designed to meet the needs of developing countries in the first area above.57 Others must probably be adopted at national or regional levels, such as application of agricultural technology, reforestation and protection of arable soils, and improved water management.

Significant non-economic factors include an educated public and its leaders, together with long-term planners. An important aspect of this is the ability and willingness of planners to recognize the signals of coming change and to interpret those signals for the institutes they serve as well as the general populace. If the scope of such anticipatory actions is to be remotely adequate to the likely extent of climate change, the burden could be substantial. While many of the actions can best be undertaken at the national level, regional cooperation could make an enormous difference, and the development of the technology and forecasting systems clearly calls for the widest possible forms of international cooperation, making use of existing U.N. organizations in particular.


But if concerted worldwide action programs can only be limited for the immediate future, it is not too early to speculate on some of the implications of the climate changes now foreseen, for the relations among nations and possibly for the development of relevant principles of international law and shared responsibility.

Here a crucial element of the problem is that a small number of industrialized countries are likely to remain the principal sources of the additional carbon dioxide, which in turn is the major agent of climate change. Between 1950 and 1976 the United States, the Soviet Union, China, the United Kingdom, West Germany and Japan were the largest carbon dioxide-producing countries.58 These and similar nations are likely to continue for many decades to be the major cumulative contributors of carbon dioxide to the atmosphere, and this fact alone suggests that developed nations will bear much of the responsibility for the new climatic regimes, should they occur.

One aspect of this is that the carbon dioxide/climate problem could become a significant source of controversy, both between one developed nation or region and another, and between the developed nations as a group and any developing nations which find that the changes have adversely affected them. The current controversy between Canada and the United States over measures to control "acid rain" suggests what might be in store on a much wider scale. As yet there is no international mechanism to consider a given country's responsibility for the climate change, let alone to set carbon dioxide standards, establish control measures, and enforce what some would assert as a legal claim for having affected a nation's climate adversely. Liability claims, sanctions, and indemnity awards are probably unworkable ways to resolve carbon dioxide-related problems such as disruptions in trade, environmental damage, and population relocations. By the time such claims are brought to court, settled, and (if possible) enforced within a future international legal framework, the damage will probably have already occurred. Hence we should not expect much of international law when it comes to resolving cases that deal with adverse climatic impacts suffered by regions or nations and attributed to increased atmospheric carbon dioxide.

But legal remedies are only a corner of the problem. As climate change unfolds, and is perceived to be the result predominantly of the energy practices of the major industrialized countries, there could be strains and controversies highly disruptive to present patterns of close relationship and alliance, as well as to the overall structure of global international cooperation. Will the major nations responsible for most carbon dioxide release show awareness of the problem and make moves to mitigate it? Will they wish to avoid potentially serious damage to their international standing and influence?

Finally, the time is almost bound to come-if not in the next decade, then surely by the turn of the century-when it will be both possible and necessary to take a much harder look at potential national and international measures to reduce fossil fuel consumption. The "first stage" of climate change postulated in this article is an average global temperature change of "only" 1°C, associated with an increase in atmospheric carbon dioxide from the present 340 parts per million to a level of 360 to 370 parts per million. Figure 1 suggests that this may happen in roughly the next 20 to 30 years. And even if concerted programs were adopted immediately, this scale of change would probably still occur within at most 40 years.

But the pace of change to a "second stage" of climate change-a doubling of the carbon dioxide fraction to a level of, say, 560 parts per million-producing an average global temperature change of 2 to 3°C (with a change of 4 to 6°C in temperate zones and even greater changes at the Poles)-is something the world may see strong reason to at least postpone for as long as possible. If one looks at the impact of continued growth rates in fossil fuel consumption over a period of 50 to 100 years, the effects are indeed striking, and are depicted for illustrative purposes in Figure 4 below.59

The first thing to note in Figure 4 is the very great impact of alternative future growth rates in the developed countries. If their consumption of fossil fuels were to resume the four-percent growth rate of 1900-73 period, the doubling would still take place not later than 2030 on any assumption of the growth rate for developing nations-while if it were to grow at a three-percent rate, the doubling would come roughly between 2030 and 2040.

