The Climate Jump Heard 'Round The World

In July 2004, nearly 5,760 meters sky-high in Peru’s Andes Mountains, the Science Bulletins film crew trailed a pair of glaciologists hiking the summit of Quelccaya, the largest tropical ice cap in the world. Cameras and sound booms in hand, we aimed to film the Ohio State University team drilling an ice core from the top of this 44-square-kilometer layer cake of accumulated snow and ice. But then Lonnie Thompson, the lead researcher on the National Science Foundation-funded project, took a detour. He headed toward a glacial lake pooling in a valley three kilometers away. Had we followed him, we may have caught on video a discovery of some importance. The find has profound implications for the fate of this ice cap and its kin that dot high-elevation sites at similar latitudes.  

Frozen in Time?

Thompson’s eureka was a clump of brittle vegetation rooted near boulders flanking the ice-littered lake. It resembled a cow patty more than a plant. Thompson’s colleague Benjamin Vicenio scooped several plants into a plastic baggie, noting their proximity to the abrupt edge of the ice cap looming nearby.

Upon return to the United States, Thompson mailed a portion of the plant to the Woods Hole Oceanographic Institution in Woods Hole, Massachusetts, for carbon-dating. The verdict: more than 48,000 years old. “There had to be a mistake, I thought,” said Thompson. “They’ve got too many digits. It’s supposed to be 4,800 years old.” Thompson’s shock is understandable. After all, three previous samples of ancient plants near Quelccaya’s edge turned up dates near 5,000 years old. But a second carbon-dating at Lawrence Livermore National Laboratory confirmed that the “new” plant was indeed extremely old; according to Livermore’s method, it was more than 55,500 years old.

Like the 5,200-year-old “ice man” found in the Austrian Alps in 1991, the plant had been cryo-preserved in ice. Thompson suspects it was exposed only in the last few years as the 0 degree C ice cap shrinks in the 3 degree C air.

Plants at Quelccaya are often rooted in soil-filled depressions in the bedrock. Thus, Thompson suspects the plant was preserved as the cap advanced over such a depression. This left a chamber of trapped air that spared the plant from being scoured to tatters by the slow-flowing glacier.

Melting Margins

To identify the plant, Thompson called on Blanca Léon, a botanist from the University of Texas at Austin. Looking at the plant under a microscope, Léon observed the cell walls of a fragile moss that was remarkably well preserved. “It looks like a plant that was taken recently from the field,” she says. “It is amazing.” Called Scorpidium, the moss still grows today along the edges of lakes in the Andes, but at warmer elevations no higher than 4,600 meters. The ancient plant was collected from Quelccaya at 5,000 meters. Léon concluded that “the plant must have existed at a time with warm conditions, enough to allow its establishment at such a high altitude.”

Thompson knows, however, that paleoclimate records derived from Antarctic ice cores show that no warm period occurred 50,000 years ago. Also, current carbon-dating technology cannot date objects older than 55,000 years or so. Therefore, Thompson suspects the plant is actually much older.  Around 125,000 years ago, temperatures were about 2 degrees warmer than they are today,” he says. “I think that’s when these plants were actually growing. Then a cooling cycle began and the glaciers grew and buried the plant.” The plant was exposed and discovered in the current warming period.

Thompson has witnessed our current warming first-hand. Since 1978, he has snapped near-yearly terrestrial photographs of the cap’s largest outlet glacier, Qori Kalis. During each expedition, he spots more bedrock that has been exposed and lakes swelling or appearing out of nowhere. He sees newly freed boulders perched on crests of gravelly moraines, the piles of rocky debris left behind by the retreating wall of ice. The soundtrack to every trip is the constant drip, splash, trickle, and rush as ice turns quickly into water.

From his data, Thompson has determined that 20 percent of the ice he measured at Qori Kalis and Quelccaya as a whole in 1978 has since melted. “In the first measurement period, which was from the first aerial photographs of 1963 until our photographs in 1978, the terminus of Qori Kalis was retreating at a rate of about 4.7 meters per year,” he says. “From about 2000 to 2002, that rate had increased to over 200 meters per year, or over 40 times faster.”

