The Great Global Experiment

As climate change accelerates, how will we adapt to a changed earth?

During a recent Alaska study cruise cosponsored by the Harvard Museum of Natural History, James J. McCarthy stopped at several islands with small native communities—Little Diomede, for example, with 150 inhabitants. At each village, McCarthy asked the elders if climate had changed in their lifetimes. In one village after another, he relates, "They said, 'Well, my grandfather said the ice used to come in November, and now it doesn't come until January.'" Wherever he went, the story was the same: "My grandfather said it used to leave in June. Now it goes out in March."

"Those are just anecdotes," says McCarthy, the Agassiz professor of biological oceanography. But even as he distinguishes anecdote from scientific evidence, McCarthy shares with virtually all his colleagues who study climate change the firm conviction that our world is warming rapidly. Understanding the rate of change, its causes, and the consequences for humans and nature engages researchers around the planet—including prominent scientists in Harvard laboratories. With the scientific consensus coming into clearer focus, policy analysts in the University, as elsewhere, are struggling to devise appropriate responses—a task revealing sharp differences of opinion over fairness and efficiency, and even wider gaps between the worldviews of biologists and economists.

Biological oceanographer James J. McCarthy has been a leader in the international effort to assess the risks to human and natural systems posed by global climate change.

Photograph by Jim Harrison

McCarthy's experience demonstrates the sweep of the global- warming challenge. The former director of the Museum of Comparative Zoology has studied the processes that control biological production in the upper waters of the North Atlantic, the equatorial Pacific, and the Arabian Sea. Inevitably, those inquiries have led him, in the past two decades, to consider the overall condition of the marine environment—where half the planet's biological production occurs—and more recently, to investigate the diminished extent of sea ice; altered ocean-atmosphere exchange of energy, air, and moisture; and the likelihood of changing ocean-circulation patterns.

In 1997, McCarthy was tapped to cochair the Working Group on Impacts, Adaptation, and Vulnerability for the Intergovernmental Panel on Climate Change. The IPCC, created in 1988 by the World Meteorological Organization and the United Nations Environmental Programme, is the mechanism for winnowing the myriad of published research to achieve consensus on what aspects of climate change scientists are most confident about. It's a conservative process, involving comments from thousands of scientists worldwide. McCarthy describes meetings at which four-fifths of the papers were rejected for insufficient data. "The possibility of having anything radical get through this process is virtually nil," he says. His panel's effort represents the work of more than 400 authors: their report of 1,000 or so pages documents the mainstream scientific consensus on climate change. The result, McCarthy maintains, "is the one that stands up to all the tests, the one that you cannot refute with published scientific findings."

The full IPCC report, released last year (see, confirmed that the average global temperature is rising, and concluded that human activity propels climate change.

What's more, McCarthy says, a warming world is a changing world. Average temperature rose a little more than 1 degree Fahrenheit during the past century—but on Alaska's north slope and in northwestern Canada, during the same period, temperatures have already risen 4 to 7 degrees. That impact dramatically illustrates a key research finding: the effects of climate change vary widely by region, and may be far more powerful than average figures suggest. It is no particular comfort that the data confirm the Little Diomede villagers' anecodotal testimony: subsistence lifestyles throughout the far north are threatened by warmer winters. All over the planet, says McCarthy, scientific work reviewed by the IPCC shows warming-induced changes "in geographic distribution of many species, and in the timing of flowering, egg laying, migrations, et cetera."


The Arctic Meltdown

How did we reach this point? The largest contributor to recent global warming, scientists believe, is carbon dioxide (CO2), a "greenhouse gas" that allows visible light from the sun to reach the earth's surface, but then retains some of that emitted light as heat in the lower atmosphere, and hence warms it. This gas occurs naturally in the environment: we exhale it, and plants absorb it during photosynthesis, releasing oxygen. The problem is that we have been adding more CO2 to the atmosphere than can be absorbed by the biosphere and the ocean, slowly but surely, since the beginning of the Industrial Revolution. When fossil fuels such as oil, coal, or natural gas are burned, they release their carbon content to the atmosphere as CO2. Atmospheric concentrations of the gas are now well beyond the steady level of 280 parts per million (ppm) that persisted from a.d. 1000 until the early 1800s. Ice cores from around the globe tell most of the story, and direct measurements were added in 1958. The concentration of CO2 began increasing about 150 years ago and is now at 370 ppm, one-third higher than the historic level—and rising. Because these additions of CO2 will persist in the atmosphere for a century or more, mitigating action taken now won't reduce concentrations for generations. These actions, however, will slow the rate of greenhouse gas warming.

Surface temperature in the last thousand years (according to evidence from tree rings, corals, ice cores, and isotopes), has been variable, but if anything was decreasing slightly, on average, until about 1900. That is what might have been expected based on historic cycles of glaciation: we are at a point in the cycle where we might anticipate beginning a long period of gradual cooling. Instead, temperatures have been rising since the turn of the twentieth century. The rate of warming in the last century was probably the fastest of any hundred-year period in the last millennium, and the trend appears to be accelerating. Since 1976, the World Meteorological Organization reports, global average temperature has risen approximately three times faster than the century-long trend. Nine of the 10 warmest years in the last 140 years have occurred since 1990—and 2002 is on track to be the warmest ever.

Looking beyond instruments, scientists have found other evidence of warming. In close to 100 areas in the Northern Hemisphere, data exist covering at least a century, often based on newspaper reports of contests and wagers to guess the ice-out date of lakes and rivers. In 95 percent of these cases, the ice-free season has lengthened an average of about two and a half weeks.

Another piece of the puzzle comes from the top of the world. Nuclear submarines have been transiting the North Pole beneath the Arctic ice since the 1950s, and measuring its thickness. When the data were declassified at the end of the Cold War, they showed that thickness had decreased by 40 percent between the late 1950s and the 1990s. Satellite data show a 10 percent reduction in the extent of the icepack over the last two decades. The United States Navy, pondering the implications for national security, worries about "scientific models [that] consistently suggest seasonal sea lanes through a formerly ice-locked Arctic may appear as soon as 2015. Summertime disappearance of the ice cap could be possible by 2050 if the trend continues."

