CLIMATE AND CIVILIZATION
The Scientific Evidence for Climate Change,
and How Our Response to It May Influence National Polic

by
Donald Kennedy, Ph.D.
former President, Stanford University
Editor-in-Chief, Science Magazine

The Inaugural Lecture of the
Cornell-Gladstone-Hanlon-Kaufmann Annual Series
in Environmental Studiesand Communications

Presented on
October 4, 2000 at the
State University of New York
College at Oneonta

© SUNY College at Oneonta

DONALD KENNEDY was born in New York City on August 18, 1931. He is a graduate of Harvard College (1952), and he received his Ph.D. degree from Harvard University in 1956. He taught at Syracuse University for several years and then joined the Biology Faculty at Stanford University in 1960. In 1976, he received the Dinkelspiel Award, Stanford’s highest honor for outstanding service to undergraduate education. He was Chairman of the Department of Biological Sciences from 1965 to 1972 and Chairman of the new Program in Human Biology from 1974 to 1977. From 1976 to 1977 he served as Senior Consultant to the Office of Science and Technology Policy, then in 1977 he took a two-year leave from Stanford to become Commissioner of the Food and Drug Administration. In August of 1979, Dr. Kennedy returned to Stanford to become Vice President and Provost, and on August 1, 1980, he became Stanford’s Eighth President. He served as President until 1992, when he resigned and became Bing Professor of Environmental Science. In April 2000, Kennedy was named Editor-in-Chief of Science, the prestigious weekly journal of the American Association for the Advancement of Science. He is a member of the National Academy of Sciences, a Fellow of the AAAS, a member of the board of directors of the Washington-based Carnegie Endowment for International Peace and has served with various other organizations concerned with health and science policy. His recent books include Academic Duty (Harvard University Press, 1997) and The Last of Your Springs (Stanford Historical Society, 1998).

 

CLIMATE AND CIVILIZATION

It is a great pleasure to be here with Bill Kaufmann, who first introduced me to Catskill fishing and has been a friend for over 40 years, on both ends of the country. It is characteristic that he is giving something of value to higher education. But it isn’t new; Bill, otherwise known as Mr. Flood, has been doing that as an academic editor and publisher who has extended a benign and sensitive influence to all kinds of scholarly work. His outdoor interests and his own scholarly commitments fuse in this lectureship, and I am honored to be here.

This afternoon in a geography class, a student asked me what I would put at the very top of a list of pending environmental challenges. Here is what I told him — or at least what I meant to tell him! Nothing in the domain of environmental science and policy looms larger right now than climate change, and for the simplest of reasons: it’s because it’s happening.

Please don’t misunderstand me here -- I don’t mean changing day by day, or week by week, or even month by month. What changes on that time-scale is weather, not climate. There is a deep public confusion about that distinction -- to which media treatments of the latest flood or hurricane often add, by including some speculation to the effect that we may be seeing the effect of global warming. That’s a stretch that can’t be made on the basis of present knowledge: you just can’t infer much about climate from weather.

But we have learned a lot about climate in the past few years. In what follows I want to focus first on what we know for sure, then on what we project or infer from models, and finally present a rather unconventional view of our prospective future.

First: it is getting warmer. For a long time we have been measuring temperature directly, day after day, month after month, year after year. Some of the measurements, more recent of course, come from satellites that sense microwave radiation from oxygen molecules in the bottom four miles of the atmosphere. Others, going back a hundred years or so, are land-based temperature readings -- now taken from hundreds of sites -- or sea-surface measurements. The various measures are in quite good agreement, despite earlier claims to the contrary, and the data show that the average global temperature has increased, by about half a degree centigrade in the last century.

But we don’t need fancy instrumentation or computer models to tell us what’s happening: the changes are noticeable in many more homely ways. Flowering seasons come weeks earlier for British plants than they did in the nineteenth century when parson-naturalists began recording the dates on which their favorite species first bloomed. Mountain glaciers all over the world are retreating, and sea levels have risen. Birds familiar to American birders have spread their ranges northward: cardinals now winter in New Hampshire, a state they never reached in summer when I was a boy. In the intertidal zones of Monterey the transects laid down by friends of Ed Ricketts and John Steinbeck have been resampled recently; they are now occupied by species that had more southern distributions in the heyday of Cannery Row. And on the other coast bluefish and striped bass, unknown north of Cape Cod when I spent my first summer in Woods Hole, now delight anglers in the Gulf of Maine.