It appears that global warming is shrinking every last tropical glacier, including those at the summits of Kilimanjaro, Mount Kenya, and the Himalayasall sites that Thompson has surveyed and cored in the past two decades. “Quelccaya will take 50 years or so to disappear completely given the current climate conditions,” he says. “And of course, the predictions are that things will get warmer over the next 50 years.” This summer, Thompson plans to return to Peru to drill more cores and sample more plants to better reconstruct the ice cap’s history before there’s no ice cap left to study.

If all the ice on, say, Greenland melted, it would raise global sea levels by about 6 meters. But the melting of tropical glaciers would contribute just a fraction to sea-level rise. Instead, Thompson considers the retreat of tropical glaciers ominous because they’re the proverbial “canary in a coal mine”: a bird taken into mines to test for the buildup of dangerous gases. “You have two choices,” says Thompson. “Either you take the warning and get out of the mine or you choose not to take any action. To me, the tropical glaciers are our canaries. And they're telling us the system is changing.”

What the ice gets, the ice keeps," said polar explorer Ernest Shackleton in July 1915. Shackleton himself was almost a keeper: He was moored for 15 months on an ice floe in Antarctica's blizzard-ridden Weddell Sea. His ship was en route to the continent, which Shackleton intended to cross via dogsled, when pack ice enveloped it. Pressure from the merging ice blocks crushed the ship, yet Shackleton and his entire 28-person crew survived. It is one of history's most harrowing and inspiring tales of endurance and rescue.

Glaciologists, our modern-day ice explorers, voyage to the globe’s frostiest, most unforgiving locales to retrieve what glaciers keep in their coffers--compacted snow and trapped air from ancient atmospheres, ash from long-quiet volcanoes, dust, insects, and pollen. By analyzing these relics, scientists can reconstruct hundreds of thousands of years of Earth’s climate. This frozen-in-time history is crucial in understanding how the climate system works today and how it may change.

Scientist-Adventurers                                   

“It is cold, it is windy, it is dangerous, it is an extraordinarily awkward environment to work in,” gasps Keith Mountain cheerfully as he stands atop the oxygen-poor Quelccaya ice cap 5,760 meters high in Peru’s southern Andes. “With such a bright surface, the possibility of radiation burns is extraordinary. Within a short time you can be in blizzard conditions, where you can’t see in front of you to take a measurement.” He steadies a hand-operated ice core drill with expedition-weight mittens. The drill is ready to bite into the 168-meter-thick frigid flooring that has existed on this peak for tens of thousands of years. With temperatures regularly below freezing, high-altitude spots like the Andes and high-latitude land such as Antarctica and Greenland preserve snowflakes. As the snow is compressed by the layers accumulating above, it gradually hardens atop the bedrock into a thick coating of glacier ice. It’s called an ice cap if it radially coats a peak, or an ice sheet if it spans a significant part of a continent and has submerged the underlying topography.

“You face a challenge between how long you can sustain in these conditions versus what you’re trying to get in terms of science,” says Mountain. The University of Louisville geoscientist is trying to extract one-meter lengths of a vertical cylinder of stratified ice 11 centimeters in diameter and several meters long. The core – and the six-day expedition – are short ones for Mountain and his colleague Lonnie Thompson, a glaciologist from Ohio State University. Thompson pioneered tropical ice coring and has participated in 47 major expeditions in 15 countries since 1974. The short core updates the 168-meter core pulled from Quelccaya’s summit in 2003, which required hauling six tons of equipment, including a solar-powered automatic drill, up the peak for a one-month stay.

After drilling is complete, local porters sled the cores to the edge of the ice cap, pack them in insulated boxes, and strap them to horses, which tote them 30 kilometers to a freezer truck waiting at the closest access road. The cores are driven across the Andes to Lima, sit in cold storage until approved by customs, and are air-cargoed to “30 degree C freezers at Ohio State University to await analysis. 