Extending the speculation, what will happen to all the organisms adapted to life in a frigid Arctic? Algae that live on the underside of polar sea ice, McCarthy explains, constitute the base of an arctic food web that ultimately supports the "signature creatures" commonly associated with the far north: fish, seals, and polar bears. Loss of ice threatens the chain from the bottom to the top, where entirely carnivorous polar bears stalk seals' breathing holes. Without ice, seals don't need breathing holes and polar bears will go hungry. "One might imagine that while this is bad for polar bears, it is good for seals," says McCarthy. But "this year, in the Gulf of St. Lawrence, many harp seal pups drowned when there was no stable ice for them to rest on. That is a massive recruitment failure."

In the summer of 2000, McCarthy got a glimpse of what the Arctic's future may look like. He and other scientists aboard the 75,000-horsepower Russian icebreaker Yamal arrived at the North Pole only to find open water in every direction for miles. All along their 500-mile journey they had encountered unusually thin ice, with large areas of open water visible at every point of the compass. That same season, a Canadian ship transited the legendary, once-impassable, Northwest Passage without touching ice. The Arctic meltdown is well under way.

The Great Carbon Sink

Although the tangible evidence of global warming is clearest at high northern latitudes, other important factors are at work much closer to home, in the forested middle latitudes where many of the industrialized world's people live. We know this in part because of work pioneered by Steven C. Wofsy, Rotch professor of atmospheric and environmental science, who has been studying the role of the terrestrial biosphere—mostly forests—in absorbing atmospheric carbon as he tries to ferret out the magnitude of the CO2 problem. "The world use of fossil fuels amounts to about six and half billion tons of carbon per year," he says. "There's a magic number that I teach my Science A-30 class: it takes 2.1 billion tons of carbon to raise the atmospheric CO2 concentration by one part per million." If all the CO2 we pumped into the atmosphere stayed there, the concentration would be rising by more than three ppm per year—but it is actually rising only one and a half ppm annually. Scientists began to search for the missing carbon, and guessed that it was going into the oceans or being taken up by forests or soils.

Of the three-billion-plus tons missing from the atmosphere each year, it turns out the oceans absorb half (for an explanation of how the "biological pump" moves carbon into the deep ocean, see "The Ocean Carbon Cycle," page 40). The rest almost certainly ends up sequestered in forests.

The process of plants taking CO2 out of the atmosphere and storing it as organic matter has accelerated in the last 20 years—a surprise, given how much one hears about deforestation. (In his lectures, Wofsy calls it a miracle.) Could it be that the terrestrial biosphere is responding to the rising CO2? If we knew why, might we control the process and use it as a management tool? He set out to understand exactly what is happening in the woods.

"There are three explanations worth mentioning," says Wofsy. "One is that when you add CO2 to the atmosphere, plants grow better." In fact, experiments conducted at Harvard have shown that plants exposed to double concentrations of CO2 usually grow faster: the so-called CO2 fertilization effect. Of course, the actual rise in atmospheric CO2 since the pre-industrial era has been only about 30 percent. "If that is enough to stimulate extra uptake of CO2," says Wofsy, "it really is a miracle."

A second possibility is that forests are growing better because they are being fertilized by pollution. "Trace metals, the oxides of nitrogen that form smog, and the sulfate aerosols that cool the earth (and kill people) are actually all fertilizers," he explains.

"Finally," he says, "it may be that what is happening is an historical artifact. In the eighteenth century a lot of forests were cut down as we converted the land to agricultural use; in the nineteenth century many forests in Northern Europe were logged." Because of modern agricultural practices that boost crop yield per acre, marginal lands have been abandoned to forest. This is true in most of the eastern United States. "In the most extreme case," says Wofsy, "if you went to New Hampshire in 1680, 95 to 100 percent of the land was in forest. In 1880, only 15 percent of the land was forested. If you go there today, it's 85 percent. In South Carolina, huge areas that were used to grow cotton now support forests that yield wood fiber for paper companies. Much the same has happened in northern Europe." And organic matter has accumulated in forests for the last hundred years as fires have been suppressed in the western United States.

Which of the three likely explanations is actually responsible for the increased uptake of carbon? A Princeton analysis found no significant increase in forest growth rates during the last 40 years, suggesting that the effect of any kind of fertilization on these forests has so far been negligible. The study attributed half the carbon uptake to fire suppression and called the other half an historical artifact.

But Wofsy, suspecting that the story was more complex, devised a direct experiment. Thinking about just how much carbon a forest absorbs, he began to wonder what else happened in the woods besides the growing of live trees. After all, he reasoned, when the New England forests were fields, there was no dead wood lying around, and even the soils had been depleted of organic matter by cultivation. He was also interested in how global warming might be affecting forest growth. "Growing seasons have gotten longer in the middle latitudes of the Northern Hemisphere," he says, "and I wanted to understand whether that was a factor."

Using techniques he had deployed for short-term measurements in Brazil and Canada, Wofsy and his research group of 15 to 20 young scientists set up a long-term experiment at the Harvard Forest in Petersham, Massachusetts, in 1989. They found that the Harvard Forest was taking up a lot of carbon. "It is 60 years old," Wofsy says of the prevalent vegetation, "and a lot of models said that it should have stopped absorbing carbon by then, but even though it is a full-height forest, the trees are still growing." He discovered first that less than half the carbon absorbed by the forest is going into the living trees. The rest is going to deadwood in the soils, and accumulating there. (The forest is also undergoing "succession": as oak replaces pine, the denser wood of the successor species holds more carbon.)

He also learned that the forest responds strongly to climate variations. In a long growing season, it will take up more carbon. In a dry growing season, carbon sequestration increases as well. "It has nothing to do with the trees' uptake of carbon," Wofsy explains, "but rather with the decomposition process of deadwood, which is slowed down when the forest dries out."

Having established that northern mid-latitude forests are sequestering vast quantities of carbon, Wofsy decided to run the same longer-term experiment in a boreal forest in central Canada and in a tropical forest in Brazil. He learned that those ecosystems are quite different.

