Sustainability Science

Tuesday, February 27, 2007


The current rapid rise in the atmospheric concentration of carbon dioxide presents us with two very different questions. The first is scientific: how will the rapid rise affect our climate? The available information has large uncertainties which lead to disagreements. Explaining phenomena such as the Ice Ages can help resolve the controversies, but will still leave us with a separate, moral question: how do we find a balance between our obligations towards future generations, and our responsibility towards those who are suffering today, the poor and infirm for example? This balance is bound to be different for developed and developing countries.

The future of our planet is too serious a matter to be left to experts in fields such as climate science and environmental economics. Everyone should be concerned about the possibility of global warming and global climate changes. Michael Crichton, a novelist, and Al Gore, a politician, therefore deserve applause for taking the time to familiarize themselves with the issues. Unfortunately Crichton, in ���State of Fear,��� and Gore in the documentary film ���An Inconvenient Truth,��� reach contradictory conclusions. Gore insists that the man-induced rise in the atmospheric concentration of carbon dioxide is already causing global climate changes, that disasters are imminent, and that immediate action is imperative. Crichton, in his assessment of the same scientific information comes to the opposite conclusion. He asserts that the rise in the atmospheric concentration of carbon dioxide poses no threat whatsoever. How can two highly educated, responsible people, both of whom assure us that they love and cherish the environment, disagree so completely about a matter as important the future of our planet?

Global warming presents us with a dilemma. The burning of fossil fuels brings enormous and immediate economic benefits, but has a potentially harmful byproduct, a rise in the atmospheric concentration of carbon dioxide that can cause serious global environmental problems in future. How do we find a balance between our obligations towards future generations, and our responsibilities to those who are suffering today, the poor and infirm for example? Although this balance will be different for developed and developing countries, everyone needs accurate scientific information concerning future climate changes. Why is there so much disagreement about the science?

Scientific advances invariably involve debates and arguments. Is the Earth flat or round? Is it a few thousand or a few billion years old? Do continents drift? For a long time these were controversial issues. Progress towards ultimate agreements depended critically on a continual interplay between observations and hypotheses. The debate about global warming is proceeding similarly. We are now close to agreement on what has happened over the past hundred years, but are at a stalemate concerning likely developments over the next century.

Towards the end of the 19th century the Swede Arhennius proposed that our industrial activities are causing an increase in the atmospheric concentration of carbon dioxide that will lead to global warming. This hypothesis was at first rejected on the grounds that the carbon dioxide injected into the atmosphere will not remain there but will be absorbed by the oceans. Accurate measurements of the atmospheric concentration of carbon dioxide, which became possible in the late 1950���s, established that the concentration is increasing rapidly, because of human activities. The debate then shifted to the consequences of this increase. Theories concerning the degree to which greenhouse gases in the atmosphere can trap heat predict a rise in globally averaged surface temperatures.

The paucity of the available instrumental records makes it difficult to determine whether global warming is underway. Thermometers were invented in the 17th century and initially were installed at only a few stations on land; coverage over the oceans remained very sparse deep into the 20th century. Satellites that provide measurements globally became available only in the late 1960���s. Over the past few decades the quantity and quality of the observations have improved markedly. Furthermore, to complement temperature measurements, we now monitor variations in the volume of glaciers, and in sea-level. These different data sets are all consistent with the hypothesis that global warming is underway: an increase in temperature has been accompanied by a decrease in global ice volume and a rise in sea level. This information convinces most scientists that global warming is underway. But is that warming a consequence of the rise in the atmospheric concentration of carbon dioxide?

To appreciate why this is a difficult question consider the hemispheric warming that occurs annually, the one commonly referred to as summer. From 21 December onwards the intensity and hours of sunlight increase steadily in the northern hemisphere and inevitably lead to spring and summer, albeit erratically. Unexpectedly warm spells towards the end of January can cause rhododendrons and azaleas to start flowering, until snowflakes abruptly return us to winter. Such transient warming is unrelated to the increase in the intensity of sunlight that starts in late December; it is part of an erratic and gradual transition to summer. The onset of global warming because of the rise in carbon dioxide levels can be similarly erratic. It is possible that the warming now underway is a transient phenomena, soon to be followed by global cooling. A thousand years ago Greenland was so warm that large numbers of Scandinavians moved there. After a few centuries the Little Ice Age forced them to abandon their settlements. Those temperature fluctuations were unrelated to changes in the atmospheric concentration of carbon dioxide. How can we be sure that the global warming observed over the past few decades is attributable to the rise in carbon dioxide levels?

To address this question scientists inspect the spatial and temporal structure of the global warming. The greenhouse gases in the Earth���s atmosphere inhibit the escape to space of heat from the Earth���s surface and thus amount to a blanket that keeps the Earth���s surface warm. In winter we similarly prevent the heat inside our homes from escaping to space by insulating the floors of our attics. As a consequence, temperatures increase inside our houses, but decrease in the attics. In the case of the atmosphere, scientists are observing that the rise in temperatures is confined to the lower part of the atmosphere, the troposphere where the greenhouse gases are accumulating. At greater elevations, in the attic (also known as the stratosphere) temperatures are falling. This finding supports the hypothesis that the increase in atmospheric levels of carbon dioxide is causing global warming at the Earth���s surface. Additional support comes from the difference in the warming as observed during the day and at night: there is no observed difference. This rules out the possibility that certain effects associated with changes in sunlight are causing the warming.