Ghosts of Climates Past

Scientists compile data from ice cores around the globe to chart Earth’s overall temperature, precipitation, atmospheric composition, and climate patterns for hundreds of thousands of years. Quelccaya’s complete core provides a high-resolution “time capsule” back to 200 B.C. A new, lower-resolution 3,270-meter core pulled from Antarctica last December may extend the ice-core record back to 900,000 years.

Ice cores reveal:

Timelines: In most tropical ice cores, the boundaries between layers of annual or seasonal snow are visible enough to count the core’s age. By analyzing the climatic signatures of each layer, researchers can date changes and events precisely. Bands of particles deposited from ancient volcanic eruptions of known age help delineate years or seasons. Frozen insects and other organic material swept into the snowfall by wind are carbon-dated, which can confirm dates obtained by counting layers.

Precipitation: Snow is compressed as it builds up into glacier ice. Thus, newer layers tend to be thicker, while older, deeper layers are thinned. After core depth, accumulation rates, ice temperatures, and ice flow are taken into account, the layer’s thickness roughly corresponds to how much snow fell at the coring site that year or season.

Atmosphere: As snow turns to ice, it traps bubbles of air. The air and frozen water provide chemical records of the composition of Earth’s atmosphere at different times. ”We can tell when lead was put into gasoline. We can see when legislation was passed to take the lead out of gasoline,” says Thompson. Polar ice cores have revealed preindustrial atmospheric concentrations of greenhouse gases: carbon dioxide, methane, nitrous oxide, and other gases that absorb infrared radiation emitted by the Earth’s surface, and trap heat in the atmosphere.

Temperature: By analyzing the oxygen atoms in the frozen water of an ice core, scientists can determine the average temperature of the air above the core site for that period. The technique is based on the fact that not all oxygen atoms have the same number of neutrons. These variations, which differ in mass but not chemical behavior, are called isotopes. Oxygen has three isotopes: oxygen 16, oxygen 17, and oxygen 18.

Here's how the technique works: As water vapor in a cloud condenses to form precipitation, slightly more oxygen 18 goes into the precipitation, leaving the remaining vapor enriched in oxygen 16. During glacial periods, the colder temperature of the atmosphere means that the moisture-laden air will lose more water to precipitation before it arrives at the glacier site. Thus the oxygen 18 to oxygen 16 ratio will be less in the snow deposited on the glacier during colder periods, providing a proxy for paleotemperature.

In Quelccaya cores, says Thompson, “you can see the medieval warm period when the Vikings settled in Greenland. You can see the onset of the Little Ice Age in the early 1500’s that contributed to the demise of the Vikings. And you can see the warming in the 20th century.”

Our Warming World

Ice core records, when combined with data from other paleoclimate proxies such as ocean sediment cores and tree rings, reveal that our climate has drastically changed many times over the last 100,000 years. It has switched from glacial periods, when ice sheets blanketed nonpolar areas, to warmer interglacial periods, and back again. These switches occurred slowly (taking hundreds of years) or abruptly (in as little as a decade).

During the last 100 years, global surface temperature has risen between 0.5 °C and 0.8 °C. This rise may have resulted from increases in greenhouse gas concentrations caused by increased burning of oil, gas, coal, and vegetation. Global warming has possible worrisome consequences, among them melting glaciers, an associated rise in sea level, regional drought, and more severe climate events. Most worrisome, however, is the possibility of changes that we cannot foresee with our current understanding of how the climate system works.

Thompson notes that the real challenge glaciologists face is not the tough expeditions, but ensuring that the information they obtain is used effectively. “If all we do is collect these records, we're just historians,” he says. “If we cannot take that data and bring about meaningful policy changes that make for a better world for the future, in some ways I think we may be wasting our time.”

Certain landscapes in Canada, the northern U.S., and northern Europe are bona fide strange: Lone boulders squat on grassy, gentle hills. Rocks are raked with deep scratches. Rectilinear piles of gravel are aligned almost too perfectly in a single direction. Swiss geologist Louis Agassiz was among the first to realize that ancient, monstrous sheets of ice spanning entire continents produced these odd geological leftovers. In 1837, such an idea was highly controversial.