In boreal forests, growing seasons are short, there is very little rainfall, nutrients are few, and the trees don't get very big. But underlying the forest there is a lot of peat, the remains of moss that has been building up for 5,000 to 7,000 years. The moss grows slowly, but it accumulates because the soils tend to be saturated with water or to be frozen as part of a discontinuous permafrost. "We call this a cold desert," says Wofsy, "because it gets only about 11 inches of rainfall a year, but the climate is cold enough that evaporation is even less." The combination of cold and wet preserves the peat. Recently, however, the climate has warmed in this area, and there is a good chance that the peat is no longer stable over the long term.

Why should that matter? "If you took all the peat in Canada and Russia and turned it into CO2" by burning, Wofsy says, "you would double the amount of CO2 in the atmosphere. It took 5,000 years to make it, but it doesn't take much to get rid of it," because it can catch fire. Intense conflagrations burn all of the dried peat in a forest. "Usually, just a foot or so of peat is dry enough to burn, but if all two meters dry out as climate warming trends continue," he cautions, "the full accumulation could be released to the atmosphere over perhaps 50 years."

Given that threat, Wofsy is now studying boreal forest hydrology, or water balance, in order to gauge the peat's long-term stability. He already has temperature measurements dating back to 1994 of the permafrost in the immediate area where he has been working. They show that temperatures deep in the soil are rising. "It is only one spot," he cautions, "and it is not something that we would be able to extrapolate to the whole region, but we don't think it is atypical."

Far to the south, in a tropical setting, he is duplicating the experiment as part of a cooperative program among NASA, the United States, and Brazil. One would expect that a mature tropical forest wouldn't be taking up any carbon on average, he says, because any organic matter decays quickly in the heat and humidity despite the tremendous growth. And in fact, his data show that the forest neither loses nor gains carbon. The miracle, then, is that the mid-latitude forests over much of North America, Europe, and some parts of Asia, all generally 50 to 100 years old, are acting as a giant carbon sink, and prior land use changes are a major factor in that.

Can we solve the carbon problem by growing more forests? "No," says Wofsy, "that is not an entire solution. But I strongly advocate the idea that managing forests for carbon should be part of a much broader strategy of managing forests for multiple gain. Forests provide a variety of economic goods, including fiber, watershed protection, and wildlife habitat. Lengthening the cycle of rotation or changing the tree species to higher-quality hardwoods would add to the amount of carbon stored there and simultaneously increase the value of the product." Sequestration of carbon by forests is "an important matter to keep in mind," says Wofsy. "But if you want a magic bullet, I don't have one—I don't think anybody does."


Spreading Seas

Not everyone is compelled by Wofsy's and McCarthy's data. But owners of oceanfront property confronted by rising sea levels are increasingly aware of global warming. Contrary to popular belief, most global sea-level rise to date is caused by thermal expansion, not melting of ice. As ocean waters warm, they expand at a predictable rate in response to temperature. During the twentieth century, driven by warming waters, the global sea level rose 4 to 8 inches. This century, sea level is expected to rise between 4 and 35 inches, according to the IPCC, with mid-range values (a little more than 18 inches) more likely than either extreme. Sea levels will continue to rise for centuries, even if new emissions of CO2 were limited tomorrow, because to date, only a fraction of the ocean—the warmest water that lies on the surface—has been warmed by higher temperatures. It will take hundreds or thousands of years for all the water in the ocean to be exposed to our warmer planet, so coastal inundation, erosion, storm damage, contamination of freshwater supplies, and rising water tables are problems that will be around for a long time.

Melting of land-based freshwater glaciers and ice sheets also contributes to sea-level rise. (Melting sea ice doesn't—because it floats, sea ice already displaces ocean water. The 12,000-year-old, half-a-trillion ton, Rhode Island-sized chunk of the Larsen B ice shelf that collapsed so spectacularly this spring off the Antarctic peninsula was sea ice.) Worldwide, 90 percent of alpine glaciers are retreating. Glacier National Park, for example, is not likely to have any glaciers by 2070. But by far the greatest reserves of fresh water on the planet are frozen in Greenland and Antarctica, where the ice forms sheets up to two miles high. At those elevations, temperatures remain consistently below freezing. Today, Antarctica has the least precipitation of any continent—but that might change as the world warms. Warmer air has the potential to hold more water, and if moisture-laden winds found their way into the polar vortex, that might actually increase snowfall at the South Pole—thereby acting as a small brake on sea-level rise.

But there are indications that the Antarctic ice sheets partially melted as recently as 14,000 years ago, and that sea levels rose 70 feet in a few hundred years. No climate model predicts melting of any of these massive ice sheets in the next 100 years, yet no model today can explain the melting of the past. Clearly, the two biggest variables governing sea-level rise—what happens to the ice in Antarctica and in Greenland—are subjects for further research.

Doomsday scenarios aside, even the incremental rise in sea level already in evidence and forecast to continue at an accelerated rate is potentially catastrophic. The horizontal extent of beach erosion is typically 50 to 200 times the rise in sea level. Mid-range IPCC estimates of sea-level rise during the next century therefore imply a corresponding loss of 75 to 300 feet of shoreline, threatening coastal settlements everywhere. In the span of one lifetime, many U.S. beaches would disappear. Low-lying areas like the Mississippi River delta (think of New Orleans) and Chesapeake Bay would suffer further inundation. Coastal habitat, including wetlands, would vanish and some species would become extinct. Millions of people in developing countries—Bangladesh, for example—would be at risk from rising waters. By 2090, lower Manhattan would be under several feet of water during storm surges every few years unless something were done (as it likely would be: the real estate is simply too valuable, the inhabitants too affluent, to do nothing).


To the Extremes

Not that it is safe to assume all such changes will be as "gentle" or as gradual. Another "robust conclusion" of the recent IPCC report is that extreme weather events have become "more frequent, more intense, and more persistent" in the last 50 years: higher maximum temperatures and more hot days over nearly all land areas; higher minimum temperatures over all land areas; an increased heat index (a combination of temperature and humidity); more intense rain over land areas; and increased summer drying and risk of drought in some areas. The projections show a broadening and intensification of these trends, plus increased tropical-storm peak winds and intensified peak precipitation; more droughts and floods associated with El Niño events; and increased Asian summer monsoon variability. "One can argue that a little bit of warmer weather may be bad for the ski industry and good for the citrus industry," James McCarthy acknowledges, "but hardly anyone can find good news in any of this because extreme events are inherently destructive."