These are some of the considerations that persuade most scientists that a rise in the atmospheric concentration of carbon dioxide, because of human activities, has resulted in global warming. In the top panel of fig.1 the key evidence is shown in a manner that underlines the correlation between the changes in temperature and in the concentration of carbon dioxide.

Figure 1. Two versions of the same data. The red curve shows changes in the atmospheric concentration of carbon dioxide over the past century. The blue line shows changes in globally averaged surface temperatures. The scale for temperature variations is much larger in the upper than lower figure.

Temperature and carbon dioxide concentration are measured in unrelated units so that the same data can be presented very differently, as in the bottom panel of fig.1. Whereas the top panel shows a significant rate of warming over the past few years, a rate that is correlated with the rise in carbon dioxide levels, the bottom panel emphasizes that the warming over the past century has been very modest -- less than a degree Centigrade. Which picture prepares us best for the climate changes that are likely over the next few decades? Gore and Crichton give different answers. To resolve the matter we can turn to computer models of the climate. They give a range of answers, all indicating warming over the next century. Should we trust the models? What tests have they passed?

To settle this debate about future global warming we need information about the sensitivity of the climate to an increase in the atmospheric concentration of carbon dioxide. This sensitivity is an immensely complex matter because of the role of feedbacks, a term that refers to processes that spontaneously amplify modest disturbances (the way a slight push from one child to another can quickly escalate into a fight.) Particularly complex are the feedbacks involving water vapor. An increase in atmospheric temperatures (because of the rise in carbon dioxide for example) increases evaporation from the oceans, thus increasing the atmospheric concentration of water vapor. This is a powerful greenhouse gas that warms the atmosphere, thus promoting even more evaporation from the ocean, causing further warming, further evaporation and so on. Such a runaway greenhouse effect apparently occurred on the planet Venus which originally had oceans but now has none. Earth (which is further from the Sun and has lower temperatures) did not suffer this fate because here the accumulation of water vapor in the atmosphere leads to the formation of clouds which produce rain, thus returning the water to the oceans.

Seen from afar, ours is the blue planet with swirling white clouds, parasols that keep the Earth cool by reflecting sunlight. The clouds also act as a blanket that keeps the planet warm by means of the greenhouse effect of water. Is the heating greater or smaller than the cooling? The answer depends on the type of cloud. The fog and low stratus clouds off California, Peru and southwestern Africa are highly reflective, and mainly cool the planet. Tall cumulus towers that bring lightning, thunder, and heavy rains have a strong greenhouse effect that warms the planet. These are but two types of clouds. From a glance at the sky it is evident that there are many, many more types. At present their net global effect amounts to a warming of the planet. Scientists have developed climate models that reproduce this warming, but can we trust those computer models to predict how the clouds will change as the concentration of carbon dioxide increases? Can computer models really cope with whimsical, ephemeral clouds?

One test for the models is the global warming observed over the past century. They simulate that warming realistically, thus passing this test impressively. But is that a sufficiently stringent test for the models? Those who emphasize that the changes over the past century were very modest ��� those who favor the bottom panel in fig.1 ��� judge this test to be inadequate. Those who favor the top panel of fig.1 disagree. The resolution of these differences requires additional stringent tests for the models. The geological records of very different climates in Earth���s distant past provide not only appropriate tests, but a great deal more; they also provide an invaluable context for our current industrial and agricultural activities.

The geological records of the past 400,000 years in fig 2(b) tell a most remarkable story of recurrent Ice Ages that involved highly correlated fluctuations in temperature and in the atmospheric concentration of carbon dioxide. The rise in that concentration from the last glacial maximum (twenty thousand years ago) to the pre-industrial era (a century ago) is as large as the rise over the past hundred years. For now, the response (the rise in temperature) to the recent, rapid increase in carbon dioxide levels is small ��� see fig 2(a) -- because it takes the Earth���s climate a while to adjust. Hence it is a matter of time before we experience disastrous global climate changes.

This inference is questionable because correlations do not imply causality. The variations in carbon dioxide and in temperature in fig.2b are both attributable to another factor, fluctuations in sunlight. The introduction of the geological record is nonetheless a very important development in the debate about global warming because that record shows that the present is a precarious moment in the eventful history of planet Earth.

Brief, temperate interglacial periods such as the present one occur infrequently; they are separated by prolonged Ice Ages of approximately 100,000 years duration. At the time of the last interglacial period our species did not amount to much, but when the current interglacial started, some 10,000 years ago, we were ready and seized the opportunity to progress impressively from the invention of farming, to the development of astounding technologies that are transforming our daily lives. So remarkably rapid were our advances that we are now geologic agents, capable of interfering with the natural rhythms of our planet, increasing the atmospheric levels of carbon dioxide to levels higher than they have been for millions of years.