However, 150 years of follow-up research is finding that Earth’s climate has actually undergone many such glaciations in the last 2.5 billion years. In the past two million years alone, Earth has experienced around 20 ice ages--cycles of advance and retreat of large continental ice sheets. Currently, Earth is between glaciations. If nature has its druthers, we’re probably not due for the next big chill for tens of thousands of years. How exactly our anthropogenically influenced global warming is forcing large-scale natural cycles, however, remains to be seen.

Plotting the Pleistocene

The glaciation that scientists in the 1800s noticed was our most recent one. At the maximum extent of its ice sheets 21,000 years ago, Earth’s air temperature was, on average, about 4 degrees C cooler than today. Around 30 percent of the land surface was covered with ice up to 3 km thick. The sheets carved the basins of the Great Lakes and bulldozed the gravelly ridge we now call Long Island.

The event took place in the Pleistocene Epoch, which began about two million years ago and ended about 10,000 years ago. For roughly the first half of it, every 40,000 years or so contained a single cycle of prolonged, extensive glaciation, then a shorter warm period. For roughly the last half, each cycle took 100,000 years.

This timing became clear only in the early 1970s. That’s when researchers working in the Indian Ocean drilled the first cores of deep-sea sediment deposited throughout the entire Pleistocene up until current times. By measuring oxygen-isotope ratios in the carbonate “rich shells of tiny marine organisms buried in the strata of these cores, scientists estimated the temperature of the ocean’s surface when these organisms lived. Using computer models, scientists were able to infer average global temperatures during the entire Pleistocene from this data.

Milankovitch Cycles

The effort to explain how glaciers retreat and advance began decades before scientists studied these cores, however. The Indian Ocean work simply confirmed long-debated speculation that astronomical cycles may have timed the Pleistocene’s glaciation.

Cyclical changes in the way Earth orbits the Sun and spins in space were worked out by Serbian mathematician Milutin Milankovitch in the early 20th century, based on calculations made by two earlier scientists.

Three changes were scrutinized: 

Orbit: Earth orbits the Sun in a slightly elliptical path. Sometimes, the orbit is more elliptical than at other times. The shape of the orbit changes the maximum distance of Earth from the Sun, and with it the amount of solar radiation Earth receives. The transition from “most circular” to “most elliptical” and back again takes about 96,000 years.

Tilt: The tilt of Earth’s axis of rotation also varies. It shifts between 21.5 and 24.5 degrees in a cycle of 41,000 years. The tilt affects where the globe is receiving the most solar radiation. During times of more tilt, higher latitudes receive more sunlight.

Precession: Earth doesn’t rotate perfectly around its axis. Instead, it wobbles like a top, a motion called precession. Precession influences the amount of solar radiation striking a given location for a given season. This causes the difference of temperature between seasons to be either large or small. For example, sometimes winters will be extra frigid and summers extra warm (large difference). Other times, mild winters are followed by cool summers (small difference). Precession operates on a 21,000-year cycle.

Milankovitch reasoned that these cycles could work together to vary the amount of sunlight a given place on Earth receives by 20 percent, especially at high latitudes. That could nudge the advance of the polar ice caps: Less radiation at the poles would mean more snow would survive until the next season. Snow would therefore increasingly accumulate into glacier ice.

Milankovitch cycles explain much about how and when climate changes have occurred in the last 2.5 billion years. But they’re not the whole story. The magnitude of large-scale climate change is influenced by many Earth-bound factors. Among them are changes in topography and plate motions, the hydrosphere, the biosphere, and the atmosphere. The concentration of atmospheric greenhouse gases is important: The higher the concentration, the more these gases trap escaping radiation close to Earth’s surface. Ice-core analysis indicates that levels of greenhouse gases were lower during glacial periods than interglacials.