The increased frequency and intensity of floods is the strongest sign of this tendency toward more extreme weather. "One has to be careful," McCarthy says, "because most of the floods that we hear about, including those in the news this summer in Central and Eastern Europe, are in systems that have been heavily modified by human use—floodplains, for example, have been developed." But one need only look at the last five years to see the increased frequency in century-scale storms, like the one in Europe. "Tropical storm Allison in Texas a year ago was the costliest precipitation disaster in U.S. history," says McCarthy. "And 1999's Hurricane Mitch—which wasn't a hurricane by the time it reached Honduras because the wind energy was gone—was a warm air mass that lifted up slowly and dumped its moisture, resulting in more than 10,000 lives lost. It was the biggest precipitation disaster in Central American history. One year later in Venezuela, in December, more than 25,000 lives were lost in the greatest precipitation disaster in South American history." In the last five years, he continues, "in China, in North Korea, in India—even places that are preadapted to flooding—people have experienced flooding beyond any historical proportions."

Might warmer temperatures increase the global food supply, offsetting coastal property losses? To a point, yes, McCarthy explains, but other problems of distribution and inequity arise in the IPCC's vision of the future. Net gains in agricultural productivity are most likely to take place in North America and Northern Eurasia with a moderate rise in temperature. But neither of those places is food-limited today. So the bounty will accrue where the people aren't, while the populous tropics and subtropics will see net declines in agricultural production. The notion of environmental refugees streaming north out of Africa is one that European nations are taking very seriously.

Further, any projected increase in crop yields assumes temperature increases of no more than 4 to 6 degrees Fahrenheit. Beyond that, productivity drops off sharply everywhere. Not surprisingly, most of the crops humans have selected over time are grown in regions near the optimum temperature for maximum food production. A detailed analysis in the U.S. Climate Action Report­2002 (, the official federal summary of observed and anticipated domestic climate-change impacts, notes, for example, that with rising temperatures, barley should benefit, but wheat will not. The report (McCarthy calls it "an excellent document...the science is accurately represented") also illustrates graphically what will happen to agriculture in one state by "moving" Illinois south to the latitude of Oklahoma and North Carolina. Beyond agriculture, the government report acknowledges that "some of the goods and services lost through the disappearance or fragmentation of natural ecosystems are likely to be costly or impossible to replace"; that alpine meadows in the Rockies, and some barrier islands, will likely disappear; and that southeastern forests are likely to experience major species shifts or break up into a mosaic of grasslands, woodlands, and forests.

But what if global warming somehow reinforces itself, accelerating climate change—or in fact triggers a sudden, sharp shock to earth's systems? McCarthy says the models that purport to predict agricultural yields "presume a very linear change, that we will gradually get warmer and warmer and warmer. [But] we know from the record in the past that climate doesn't gradually shift from one stage to another. It does so with large swings, and one manifestation of these swings is in extreme events."

An example of such reinforcement occurs in the Arctic. When sunlight hits ice or snow, 90 to 95 percent is reflected. But when the snow melts or the ice retreats, almost the inverse is true: only 5 or 10 percent of the sun's energy is reflected. As the sun warms bare ground and open ocean, this warms adjacent areas of snow and ice, causing further melting and increased absorption of the sun's rays. "This drives a greater pace of change in the Artic," says McCarthy. Another imponderable feedback—he calls it "the juggernaut in all this"—is what will happen to cloud cover. A warmer atmosphere is able to hold more moisture. But will there be more clouds? If so, will they be dark clouds that absorb heat, or reflective clouds that reduce sunlight that reaches earth, thereby mitigating the impact of global warming? Nobody knows. And what kind of impact will an ice-free Arctic Ocean have on world ocean-circulation patterns and the variety of their oscillations and harmonics, most of which are poorly understood? "It is an experiment that is really wide open," says McCarthy, "and many of us wish we were doing it in a laboratory rather than in the real world."


"Climate Switches" and a Permanent El Niño

Butler professor of environmental studies Michael B. McElroy focuses on just such questions of larger-scale effects from climate change, a principal interest of Harvard's Center for the Environment, which he directs. The center, an interfaculty initiative, includes experts in oceanic ecosystems, development, governance, public health, and atmospheric chemistry.

"The IPCC approach," he says, "has been to focus in large measure on the ability of models to reproduce the global average temperature changes that were observed in the last 150 years." When early models overstated temperatures, some scientists were concerned. But McElroy wouldn't throw the models out—he just wouldn't ask so much of them: "I don't believe that any of these models should be expected to reproduce in detail what happened over the last 150 years because there is natural variability in the climate system."

He focuses less on the degree of warming, which could easily be higher or lower than IPCC estimates, he says, and more on the risk of sudden changes to the climate system. There are indications that some of the major circulation patterns that drive that system are starting to change, with potentially serious consequences.

Michael B. McElroy directs Harvard's Center for the Environment. An atmospheric scientist, he is concerned about the possibility of sudden, major disruptions to the global climate system.

Photograph by Jim Harrison

One hint comes from a general circulation model—a giant climate simulator running on a supercomputer—produced by the Hadley Centre in Britain. The Hadley model, one of several worldwide, "though not necessarily a uniquely credible projection," stresses McElroy, "is as good an indicator of the changes that could happen in the future as any that we have." What it shows (see graphic on page 42) runs counter to the conventional wisdom on climate change—that the highest temperature increases will occur at the highest latitudes. McElroy was sufficiently impressed by the model that he put it on the cover of his latest book, The Atmospheric Environment: Effects of Human Activity. "This shows what the world might look like 50 years from now, with temperature changes of 7 to 11 degrees, and in some places 14 to 18 degrees, which are bigger than the interglacial changes that occur in nature," he says. "But what is really notable about this projection is that it shows significant warming, and simultaneous drying, in areas of the equatorial tropics, such as the Brazilian rainforest."