If we extend the geological record further back, to the time of the demise of the dinosaurs some 65 million years ago, we gain yet another valuable insight. From figure 2(c) we learn that the Ice Ages are relatively recent phenomena. They were absent until nearly three million years ago, then started appearing as modest oscillations, gradually amplifying into the dramatic climate fluctuations of the recent past. Why is this an era of such huge climate fluctuations? What caused the onset of Ice Ages 3 million years ago? The answer involves an interplay between two separate sets of processes.

Planet Earth, with the regularity of a metronome, spins about its tilted axis once a day, and orbits the sun once a year, giving us the cycles of night and day, and of summer and winter. We are all familiar with those rhythms, but few know that our planet has additional ones, with much slower beats. For example, the axis rocks back and forth with a period of 41,000 years so that the tilt gently oscillates between 22 and 24 degrees. The remarkable correspondence between these precise periodicities in sunlight variations (as determined by astronomical calculations) and the periodicities in the global ice volume fluctuations, convinces scientists that the changes in sunlight cause the Ice Ages. But why do those changes induce Ice Ages during some eras but not others?

Figure 2.

Ours is a continually changing planet. Earth���s richly varied landscape -- its spectacular mountains, vast, fertile prairies and pampas, lush tropical jungles, and barren deserts -- is but a snapshot of a constantly changing panorama. The agents of change are the drifting continents. The past 60 million years saw Australia separate from Antarctica, India marry Asia, the Atlantic expand, the Pacific contract, the Rockies rise, and the Appalachians fall. Because of these gradual changes, which were punctuated by sudden, violent earthquakes, and the sporadic eruption of volcanoes that altered the composition of the atmosphere, our planet experienced global cooling. At the time of the sudden demise of the dinosaurs, 65 million years ago, there were no glaciers anywhere; palm trees and crocodiles flourished in high latitudes. By 3 million years ago both poles had ice-caps and cold surface waters started to cover large parts of the tropical oceans, around the Galapagos Islands for example. That occasion was auspicious -- a change to a more temperate climate contributed to the evolution of African hominids, and subsequently of our species, homo sapiens. That occasion was also ominous -- the appearance of glaciers and of cold tropical waters introduced feedbacks that enhanced the sensitivity of Earth���s climate to such a degree that modest, periodic variations in sunlight are now capable of inducing recurrent Ice Ages. The drifting of continents created opportunities for our evolution, and also created the current era of high climate sensitivity.

The present is a precarious moment in the history of our planet because of an intricate interplay between perfectly regular variations in sunlight, and the jerky, irregular drifting of continents. We are currently experiencing interglacial conditions but, if the geological record is a guide to the future, are about to slide into another Ice Age, any millennium now. That inference, however, fails to take into account human activities. We are causing an exponential rise in the atmospheric concentration of carbon dioxide. We are doing this at a time when the climate of our planet is known to be very sensitive to small perturbations such as slight changes in the distribution of sunlight. Our activities are likely to inhibit the onset of the next Ice Age, but risk returning us to the conditions of 3 million years ago, risk eliminating the very conditions that favored our evolution. The next Ice Age is at least a few millennia away and hence not an urgent matter, but a world significantly warmer than the present one is another story that could unfold over the next few decades.

Both Crichton and Gore claim certainty for their respective (and contradictory) assessments concerning developments over the next several decades. In reality the uncertainties in estimates of future global warming, by means of climate models for example, are large. Efforts to reduce the uncertainties, and to reach agreement on the scientific issues, will benefit enormously from an explanation for the Ice Ages, and hence information about the degree to which carbon dioxide variations contribute to the transformation of slight fluctuations in sunlight into a huge climate signal. Scientists are making rapid progress on solving these puzzles, but as yet have no definite answers.

We now are obliged to make policy decisions in spite of the scientific uncertainties. Fortunately the geological record provides us with valuable guidance. The salient results from that record are that planet Earth has had an eventful past, and is currently in an era of high climate sensitivity to small disturbances. To persist with a perturbation that grows exponentially is foolhardy. This is a time for circumspection. We are in a ship, in a dense fog, in treacherous waters. It is therefore prudent to slow down, to reduce the rate at which we inject carbon dioxide into the atmosphere. Uncertainties in the global warming predictions underline the need to be cautious. Will the economy be ruined or stimulated by adopting measures to conserve energy, and by switching to alternate energy sources? Those who oppose such policies emphasize the uncertainties in global warming predictions, but downplay the even larger uncertainties in forecasts of how lower taxes, higher interest rates, etc. affect our economy. All of us are experienced in making decisions on the basis of uncertain information and realize that a wise strategy is to implement adaptive programs whose evolution is determined by their results, and by new information as it becomes available. Fortunately, the available technologies permit the immediate implementation of such programs. Now that we, at an unusual moment in its history, have become the stewards of a most remarkable planet ��� the only one known to be blessed with a great diversity of fauna and flora ��� we should remember that we call ourselves homo sapiens. This reminder is timely because balancing the possibly conflicting needs of the present and of future generations will require wisdom.

S. George PhilanderPrinceton University / University of Cape Town / CSIR South Africa

February 2007