Welcome to the Holocene

After continental glaciation reached its largest extent 21,000 years ago, the Pleistocene began to warm. Prodded by Milankovitch cycles, the ice sheets shrank. After about 11,000 years ago, humans began to cultivate food, domesticate animals, and build cities in the continuously stable climate. Enter a new interglacial period, and with it, a new epoch: The Holocene. This, as they say, is our time.

“There’s no danger of an ice age popping in now,” says Penn State glaciologist Richard Alley. “I believe that most people studying this field think that, without any human intervention, a new ice age should arrive 20,000 years into the future.” Likewise, it’s generally accepted that our civilization's increase of atmospheric greenhouse gases could make it tougher for that ice age to get going. However, William Ruddiman, a climatologist at the University of Virginia, is one scientist that suggests that the transition to the next glaciation should have already begun. Ruddiman's research says that early, pre-industrial societies generated enough greenhouse gases to actually stop a current-day ice age from happening altogether. But whether nature or humans avoided an ice age recently, we still have 20,000 years or more to wait before the next one.

“It’s a fascinating idea,” says Alley, “but it doesn’t mean anything for the future. If you’re concerned about staving off a new ice age, the wise thing would be to save your fossil fuels for 20,000 years from now and burn them then. But nobody has 20,000-year planning in their blood.”

Climate change isn’t always slow, small, and imperceptible in a human lifetime. One of the most important lessons from ice core analysis is that Earth’s climate in some places can also change rapidly and dramatically, such as a 15-degree temperature change in a decade. This, you’d notice.

About 12,900 years ago, Earth was warming steadily, in resolute recovery from our most recent glaciation. Then, suddenly, things got a lot colder and drier. The ice sheets reversed direction. Much, if not all, of the North Atlantic Ocean froze over. The landmasses flanking it became colder. Elsewhere in the world, Asia and Africa were beset by dust storms, and tropical wetlands dried up. In some areas, flora and fauna--and likewise, early humans--responded to the climate change in as little as two human generations.

After about 1,300 years of odd behavior, this interruption, called the Younger Dryas, ended. Earth resumed its warming course. One important study of Greenland ice cores showed that in a single decade of the Younger Dryas’s departure, temperatures over Greenland shot up about 15 degrees C. That’s as if your local climate changed to one typical of that 1,500 kilometers south of you in the geologic blink of an eye.

Some mechanism, then, was overriding the gentle long-term astronomical rhythms, called Milankovitch cycles, that govern when glaciations come and go.

Could climate switch this abruptly again? It’s not impossible. But global warming complicates the scenario. Interestingly, human impact on the planet may have the power to both force such a rapid changeor prevent it entirely.

The Heat Is On

According to the Met Office, the UK government’s meteorological branch, nine out of the ten warmest years on instrumental record (since 1861) were in the last decade. Today’s temperatures, however, aren’t unprecedented in recent geologic history. About 10,000-6,000 years ago, after the Younger Dryas left and the current interglacial period continued, temperatures in many northern locales were roughly 1 degree C warmer than recently. Epochs previous to the Pleistocene saw warmer periods as well.

Warm periods and high concentrations of the greenhouse gases methane and carbon dioxide usually occur together, although cause and effect are not clear. However, current-day levels of greenhouse gases may force the issue. Carbon dioxide is at its highest concentration in 260,000 years. And no mechanism other than our unprecedented burning of organic material such as fossil fuels and forest biomass can explain carbon dioxide’s increasingly rapid rise of late.

Computer models cited by the International Governmental Panel on Climate Change in 2001 predict that if greenhouse gases double their pre-industrial levels, average global temperatures will rise between 1 and 5 degrees C. A February 2005 modeling study in the journal Nature makes a more aggressive claim. Using nearly a hundred-thousand personal computers, the simulation produced a 2 to 11 degree C rise in global surface temperature in response to doubling of atmospheric carbon dioxide. In most IPCC model predictions, doubling will be reached before 2100.

Conveyed by the Ocean

We can already spot earthly changes from global warming. Arctic and subarctic permafrost is thawing. Winter ice on lakes and rivers breaks up earlier in the season. Trees flower earlier.