"The way the tropics work at the moment," he explains, "is that you have the strongest rainfall over Indonesia." Warm, moisture-laden air releases rain over the region as it rises into cooler heights of the atmosphere. This fountain of air, once drained of its moisture, descends over the Pacific, where the cycle begins again. Another patch of air rises over Brazil, and a third in central Africa, "so that in the tropics today, there are three fountains blowing air up, and everywhere else air is descending and it is not raining. The model suggests that by 2040 or 2050, there might be a coalescence of these three centers of precipitation in one enormous fountain of rising air somewhere near Indonesia, and drought everywhere else. The implications of this are devastating rainfall and floods in Indonesia, massive drought in Brazil, and destruction of the rainforests there and in Africa. What we'd be doing," he says, "is causing a climate disaster in the poorest countries in the world."

An unconventional prediction suggests significant warming and drying in the equatorial tropics as early as the 2050s. Brazil's rainforest could dry out and disappear. Source: Hadley Centre for Climate Prediction and Research.

McElroy is describing what are sometimes called "climate switches," like the El Niño phenomenon. "There are currently two modes of wind and ocean interaction in the tropical Pacific," he explains. In one, trade winds blowing from the southeast and the northeast drive warm surface water across the Pacific and pile it up in Indonesia, so that the ocean's surface is actually several meters higher there than it is along the coast of South America. This exposes the cold, nutrient-rich waters off the coast of Peru, and the fisheries thrive. Once the elevation of the water gets high enough, though, it becomes unstable and collapses. "When that happens, you have a reverse flow of all this warm water back across the Pacific where it piles up against the coast of the Americas." The cycle repeats itself periodically like a pendulum. "When El Niño begins," McElroy says, "the warm water returning to the coast of South America caps off the colder water, nutrients don't get to the surface, the fish die, and it rains cats and dogs in Peru." Brazil and Indonesia suffer a drought.

What McElroy thinks is happening—and what the Hadley model also suggests—is that as the earth warms, "we are slowly increasing the amount of warm water that is stored in the surface levels of the Pacific Ocean, and could reach a point where the cold water is never exposed. Once you get to the point where you have enough warm water, then that's it," says McElroy: "a permanent El Niño. This is a serious issue that could give you a globally significant climate change in a matter of years."

Some evidence suggests this change is underway. One of McElroy's colleagues, professor of earth and planetary sciences Daniel P. Schrag, a geochemist, has developed ways of assessing the history of temperature in the Pacific Ocean by using the isotopic composition of coral as a proxy. "His analysis suggests there may be a real change in the last 25 years," says McElroy, who calls the work "incredibly important. If you can show that the rhythm of climate really is statistically different now, that is a big deal." Schrag's data show that the cold and warm periods regulated by El Niño were more evenly distributed in the past, whereas the last 25 years have brought more frequent warm spells.

An equally important regulator of climate is the North Atlantic circulation, sometimes referred to as "the global conveyor belt." In both the Northern and Southern hemispheres, westerly winds that carry Pacific Ocean moisture east are largely stripped of their water content before they reach the Atlantic by the line of mountains—the Rockies and the Andes—that runs from north to south along the continents. But near the equator, where trade winds blow in the opposite direction, from east to west, there is no barrier near the Panamanian isthmus, so winds there transfer moisture from the Atlantic to the Pacific. Likewise, westerly winds blowing from the Atlantic across the Eurasian continent encounter no particular barrier before reaching the Pacific. This net loss of fresh water makes the surface waters of the North Atlantic among the most saline in the world. Saltier water has a lower freezing point, so as it cools and reaches high density inside the Arctic Circle, some of it sinks, descending to the ocean floor and flowing south through the Atlantic at great depths to supply the oceans of the world. This sinking column of water draws the warm Gulf Stream north along the coast of Western Europe where it greatly tempers the climate all the way up into Northern Scandinavia, well inside the Arctic circle.

McElroy worries about what would happen if higher temperatures melt the land-based ice in Greenland. "You might add enough fresh water to the North Atlantic to remove the salinity contrast, in which case deep water will stop forming and the Gulf Stream might go straight across the Atlantic. It is much more complicated," he says, "than just talking about global temperature changes." The IPCC assigns a low probability to disruption of the North Atlantic circulation in this century, but an increasing probability after 2100.


The Human Variable

How credible are these scenarios? One criticism of climate-change science is that its predictions seem so uncertain. The IPCC has created 35 different scenarios for changes in CO2 emissions and the ensuing atmospheric concentrations over the next century, with a consequent rise in temperature ranging from 2.5 to 10.5 degrees Fahrenheit. One scenario assumes very rapid economic growth, a global population that peaks in mid-century and then declines, rapid introduction of new and more efficient technologies, and a convergence of developing- and developed-world standards of living. On this particular set of assumptions, there are three variant futures: one that is fossil-fuel intensive, a second that assumes a future emphasis on non-fossil-fuel energy sources, and a third that uses a balance among many sources of energy. "These last three variants represent the biggest uncertainty," says McCarthy. "What are we and our descendants going to do?" It is the future actions of humanity that will have the greatest impact on CO2 concentrations and temperatures. Humans, rather than climate, are the biggest variable, the factor that introduces the biggest uncertainty into the models.

Assuming that humanity makes at least some attempt to rein in its use of fossil fuels, the IPCC forecasts that temperatures will rise in the northeastern United States on the order of 6 or 7 degrees over the coming century. The change will be greatest at night, and in the winter, McCarthy says, as lows fail to dip as low as they once did. Already, winters are warming more than summers, and nights are warming more than days. McCarthy calls these changes "the fingerprints of anthropogenic climate change." They are the result of the increased insulation in the atmosphere caused by greenhouse gases.

The consequences for nature seem profound. Because the rate of temperature increase is projected to be 2 to 10 times that of the last century, some species will be stressed or displaced beyond their limits for survival, the IPCC has concluded. Species unique to the Arctic and to heat-sensitive coral reefs—the tropical forests of the ocean world—are especially vulnerable.

For humanity, presumably a more adaptable species, one of the most startling conclusions of the IPCC report is that no proposed mitigation strategy will preclude some harm to natural and socioeconomic systems from the climate change already underway, so adaptation is not an option—it is inevitable. We can diminish the vulnerability of lives, livelihoods, and properties to anticipated climate change by planned adaptation, like changing agricultural practices and improving public health capabilities. But adaptive capacity is highly dependent upon the state of a nation's socioeconomic development. This country's highly developed food-distribution system, great wealth, and diversity of climate zones will confer relative advantages in adapting its agriculture. But because of extreme events—which are never included in economists' projections of food production, says McCarthy—"the idea that climate change is going to be a win for some and a lose for others, rather than a lose for everybody, is very, very naive."