The date of doubling of the carbon dioxide fraction is somewhat less sensitive to assumed growth rates for fossil fuel consumption in the developing nations. Yet even for these countries, if the growth rate were to be as high as six percent annually-which happens to be close to the World Bank's latest projection for the next ten years-the doubling should take place shortly after 2030, with even as little as a one-percent growth rate for developed countries.

Moreover, if we look at possible lower growth rates, the differences are still substantial. If we assume growth rates of approximately two and three percent per year in developed and developing countries, respectively, the time of doubling would be approximately 2050, while rates of one and two percent per year for developed and developing countries, respectively, would defer the date of doubling to 2080. And if fossil fuel consumption could be levelled off completely or reduced, the advent of the "second stage" would of course be further postponed.

So it is clear that "low growth" or "no growth" scenarios for fossil fuel consumption would substantially delay the "second stage" impact of a carbon dioxide-induced warming trend. The time available and needed for policymakers to consider implications of the climate change in a deliberate, purposeful policy design would then be extended considerably. It may even be sufficient to enable the transition in energy production to non-fossil sources to take place without massive adjustments toward non-fossil supply expansion or lower economic growth.

And here one returns to the probability that many of the adverse effects of the kind of climate change that appears likely would be felt most seriously in such countries as the United States, the Soviet Union and China, particularly in terms of their food-growing capabilities. These three countries happen to be those with the largest reserves of coal, as well as being major users of energy.60 Thus, many of the countries chiefly responsible for the climate change would also be its principal victims, and at some point this fact could be a very great inducement to concerted and drastic action to deal with such factors as the expanded use of coal as a major energy source. So far, carbon dioxide and climate change have not figured significantly in the setting of energy policy in the industrialized nations. As the climate change and its impacts are more clearly perceived, this situation could and should change.


We must view the prospect of a carbon dioxide-induced climate change against a backdrop of other equally important environmental and societal problems facing the people of the world. In the first half of the next century, as the environmental changes will probably become increasingly more apparent, the world could well contain twice as many people, consume three times as much food, and burn four times as much energy. Our future problems will be serious even without a shifting climate.

It now appears that the possible global climate change may be very disruptive to some societies, though not generally disastrous. It may trigger shifts in agricultural patterns, balances of trade, and habitual ways of life for many people-and eventually, a few centuries from now, may even force abandonment of low-lying lands due to a rise of sea level. Ideally, we may hope that the countries of the world would unite to control and limit the use of fossil fuels, thereby averting or at least delaying this disruption.

The political motivation to do this is apparently lacking, however, nor is it likely to develop in the foreseeable future. There are powerful transnational as well as local interests that would surely oppose such measures, and there is no international machinery that could make such a drastic decision, much less enforce it.

The alternative, then, is to prepare for the climate change in store for us. We can hope that the necessary measures to do this are adopted at all levels of society. The outline in this article indicates a number of measures that make sense right now, even if there were no impending long-term changes. In addition, the scientific community must continue to probe the many factors involved, factors concerning the physical system that determines climate and also the factors governing the impacts of climate change on human activities.

2 William W. Kellogg and Robert Schware, Climate Change and Society: Consequences of Increasing Atmospheric Carbon Dioxide, Boulder, Colorado: Westview Press, 1981.

4 Kellogg and Schware, op. cit., p. 148; Rotty, op. cit., p. 125.

9 T.C. Chamberlain, "An Attempt to Frame a Working Hypothesis of the Cause of Glacial Epochs on an Atmospheric Basis," Journal of Geology, Vol. 7, 1899, p. 545; S. Arrhenius, "The Influence of Carbonic Acid in the Air Upon the Temperature of the Ground," Philosophical Magazine, Vol. 41, 1896, p. 237; also his Lehrbuch der Kosmischen Physik 2, Leipzig: Hirzel, 1903.