Another significant effect is the rapid melting of glacier ice. The overwhelming majority of it – including ice on tropical peaks like the Andes and the Himalayas, Antarctic ice shelves, and portions of the Greenland ice sheet – is diminishing. In many places, the ice is melting at an increasingly faster clip each year.

Similar mass melting occurred during the interglacial period preceding the Younger Dryas. Melt water from the ice sheet covering central and eastern Canada collected in a gargantuan lake that dammed behind the ice. As the ice melted further, the dam burst. The water rapidly drained eastward to the sea via the St. Lawrence River. This rapid flooding covered the North Atlantic Ocean with a layer of fresh water.

These freshwater floodgates appear to have been the trigger that set off a complex chain of events that directly resulted in the Younger Dryas’s cold and dry plunge. The first step was the fresh water’s influence on the ocean: specifically, its thermohaline circulation. This large-scale circulation between oceans is intimately tied to water temperatures and saltiness, two parameters that affect water density.

In this conveyer-belt-like circulation, warm, relatively salty tropical Atlantic Ocean water flows northward. Upon arrival in the North Atlantic, the water warms the atmosphere and keeps northern Europe comfortable. As this salty water cools in the winter, however, it sinks. Now denser, this water flows nearer to the ocean floor back down to the South Atlantic. There, it is joined by sinking cold water from Antarctica and flows as a deep current into the Pacific. The Pacific water eventually returns to the Atlantic in surface currents in a global circulation that takes perhaps a thousand years or more.

How is this connected to the Younger Dryas? Fresh water is less salty and hence less dense than seawater. The sudden influx of fresh glacial melt water into the North Atlantic floated atop the warm salty water already there. The top fresh layer wasn’t salty and dense enough to sink. Thus, it prevented the salty water underneath from cooling in the winds and sinking. This stopped the thermohaline circulation in its tracks. It’s as if you stuck a fork in the “down” end of that grocery conveyor to cease its descent.

With no cold water sinking, the warm surface water flowing northward in the Atlantic stopped arriving. Without warm replenishment, the temperature of the North Atlantic Ocean surface dropped drastically. Ice formed on the surface. The Europe-bound winds were now cooled by the frozen ocean.

This freeze-over is a major switch for global climate, according Richard Alley, a Penn State glaciologist whose ice-core and glacier work has contributed significantly to our understanding of Younger Dryas events and mechanisms. “When you put ice over the top of seawater and let it get really ridiculously cold, you’re turning the ocean into a continent in the winter time.” After about 1,300 years, the saltiness of the North Atlantic had increased enough to allow the conveyor to start up again. The Younger Dryas was finished.

Could today’s global warming cause a conveyor shutdown? Models by the Met Office and other climate-change groups don’t predict a total switch-off within the next century. “Now what [the scientific community] needs to do is put probabilities on these events,” says Alley. “But we’re simply not quite good enough at doing that yet.” He likens the gamble to that of buying insurance. “You sort of know how much to spend on car insurance because you sort of know what the odds are of someone running over you with an SUV,” says Alley. “But we don’t know how much to spend on North Atlantic shutdown insurance.”

About 12,900 years ago, the North Atlantic region quickly--in years to a decade--fell into a deep freeze. The ice sheets on Greenland, North American, and Eurasia advanced. Forests in eastern North America, Europe, and Scandinavia turned to tundra, and subarctic animals such as caribou migrated southward into these new landscapes. Then, after about 1,300 years, the area rapidly warmed again. In Greenland, the jump was 15 degrees in just a decade.

The unusual and rapid climate event, called the Younger Dryas, has been well-studied by climatologists seeking understanding of how, and how quickly, climate can shift dramatically. They’ve discovered over the last 15 years that the Younger Dryas was felt not just in the North Atlantic, but over half the globe. But it’s only been in the last two years that scientists have begun to suspect why.