We are less sure about the human hand-print on global warming. Carbon dioxide and methane concentrations, as well as that of other gases that enhance the well-known greenhouse effect, have been rising dramatically since the beginning of the Industrial Revolution. Is that the cause of the increase in average global temperature? An emerging consensus says it is, and the Intergovernmental Panel on Climate Change — a huge consortium of experts on climatology, atmospheric physics, and other disciplines — is now calling it probable. But climate is variable on the scale of decades and centuries, and it is still not certain that this change is not part of ‘‘natural’’ variation.

As to the future: it is particularly encouraging that the General Circulation Models — large computer models used to predict future climates on the basis of known properties of oceans, atmosphere, and imposed changes in the latter -- when turned to the task of predicting past climate history, track the real data very closely -- including the two-year period in the early 1990’s following the eruption of the Phillippine volcano Mt. Pinatubo, aerosols from which hung around in the atmosphere for long enough to cool things off measurably.

That takes us naturally from the past into the future, where, for obvious reasons, our concern and that of the modelers is mainly focused -- in particular on global warming and the rate at which it may be expected to continue. That is a huge public policy issue, and it is one on which the lobbyists are already lined up. Economic estimates friendly to American industry are asserting that limiting the emissions of carbon dioxide, the product of fossil fuel combustion, will cost more than we can afford

But .industry is not of one mind on this matter. The insurance industry, which has experienced a huge increase in damage claims due to intense weather events -- a predicted corollary of global warming -- is deeply concerned. A number of major companies, led at first by British Petroleum and Shell, have abandoned their opposition and are now instituting emissions limitation programs of their own, voluntarily. The so-called Kyoto Protocol, executed in 1998 but still awaiting ratification by the U.S. Senate, calls for the industrial nations to reduce emissions by about 7% from 1990 levels by 2012. For the U. S. that would mean a reduction of about 30% compared to what we would be emitting if business as usual continued. It is a difficult political challenge for us, for the rest of the developed world, and for the developing countries as well —who have so far been unwilling to participate in emissions-reduction commitments. That’s why climate change is enveloped in global controversy.

Now we enter a domain in which we are less sure of anything, and more dependent on the predictions of models. Economic modelers are hard at work calculating what impacts more C02 and higher temperatures will have on the world’s agriculture and what the costs of limiting C02 emissions would be on the world’s industries. And experts on atmospheric physics, oceanic circulation and heat transfer, energy economics, computer modeling, geology, and biology are hard at work developing increasingly sophisticated projections.

At the first level these entail estimates of how rapidly the CO2 content of the atmosphere will rise. That is no small task. It requires an analysis of how the world’s population will grow --itself a complex set of predictions about how fast people in different parts of the world will attain their desired family size, how likely they are to change that desired size over time, and the degree to which "population momentum" will permit growth even after replacement fertility is attained. It requires estimates of the likely increase in per capita energy demand, along with guesses about how "carbon intensive" that demand will be -- that is, how much substitution of low carbon emission sources (natural gas, nuclear) for coal and oil will take place in the energy economies of various countries. Finally, it will demand much better estimates than we now have of the rate of deforestation in the world, since removal and burning of wood releases C02 from its storage sinks in the Earth’s biomass.

That is only the beginning. Once we have an estimate of the newly-attained C02 concentration in the atmosphere, the modelers must try to estimate by how much that concentration (say, a doubling from the pre-industrial concentration of 280 ppm to about 560 ppm) will cause the average global temperature to rise.

There is enough uncertainty in all this to satisfy anyone. At the first level, we may find that we can make economies both more efficient and less carbon intensive, slackening the growth in per capita emissions. Or we may succeed, as we appear to be doing in many places, in educating women in developing countries and expanding their economic opportunities, so that they have fewer children -- as a result of which, world population growth estimates might fall, as they did between 1996 and 1998.

At the second level, labilities and complex feedbacks abound. Aerosols -- dust and droplets, from volcanic eruptions or industrial sources -- can provide an overriding cooling influence, as did the Mt. Pinatubo eruption. As the Earth heats up evaporation from the surface of oceans and lakes will occur more rapidly, and as a result more clouds may form, reflecting some heat. But at the same time ice and snow deposits will melt, and as snow cover disappears and glaciers recede the Earth becomes darker -- technically, its albedo is decreased. As that proceeds, Earth absorbs more heat. Furthermore, say some ecologists, warming of the ubiquitous peat bogs in the far North will accelerate the metabolism of anaerobic soil bacteria and they will therefore release more methane -- a greenhouse gas twenty times more powerful per molecule than carbon dioxide.