12 These various studies are reviewed in Kellogg and Schware, op. cit., Appendix B. See also Carbon Dioxide and Climate: A Scientific Assessment, Report of Ad Hoc Study Group on Carbon Dioxide and Climate, Woods Hole, Mass., Climate Research Board, National Academy of Sciences, Washington, D.C., 1979. A new Ad Hoc Study Group under the Climate Research Board met in July 1981 and essentially confirmed the main conclusions reached in 1979; its report will appear in 1982.

19 Gilliland, loc. cit.

21 This Figure is adapted from Malin Falkenmark and Gunnar Lindh, Water for a Starving World, translated from Swedish by Roger G. Ranner, Boulder, Colorado: Westview Press, 1976.

24 Details on its construction will be found in Kellogg and Schware, op. cit., Appendix C.

26 Charles R. Bentley, "The Polar Regions as a Concern for the Future," The Polar Regions and Climatic Change, Polar Research Board, National Academy of Sciences, Washington, D.C., 1982, Appendix C.

27 Kukla and Gavin, loc. cit.

34 Cooper, op. cit., p. 515.

41 Harvey Nichols, Palynological and Paleoclimatic Study of the Late Quaternary Displacement of the Boreal Forest-Tundra Ecotone in Keewatin and Mackenzie, N.W.T., Canada, INSTAAR Occasional Paper No. 15, Institute for Arctic and Alpine Research, University of Colorado, Boulder, 1975.

43 Environmental and Societal Consequences of a Possible CO2-Induced Climate Change: A Research Agenda, prepared by the American Association for the Advancement of Science for the Department of Energy Carbon Dioxide Effects Research and Assessment Program, Rept. No. DOE/EV-10019-01, Washington, D.C., 1980.

47 Warrick and Bowden, loc. cit.; M.L. Parry, Climate Change, Agriculture and Settlement, Polkeston: Dawson, 1978.

48 Butzer, loc. cit.

50 The Global 2000 Report to the President, Vol. 1, Council on Environmental Quality with the U.S. Department of State, U.S. Superintendent of Documents, Washington, D.C., 1980, p. 31.

52 Ibid., p. 56.

53 Ibid., p. 66.

54 See, for example, Amory B. Lovins, loc. cit.; and Energy in a Finite World, Vol. I: Paths to a Sustainable Future, Program Leader Wolf Häfele, International Institute for Applied Systems Analysis, Laxenburg, Austria and Cambridge, Mass.: Ballinger, 1980.

57 Outline Plan and Basis for the World Climate Programme, 1980-1983, WMO No. 540, World Meteorological Organization, Geneva.

58 Robert Schware, "Energy Development and Major Carbon Dioxide Producers: Modelling the Cumulative Responsibility for Future Climate Change," in Energy and Ecological Modelling, The International Society for Ecological Modelling, Copenhagen, Denmark (forthcoming).

60 Carroll L. Wilson (Project Director), Coal-Bridge to the Future: Report of the World Coal Study (WOCOL), Cambridge, Mass.: Ballinger, 1980.



You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • WILLIAM W. KELLOGG is a Senior Scientist in the Advanced Study Program of the National Center for Atmospheric Research, NCAR (sponsored by the National Science Foundation), in Boulder, Colorado. While writing Climate Change and Society with Robert Schware, he was also associated with the Aspen Institute for Humanistic Studies. In 1978-79 he served as Advisor to the Secretary General of the World Meteorological Organization in Geneva, in which capacity he was one of the chief architects of the World Climate Programme. He is a past President of the American Meteorological Society, and has served on numerous national and international committees-including currently the International Climate Commission of the International Council of Scientific Unions.
  • ROBERT SCHWARE is a Post-Doctoral Fellow with the Advanced Study Program of NCAR. In 1980-81 he was on the staff of the Project on Food, Climate, and the World's Future of the Aspen Institute for Humanistic Studies.
  • More By William W. Kellogg
  • More By Robert Schware