To the scientists who work with climate models, the risks of loss and the pressures to adapt point to action now, in spite of uncertainty. Of course, there are skeptics who question projections like the Hadley center's model of devastating drought and rainfall in the tropics by 2050, or who discount unusual events as part of natural variability. "It is fine to be skeptical," says McElroy. "But give me a sense of the probability that this particular, reasonable model is wrong. If there is just a 10 percent chance that the model is right, could we risk condemning people to disaster? The precautionary principle that operates here says that unless you are sure that you are not causing a serious problem, don't do it, or at least moderate your behavior."

McElroy's frustration with widespread American apathy is evident. "Shouldn't we react to the fact that we had an incredibly hot summer here in New England?" he asks. "Shouldn't we react to the fact that, for the last several years, the western part of the United States has been up in flames? Shouldn't people react to the fact that we had one of the warmest winters on record last year, and that Central Europe was devastated with unprecedented floods this summer? Is that a smoking gun?" he continues. "If somebody wants to be really skeptical, play roulette, and say we just happen to have spun a thirty-third consecutive red, I can accept that. And I will answer, given the evidence for the likelihood of significant changes in the rhythm of the climate system, that this is not untypical of what you might expect to see. So these events should add to your sense of unease."

"In the next few decades, when we go to atmospheric concentrations of CO2 above 700 ppm," warns McElroy, "we will be going to a place where we have not been for perhaps the last 30 or 40 or 50 million years. This is a uniquely important disturbance of the carbon cycle."


A (Scientific) Bias for Action

So far, American policymakers don't seem to be listening to McElroy or his colleagues. It is as if the scientists and the policymakers speak a different language, and operate on a different clock. Perhaps nonscientists don't understand the nuanced differences of opinion over climate change that exist even among scientists.

Wofsy, for example, says that adding greenhouse gases to the atmosphere will indisputably warm the climate eventually, but that we can't know how much or how long it will take. He believes that anthropogenic forcing is likely to have played a significant role, but is not certain there is no other explanation. "The scientific paradigm that you learned here at Harvard or in high school is that you make a hypothesis and then do an experiment to test it," says Wofsy. "Imagine lining up 10 identical Earths and, because there is a lot of fluctuation, subjecting seven of them to greenhouse gases and leaving three as controls. I don't think we're going to be able to do that experiment, and failing that, what we would ordinarily regard as scientific proof is going to be very hard to come by."

Despite Wofsy's purist views about the scientific method, he regards the very impossibility of ever doing such an experiment as sufficient reason not to wait any longer to take action. "We can't really attribute climate change with certainty to any particular cause," he says. "The political response is, 'We shouldn't do anything until we know more.' The problem is that we won't know more until it is too late. Unless you have an unexpected catastrophe—which is not out of the question—an event so large that it can't be attributed to natural fluctuations, we won't really know. So it is simply irresponsible for policymakers to say that they have to wait for perfect knowledge, because we will never have perfect knowledge."

Wofsy's reaction to policymakers ranges from bewilderment to angry frustration. "We seem to be unable to communicate very effectively with [them] about this," he says. The true conservative approach, he argues, would be to say this looks like a big problem, it seems to be a significant risk, and we want to minimize it by taking steps to reduce the growth in carbon concentrations. But "the current policy is to adapt to climate change, which we can't predict. How can we adapt to something we can't predict?"

Wofsy is not naive about the political process. He and McElroy have worked on the stratospheric ozone problem for 30 years, teasing out how chlorofluorocarbons might destroy stratospheric ozone. But it was impossible to persuade anyone to act until the ozone hole over Antarctica was discovered. By chance, the unusual chemical reactions that destroy stratospheric ozone take place rapidly only at very low temperatures. "But the people who made these compounds didn't know that," says Wofsy, "and if those reactions were fast at higher temperatures, they could have removed all the ozone above the whole globe. The fact that they didn't was not due to careful planning or prudent action. It was dumb luck.

"One had better hope that this doesn't happen with respect to climate," Wofsy continues. "We have never had a situation where the CO2 concentration in the atmosphere has been increased by the type of process that is going on. Generally speaking, we have warm periods in the geological record and high levels of CO2. Most people think the warmth came first and the CO2 was just an amplifier, but we don't know that. And during some of these extraordinary warm periods we had observations of mass extinctions in the biota. Life persisted, but not for everybody. So let's hope that doesn't happen."


An Emissions Cap?

Economists' views on appropriate policy responses to climate change are shaped partly by their interpretation of the science. Most are convinced that our world is warming, but their primary allegiance is to protecting or enhancing economic growth. At what point, they ask, do the damages engendered by climate change outweigh the costs of mitigation? Much of the damage envisioned by scientists lies in the future, and has so far proven impossible to predict with enough precision on a regional basis to allow economists to account for specific effects when weighing the economic costs versus the benefits of a particular approach to controlling concentrations of carbon dioxide. Economic growth, they argue, is therefore one of the best defenses against an uncertain future, since the ability of both nations and individuals to adapt is determined primarily by wealth.

What then, should the ideal policy approach look like? McElroy believes that the process must begin with a long-term target at which atmospheric concentrations of CO2 could be stabilized—perhaps 550 ppm, perhaps 700 ppm—by a specific future date, 50 or 100 years from now. "From there, one can work backward," he argues, and "set a global cap on annual emissions in order to achieve the goal." The key is to have long-term targets, so that investment decisions today are made with the understanding that CO2 emissions will be costly in the future.

But how should emissions rights, which would become extremely valuable under a global cap, be allocated? Ultimately, says McElroy, the fairest approach is on a per-capita basis, which he would make a long-term goal that might be implemented over a hundred years. Even so, that would mean big changes in the United States, which currently emits one quarter of all the CO2 released annually into the atmosphere, but has just 4 percent of the global population.