So the debate runs on among the climate experts. But as the models get more sophisticated and the measurements better, consensus about the future is growing a little clearer. Atmospheric C02 will probably reach twice its present value very late in the next century, and that will raise the average global temperature between 1.5 and 4 degrees Centigrade. That is the consensus view of the Intergovernmental Panel on Climate Change. The associated rise in static sea level, a projection that contains additional uncertainties, could range from 15 to 90 centimeters.

Does that mean that we can count on a slow ramp of climate change, gradually forcing agriculture, industry and low-lying coastal residents to adapt? That is the prevailing wisdom, and the economic modelers are counting on it. The discussions that are preparing the nations for Kyoto all are based on projections of steady change, with initially small but increasingly significant economic consequences. My own view is rather different, though no less portentous for human society; and that is what I want to discuss in the rest of this argument.

Before turning to history, let me emphasize that much is already at stake -- and the more we learn about our already-changing climate the more important it is turning out to be. For example, the El Nino events -- the changes in global weather associated with temperature anomalies in the southwest Pacific -- have enormous effects on the welfare of even distant societies. In Zimbabwe in an El Nino year, corn yields experience nearly 10% losses. These events can be reliably predicted by observations of sea surface temperatures in the right region of the southwest Pacific, offering the opportunity to make advance preparations for shortages or even to "hedge" by substituting crops that are relatively unaffected (root crops, tobacco) for those that are severely impacted (maize). In Rwanda temperature and rainfall extremes associated with El Nino events are associated with marked increases in malaria incidence. So we already know that the impact of climate change, even in this stable era, can be heavy -- as we discovered in California during the winter of 1997-8, one of the biggest and wettest El Nino years ever.

And as we ponder the future of humankind’s influence on the planet, we find ourselves in another one of those situations in which history has some important lessons for us. Those who study earth systems have long wavered between a view of change that emphasizes linear, steady progression -- the uniformitarian view -- and one that stresses dynamism and sudden non-linear surprises -- a modem form, perhaps, of catastrophism. Indeed, this contest is but one version of a struggle that is as old as human thinking about the natural world: it is the war between statics and dynamics. Once we thought Earth was stable, and that the sun revolved around it; now we find it instead in constant rotation. Once we thought that the cell membrane was a fixed skeleton, and now we know that antigens and pores and other assemblies float around in it. Once we thought that the Earth’s skin was solid, and now we know that buildings fall down because its great plates cruise around and crash into one another. In one case after another we find dynamics triumphant over statics. And now, new interpretations of recent geological history have similarly damaged the relatively static, uniformitarian view of climate -- and that has caused me to revise how I think about global warming.

To explain how that came about, I need to review some of those changes in our notions of what climates were like in the recent past. The bottom line is that there is nothing new about climate change. Major cycles of glaciation and deglaciation, of the order of a hundred thousand years, have occurred with a basic periodicity that appears to be established by regular variations in the Earth’s orbit around the sun and by precession, that is, eccentricities in the Earth’s own rotation. Within the past 75,000 years -- a prolonged period of cold average temperatures that ended with the Last Glacial Maximum about 20,000 years before present -- there have been repeated, major fluctuations in climate, and these have invited an intensive interpretation of the record they left.

The result has been a revolution, largely accomplished in the last few years, in how we view that record. The technology is difficult, and it takes place under daunting conditions. It involves drilling deeply into glaciers and extracting cores of ice; when these are examined they are seen to contain layers, like a roll of coins except that the coins are of varying thickness. Variations in the properties of the layers allow comparisons between cores from different places, and permit dates to be associated with particular layers. The average temperature at a given level is determined using isotope ratios, and associated with the date assigned to that layer. Such data have been obtained from Greenland, from Andean glaciers, and from Antarctic ice. Isotope records from marine coral, faunal remains in North Atlantic sediments, and pollen analysis from lake sediments also support many of these observations.

There is excellent correspondence of reconstructed climate series for the past 150,000 years, indicating that the observed changes were global in scale. More surprising has been the discovery of very rapid changes, of ca. 7-8 degrees C, in global temperature; these have a time scale of a very few decades.

This pattern characterized the last period of major glaciation (sometimes called the Wisconsin by geologists) until about 12,000 YBP, when a rapid warming process began that introduced the Holocene period -- our present climate state. As the warming trend began, these abrupt climate "flickers" continued, but stabilized beginning around 10,000 YBP.