A World of Change

Generally, the world became colder, drier, and more dusty during the Younger Dryas. Dust blown from increasingly stormy deserts in Asia scattered over glaciers in Greenland, revealed by ice core records taken there. By analyzing methane concentrations in air trapped in ice cores, scientists realized that tropical wetlands, which produce methane naturally, had significantly shrunk. Ocean sediment records from waters off California, Venezuela, and Pakistan demonstrated that the Younger Dryas’s climate jumps impacted the ocean and atmosphere of these regions in significant ways. 

The temperature decreases and increases in the North Atlantic that marked the beginning and end of the Younger Dryas also appeared in other areas of the globe. Determination of the ratio of the isotope oxygen-16 to oxygen-18 in stalagmites in China’s Hulu cave revealed a temperature record for 75,000 “ 11,000 years ago. Its ups and downs matched those in the North Atlantic, as revealed in ice cores there. The data also suggest that the strength of China’s monsoon rains shifted suddenly during the Younger Dryas.

Understanding the System

“Since the beginning, the trigger for the Younger Dryas’s climate jumps was attributed to reorganizations of the Atlantic Ocean’s conveyor-like circulation,” says paleoclimatologist Wallace Broecker of Columbia University’s Lamont-Doherty Earth Observatory. Broecker himself first suggested the idea during the 1960s. “The idea is that shutdowns and startups of the circulation somehow drove large changes in climate,” says Broecker. “For at least a decade, scientists struggled in vain to characterize the ‘somehow.’ Then two years ago, two breakthroughs occurred.”

Broecker points first to a field study in Greenland’s mountainous eastern fjord region organized by Gary Comer, the founder of Land’s End clothing. Comer, a frequent financial backer of science research, himself participated. He took climatologists Richard Alley of Penn State and George Denton of the University of Maine aboard his yacht and floatplane to a deep inlet called Scoresby Sund. Aerial and ground analysis revealed new information about a set of previously studied glacial moraines: that they were likely deposited during the Younger Dryas. The positions of the moraines, which depend on the positions of the glaciers from which they were deposited, indicated to the group that Greenland temperatures during the Younger Dryas were only 4 to 6 degrees C colder than they are now.

“This stunned them,” says Broecker. The temperature differences were expected to be much greater. In fact, Jeff Severinghaus of the Scripps Institution of Oceanography had shown from isotope measurements of air trapped in ice on Greenland’s summit, 500 miles to the west of Scoresby Sund, that the mean annual temperature during the Younger Dryas was not a mere 4 to 6 degrees C, but 16 degrees colder than now. Then Comer’s team realized that the moraines must have formed in the summer. The 4 to 6 degree C, then, was the shift in summer temperature, not the winter temperature. This means that Greenland winters must have been much colder--perhaps 30 degrees or so--than they are now.

“The only way in which such frigid winter temperatures could be maintained was to cut off all heat release from the surrounding ocean by freezing over the entire northern Atlantic,” Broecker notes. “This created the Siberian-like winter conditions extending from Canada to northern Europe.” In other words, the formation of sea ice “amplified” the cooling to create the extreme winter conditions extending from Canada to northern Europe.

A second breakthrough came from climate models that suggested the large cooling caused by sea ice cover in the northern Atlantic would push the tropical rain belt further to the south and also weaken Asian monsoons. These changes in atmospheric circulation caused cooling throughout much of the northern hemisphere, as evidenced exactly in the Hulu Cave and ocean sediment climate records. So when the Atlantic’s conveyor circulation shut down, it was the formation of sea ice that not only amplified the cooling in the north Atlantic but also caused the climate signal to be transmitted via the atmosphere to the tropics.

“These findings have important implications for the future,” says Broecker. “Some ocean models suggest that, if we were to add enough greenhouse gases to the atmosphere to warm the planet by 4 to 6 degrees C, then the extra precipitation at high northern latitudes would lead to a shutdown of conveyor circulation.” Since shutdowns seem to accompany cold periods in the past, some scientists have warned that a future shutdown could plunge northern Europe into an ice age. “Not likely,” says Broecker. “A 4 to 6 degree C warming would eliminate the possibility of sea ice formation.” And without sea ice, there would be no way to amplify a cold signal throughout the North Atlantic!