Robert N. Stavins, Pratt professor of business and government at the Kennedy School, has advocated a framework for constructing an international agreement to reduce greenhouse gases that would include a cap on global emissions, as McElroy suggests. The so-called "cap-and-trade system" has been deployed domestically to control air pollution, but has never been tested internationally. Under Stavins's plan, a global emissions target would be set and then rights to emit distributed to participating nations. The emission rights would be tradable, so nations that received permits but didn't need them would sell them to big emitters that do, like the United States. Stavins's global climate policy posits that all nations must be involved, even if they can't pay in the short term—"otherwise, production of carbon-intensive goods will shift" to non-participating countries, he says, undermining the agreement and making the costs of joining later much higher.

A second key element of Stavins's plan would be to include long-term targets and timetables. "Private industry listens to these signals," he says. "Electric utility executives even now are thinking about anticipated regulations, like the Kyoto Protocol, when making new investment decisions" (see page 43). The third component of Stavins's plan is to use market-based instruments within countries (just as he would among them) to reach the emissions targets. In some countries, that would be best achieved with carbon taxes (see "The China Project," page 87), and in others by using a tradable permit system.

Tradable permit systems are popular domestically, explains Morris University Professor Dale W. Jorgenson, because permits are often distributed free to emitters, and traded among them, so that all the revenue stays within the private sector. Because the rights are valuable, this can create a powerful political constituency in favor of this approach. Carbon-intensive producers would buy permits, while "clean" producers would sell them, creating incentives for development of new technologies to reduce CO2 emissions. These tradable permit systems have worked domestically in the past, sometimes efficiently, such as when they were used to remove the lead in gasoline, and sometimes not, says Jorgenson, citing the Clean Air Act, which did clean the air, but at a relatively high economic cost.

One advantage of Stavins's approach is that it sidesteps the question of how to allocate emission rights. Under a trading system, no matter how the permits are initially distributed, the final allocation "ends up the same," he says. The allocation does affect the ultimate distribution of the burden of costs. "That's why giving extra permits to developing countries," Stavins says, "can make sense politically, economically, and ethically."

But even if they do work domestically, Boas professor of international economics Richard N. Cooper thinks international-scale cap-and-trade programs would be politically unacceptable to most Americans. Because the United States is a big emitter, it would be a big consumer of such permits. That means a lot of money, on the order of 10 times the current foreign-aid budget, would flow from the United States to other countries. Cooper calls some of those countries, particularly in Africa, "the biggest kleptocracies in the world. Middle-class Americans would be subsidizing the lifestyles of the world's super-rich." Stavins acknowledges Cooper's criticisms, but says, "I've studied cap-and-trade programs extensively, and agree that they are the worst possible approach—except for all the others."

Stavins's plan actually allows for emissions increases in the short term, but these would be slightly below the business-as-usual baseline. Over time, emissions targets would slowly begin to curve down and away from the baseline projection. Superficially, the current Bush administration plan to "slow, stop, and reverse" CO2 emissions is similar, but the Bush plan lacks a critical element: long-term targets and dates to reach them. "Unless you put in long-term targets now," Stavins says, "you'll just have the same problem 50 years from now—new power plants that you don't want to make obsolete and close down."

Taxing Carbon

Cooper, who has specific doubts about the predictive capabilities of the science, advocates an alternative, tax-based approach. "I have no trouble believing that we are warming the planet, but I do not find the future projections of climate models persuasive," he says, noting their similarities to the economic computer models that he and his colleagues generate. "As a research tool, they are fine, but as the basis for public policy, they are extremely problematic."

Despite these caveats, Cooper believes that climate change is a very big potential problem, and that there are all kinds of appropriate policy responses. "I would put a lot of resources into learning as much as we can, and I would begin contingency planning for adaptation, both for human society and for the non-human, ecological environment." He also suggests exploratory research into geophysical engineering, such as deflecting solar radiation, undertaking massive sequestration programs by planting trees very efficiently, or even (in an emergency situation) "seeding" parts of the ocean that now have very little life in them. (Oceanographer McCarthy cautions that ocean seeding would disrupt the food web by favoring organisms that are now relatively unimportant in the ocean's ecology. "It should be looked at long-term," he agrees, "but in the near-term, when you compare it with the ease with which we could get another two or three miles per gallon out of U.S. automobiles, it becomes almost unimaginable that we might try to fertilize the southern oceans to compensate for our excesses.")

A critical component of contingency planning, Cooper says, is to make the poor less poor, because this enhances the ability of societies and individuals to adapt. He warns in particular against doing anything in the name of mitigating climate change that would compromise growing incomes in poor countries. "That," he says, "is their safety net for the future."

Economist Richard N. Cooper, undersecretary of state for economic affairs from 1977 to 1981, and chair of the Federal Reserve Bank of Boston from 1990 to 1992, warns against making public policy on the basis of climate-model predictions. In the face of uncertainty, he says, economic growth is an important safety net for developing countries.

Photograph by Jim Harrison

Cooper would begin with a small increase in the tax on oil. Even though he remains unconvinced that catastrophe looms, he would support such a tax for a variety of reasons beyond climate change—energy security chief among them. He would be willing to accommodate the climate-change issue by taxing natural gas and coal—the other major fossil-fuel sources—as well, on the basis of their energy content—in effect, levying a carbon tax.

Fellow economist Dale Jorgenson agrees that taxing the carbon content of fossil fuels has some important advantages over other approaches. "Carbon taxes are easy to administer, easy to understand, and very effective," he says. "We have a lot of evidence from the oil crises of the 1970s and the subsequent decline in prices that energy—fossil fuel in particular—is highly price responsive." Revenues from such carbon taxes would accrue to national governments, and be used to reduce taxes on capital or labor at the individual and corporate levels, thereby providing a counterbalancing economic stimulus. But "Republicans do not want to be labeled the party of the rich and Democrats would probably not do something that could be interpreted as favoring the wealthy," Jorgenson acknowledges. "I think that's why that approach has had relatively little political success in this country."


We're from Missouri

There is no shortage of ideas, scientific or otherwise, about how to deal with climate change. But whether the political will exists to deal with the problem may be the biggest question of all.

"Most Americans are from Missouri on this—'Show me,'" says economist Cooper. "We have to have a flood before they'll believe the river can flood—often it takes two." Cooper has been an economic policymaker in Washington and has more recently studied responses to climate change. "When ordinary citizens hear scientists talking about climate catastrophes," he says, "they don't know what to make of it, but when a politician hears 'in a hundred years' and the time horizon in Washington is at maximum two years to the next election, is the rare politician who thinks beyond the next election."