The exit from the last glacial maximum -- the rapid retreat of the ice sheets following the dying cold burst we know as the Younger Dryas-- was a dramatic event. Recall that at their maximum, the Northern Hemisphere was covered in a way that changed geography and climate almost beyond recognition. Present Chicago was covered to a depth of three kilometers, and the edge of the ice went as far south as the Middle Atlantic states. Mountain glaciers came far down the valleys. So much of the earth’s water was tied up in ice that nearly all the continental shelves were exposed; you could walk from London to Paris. The mountain of ice changed weather circulation patterns, splitting the jet stream in North America into two limbs, the southernmost of which brought Oregon rain to Arizona and made it a verdant country, dotted with lakes.

As the ice withdrew the climate changed and the sea rose. How it rose! Recall that by 12,000 years ago or so the first human hunters and gatherers were well established in North America, having come across the Bering land bridge made possible by the low sea level, but perhaps by other routes as well. They had already gotten as far south at Monteverde in Chile and to the Amazon basin, where they had painted caves and left evidence of food collected from the forest. They had established the famous Clovis settlements and were well ensconsed on the coast of British Columbia. There the emergence from the grip of the Ice Age was particularly dramatic. Between 9500 and 9000 years ago the sea level rose by about 5 cm. a year -- three meters during the lifetime of a person -- requiring the founding of new settlements and the drowning of others. It isn’t so surprising that the oral history of contemporary Haida still includes references to the changing sea.

But at the end of this deglaciation we entered a comparatively benign and salubrious period. It has seen relatively few shifts -- most recently, the Medieval Warm Period and the Little Ice Age - a 16th and 17th century period when Alpine glaciers were much more extensive than now. These changes appear to have been both much smaller and more local than those in pre-Holocene times; they appear as minor perturbations on a climatic record that is unique in human history for its constancy -- an unprecedented era of stable warmth.

Just then something remarkable was happening in the Fertile Crescent. Where it happened first isn’t certain, but one of the early places was Jericho, of Old Testament fame. It was in fact one of the first agricultural villages. We know this because the physical remains suggest schemes for water management, and because there is clear evidence that they were cultivating two kinds of grain -- barley and einkom wheat -- and that they had domesticated sheep. Botanists recognize the Jericho wheats as cultivars because they bear plumper kernels in multiple rows compared with local wild relatives; and domesticated sheep are significantly smaller than the local variety.

What events caused peoples who gathered wild grains and hunted sheep to bring them into their settlements and help Nature along? We do not know. Perhaps the last cold gasp of the Pleistocene -- the Younger Dryas -- had ushered in a high-risk climate that rewarded husbandry. We don’t know. But we do know that what followed this transition is the great story of community formation, and the beginning of civilization. Agriculture demands other-culture: the development and transmission of knowledge about water, and planting regimes, and plant protection. It requires stability of place, and the development of new rules about what belongs to whom -- especially land, which for the first time suggests both the prospect and the problem of private ownership. Once the plants came into the village, much else followed.

Much more recently, we have learned that the middle East was not the only place where this was happening. Recent investigations have established that the earliest rice cultivation in a South China village also took place around 10,000 years ago. New dates for the first cultivated squash seeds in the New World place that transition at about the same time. It can surely be no accident that in three different parts of the world -- the Fertile Crescent, South China, and South America -- modern dating techniques have demonstrated the first domestication of wild plants at the same time. The origin of agriculture and the subsequent evolution of settled civilizations thus are phenomena of a relatively warm climate.

Now we must return to the earlier, less stable, pre-agricultural period. The cause of the climatic oscillations -- the rapid "flickers" that characterized the late Pleistocene -- is uncertain. But the most intriguing, and perhaps the most likely, explanation for the abrupt changes centers on disruptions of the oceanic "conveyor belt" -- the circulation pattern responsible for poleward heat transfer from tropical latitudes. The movement of warm surface waters in the North Atlantic northward in the Gulf Stream keeps northern Europe 5 to 10 degrees warmer than it would otherwise be. This water, having given off its heat and become saltier through evaporation, becomes denser. It sinks to form a return current of deep, cold water -- that is the thermohaline conveyor belt.

Suppose now that the melting of ice sheets in the North injects large amounts of dilute, therefore low-density water onto this ocean surface. Or suppose that warming causes more evaporation and cloud formation at lower latitudes, increasing precipitation farther North. Now the diluted surface waters cannot sink, and the conveyor belt is blocked because the surface retains water that is colder and "belongs" at depth. By stopping the conveyer, gradual warming could thus trigger a sudden shift to a colder state.