Fellow economist Robert Stavins puts the problem of dealing with climate change in a democracy another way: "You and I don't observe the climate. We only observe the weather. The changes that Mike McElroy and Jim McCarthy are talking about," he says, "even the most serious ones, are less than the current variability of the weather from year to year on July 31st. So I don't expect to hear the masses crying out for global climate-change policy."

"Look at the history of environmental policy in the United States," says Stavins, who directs Harvard's environmental economics program, based at the Kennedy School. "Whenever we have taken action, it is because we have experienced the equivalent of witnessing a child hit by a car—and then we've put in a streetlight at that corner. When the Cuyahoga River caught fire in 1969, no one said, 'Well, rivers periodically catch fire for natural reasons, so who knows?' They don't—unless they have a lot of petroleum residuals in them." This is not a good way to create public policy, Stavins says, but "we got the Superfund program as a result of Love Canal, and we got the Clean Water Act partly as a result of the Cuyahoga River incidents, so we have a history of this."

There are further political difficulties in international policymaking. One is the so-called "free-rider" problem. "Whether or not China participates, China benefits," Stavins points out. "It doesn't matter where the emissions come from, if the rest of the world signs the agreement." That means, Stavins says, "there is a tendency to have below-optimal action, which is what is happening now. We have a federal government with coercive powers to resolve such disputes, but we don't have a world government." (There is a domestic analogy: when downwind New England states seek improvements in the quality of air emanating from polluting Midwestern states.)

Although Stavins and Cooper agree about the obstacles to action, they have different visions of how the political process should unfold. Stavins believes action will have to come from the elites, as it did when policymakers addressed the stratospheric ozone problem. Cooper's sympathies lie with the average citizen: Accepting that a disaster looms requires a big leap of faith, he says, and "Americans are not used to taking big leaps of faith. We're betting on what we know are some extremely imperfect guesses about the future."

In fact, Cooper feels that no one discussing solutions to climate change is being honest about the implications. "We're talking about a change in lifestyle," he says. "That's why it is politically difficult." Some climate-change scientists, essentially, are asking us to spend a lot of money to reduce an uncertain risk. Cooper, however, thinks the core problem is a security issue. He draws an analogy to the Cold War, when even with imperfect knowledge about the Soviet Union's capabilities, the nation spent hundreds of billions of dollars and devoted tremendous national resources to reducing the risk of an attack. "Who is to say we might have spent less" and still achieved a safe outcome? Americans' willingness to spend that kind of money was formed in two world wars, in particular by what happened at Pearl Harbor, he argues. "It is hard to mobilize any society on the basis of a hypothetical. There are too many."

Embracing a Nuclear Future?

The economic costs associated with reduced use of fossil fuels are, as Cooper points out, unimaginable without extraordinary gains in efficiency, or more likely, massive increases in noncarbon-based sources of energy. Where will the power come from? The answer may lie, in part, in various alternative energy sources such as solar and wind power, hydroelectricity, and geothermal energy. These are the kinds of clean power normally associated with environmentalists. But climate-change science has also made for strange bedfellows: nuclear power will have an important role to play, McElroy and Cooper agree. Coal, which causes both gaseous and particulate air pollution, "has killed many more people than nuclear power ever has," says McElroy, even counting the Chernobyl meltdown—which was caused by a uniquely flawed power-generating technology that will never be used again. There is more background radiation coming from the granite in the Rocky Mountains, he says, than has ever been released from an American nuclear power plant.

On balance, it seems to be asking a lot of the average American to accept nuclear power as the major source of electricity in this country, as it is in France. The obstacles are daunting.

But they are perhaps not insurmountable. When James McCarthy was on the ship off Alaska this past summer, he was asked (as he inevitably is) if there was any reason to be hopeful. He thinks so. Last May, McCarthy and a few other scientists presented a report on climate change to the CEO of Shell and were encouraged by his response. "It's credible," he told them, "the sort of thing that should make sense to any CEO." British Petroleum has led the oil industry in cleaning up emissions and greening its image: BP set internal targets for greenhouse-gas emissions consistent with the Kyoto Protocol and adopted a system for trading emission rights among its major business units. (Its actions are the subject of a Harvard Business School case study.) BP met its targets this year, eight years ahead of schedule, and has even rolled out a new moniker: "Beyond Petroleum."

"As far as market opportunities," says McCarthy, "I think this field is wide open. Detroit once said they could not sell safety, that no one would pay for air bags and antilock brakes. Given the choice, would most people today say, 'Give me the radio, but keep the air bag?'" McCarthy believes that the climate issue is changing, that people are beginning to realize that they should bear the costs to minimize climate change, and that the costs are never as steep as predicted. The people who are saying of mitigation today, "'This is too costly, we can't do anything about it,'" says McCarthy, "are the same people who in 1990 said, 'Nothing is changing,' and in 1995 said, 'Well, it is changing a little bit, but it is not due to human actions and there is nothing we can do about it.'"

"They have been forced to move their argument along the way," says McCarthy, "to accommodate the irrefutable evidence that has accumulated in an enormous mass."


Jonathan Shaw '89 is managing editor of this magazine.        

Read more articles by Jonathan Shaw

You might also like

Former Women’s Hockey Coach Sues Harvard

Katey Stone alleges gender bias in handling of abuse allegations that led to her retirement.

Remembering Helen Keller and Anne Sullivan

On a Radcliffe-Harvard memorial to remarkable figures

Harvard Confers 11 Undergraduate Degrees

Protestors now found in “good standing.”

Most popular

Former Women’s Hockey Coach Sues Harvard

Katey Stone alleges gender bias in handling of abuse allegations that led to her retirement.

Mechanical Intelligence and Counterfeit Humanity

Reflections on six decades of relations with computers

Remembering Helen Keller and Anne Sullivan

On a Radcliffe-Harvard memorial to remarkable figures

More to explore

Broadway Director from Harvard Adapting Disney

Broadway music director Madeline Benson on art and collaboration

How Political Tension on Campus Creates Risk Aversion

How overheated political attention warps campus life

Harvard Professor on Social Psychology for Understanding War

Two scholars’ extracurricular efforts in the Middle East