Why these oscillations stopped, permitting the development of settled agriculture, is as unclear as the stimuli for the agricultural transformation itself. But we know from longer temperature records from Antarctic ice that previous interglacials were both less stable and shorter-lived than the Holocene; and it now appears that we are now at or near the critical ice volumes that characterized the end of previous interglacials.

Thus the historical data indicate unambiguously that the climate system is highly non-linear --even though it does not permit us to describe the mechanisms with certainty. What does the history tell us about the present human predicament? It suggests that we may be unwise to rest our economic projections on the assumption that the probable climate future is a slow, steady ramp of increasing average global temperature. That assumption by itself generates plenty of argument: technological optimists insist that we can adapt with minimal economic losses to a change in average global temperature of two or three degrees Celsius. Others argue that the damage functions will be much more serious, and that in any event the ramp may be steeper than the current estimate of the Intergovernmental Panel on Climate Change. In short, we don’t know how serious the consequences will be on the assumption that the temperature change will be gradual and steady, though it is plain that sea level change, agricultural modification, and other consequences will reshape human civilization to a significant degree.

But suppose the modern catastrophists are right again, and the uniformitarians wrong. If so, it is possible that we will reach the threshold of a region of instability -- a domain in which a new cold phase is as likely as more gradual warming. What then? It gives us cause for one more speculation about pre-Holocene history. If global conditions produced multiple agricultural transitions at the beginning of our current warm phase, why not before? Perhaps, after all, Jericho was not really first. Perhaps starts were made elsewhere, but vanished in the next cold phase leaving no record -- at least none that we have found yet. (A parallel lesson in perishability is found in our own settlement history: We remember the Massachusetts Bay Colony because it survived, and forget earlier Roanoke -- along with how many others? because they did not.)

If the climate system is as non-linear as it has been in the past, and if we trip a switch that takes us into a new temperature regime, what will it be like? It could, of course, become suddenly much warmer, because melting ice frees the metabolism of anaerobic bacteria in Arctic tundra and gives the atmosphere a sudden dose of methane. More likely, I think, is that we will rearrange the oceanic circulation, and plunge ourselves into a Pleistocene-like cold state.

Thoughts like that make me wonder about those others who, fifteen or twenty thousand years ago, may have tried to cope with something of the same kind and failed. What did they do? And what would we do if it happened to us?

I am not pessimistic about our coping capacity; I have too much faith in our resilience and our technology. What I do want to emphasize is the tenuous character of our assumptions about the character of global climate change. From the historical record, continuous slow warming at the present slow rate is not a pattern likely to continue for long. Thus I spend little time evaluating economic models based on that assumption, at least past the latter part of the next century. And I think our society ought to entertain, now, the prospect that a sudden shift in the climate regime is a highly possible consequence of the warming trend we are experiencing now. In a little-read addendum to their summary, the IPPC wrote: "Future unexpected, large and rapid climate system changes (as have occurred in the past) are, by their nature, difficult to predict. This implies that future climate changes may also involve "surprises." In particular these arise from the non-linear nature of the climate system. When rapidly forced, non-linear systems are especially subject to unexpected behavior."

Just so. Now to the domain of politics and policy. How should we prepare for the climate future, if at all? Stabilization of emissions is a strategy that can help make us ready for either the gradual or the catastrophic outcome, and we need to move toward that goal. That and other steps will depend on broad public recognition of the one certainty in this business: we are engaged in a giant experiment, in which Earth’s climate system is being deliberately manipulated according to no planned protocol, based on no working hypothesis. Not surprisingly, we have no idea how it’s going to turn out.

To tell people to prepare for slow, gradual warming is, it turns out, something less than a clarion call. "We’ll adapt," say those who want to preserve the status quo. "Don’t underestimate the capacity of our technology to provide new solutions, ones that aren’t visible now." Those arguments are hard to answer; the unifonnitarian vision just isn’t very dramatic. And right now, in the last weeks before a Presidential election, Kyoto and climate change is off the radar screen: not a word was said about it in the 90 minutes of the first Presidential debate.

But if the catastrophist vision is right -- that is, even if there is only a 20% chance or so that we will cross the threshold for a dynamic climate response -- then it seems to me that the case for a strong preventive strategy is much stronger. Alas, that is not part of the present political dialogue about global warming, and I think it should be.