by Jean Jouzel
The position of this terminal tongue, the front of the glacier, fluctuates over time. And the length of this glacier, a relatively easy parameter to determine, was measured directly in many sites in the Alps beginning in the 1870s. This is not, however, a very good indicator of climatic variations. Bernard Francou and Christian Vincent emphasize this:1 two neighboring glaciers located in the same massif and having an analogous climatic history can undergo very different fluctuations in length that are asynchronous. In fact, the front of a glacier evolves as a function of its size, the characteristics of the local topography, such as the slope and orientation to the Sun, and its flow conditions, in particular near the bedrock. From a glacier’s variations in length, we cannot deduce whether the entire mass of ice is increasing or decreasing. Even if, outside periods of advance (1890–1900, 1910–30, and 1960–90), the retreat of Alpine glaciers is rather homogeneous, some do exactly as they please. However, if we look at rather long periods of time, that is, more than a decade, using a statistical approach over a given region, variations in length (Figure 2.1) provide a first rough indication—which must be used with caution—of the evolution of the climate. This approach, taken by Hans Oerlemans, reveals that a retreat of glaciers was beginning rather homogeneously on a global scale at the beginning of the nineteenth century, became almost universal beginning in 1850, and continued throughout the twentieth century.
Mass Balance: The Health of a Glacier
More than a glacier’s advance or retreat, its health depends on its mass balance, that is, the difference between accumulation and ablation. At the end of summer the quantity of snow that has accumulated during the year is measured using markers or drillings a few meters deep, spread out on the surface: that quantity indicates any increase in the mass of the glacier. Markers and altimetric measurements in the zone of ablation thus enable an evaluation of any loss of mass. The oldest series is that of the Swiss Clariden Glacier, begun in 1914. However, such monitoring, which is very cumbersome, is restricted to only a few glaciers; in 2005, only a dozen throughout the world had been observed for more than forty-five years. A more sophisticated approach involves estimating the altitude of the entire surface of a glacier at a given moment using topographical maps drawn from aerial or land photography. This technique, called photogrammetry, enables the reconstitution of past variations in volume. For the French Alps, some photographs from the 1950s are useful in this regard. They are, however, much less precise than current aerial views that are of excellent quality. To go back further in time, to the beginning of the twentieth century in the Mont Blanc massif, old maps, drawn with extreme minutiae, can also be used. In the near future satellite images should be good enough to systematically determine the mass of glaciers.
This annual assessment depends on climatic variables, in this case on a regional scale, as seen in the similar tendencies observed in different glaciers on the same mountain massif. The response time, due to their inertia toward climatic change, depends on the characteristics of the glaciers. By comparing the relationship between accumulation or ablation to the mass of the glacier, we arrive at values for the response times of about a dozen years for the smallest glaciers and of a century for the largest. More than the annual mass balance of glaciers, it is the cumulative mass balance since an initial date that is most often used to characterize a glacier. This cumulative mass balance is expressed in terms of variation of average thickness compared to the surface area of the glacier. Thus the Glacier de Saint-Sorlin has a negative mass balance; it has lost the equivalent of 44 meters of ice during the last 100 years. The mass balance is also negative for the other glaciers in the French Alps. This applies to all the Alps with a remarkably parallel trend in Austria, Switzerland, and France. In the Swiss Alps, it is estimated that the volume of the glaciers was reduced by nearly 50% between 1850 and 1999, from 107 to 55 km3, and that over the entire Alpine massif the surface areas of the glaciers have decreased by 40%, going from 3,800 to 2,500 km2.
Figure 2.1. Smoothed variations in glacier lengths on the scale of large regions since the beginning of the eighteenth century. Glaciers are grouped as follows: Atlantic (South-Greenland, Iceland, Jan Mayen, Spitzberg, and Scandinavia), Alps, Southern Hemisphere (tropical glaciers, New Zealand, Patagonia), Asia (Caucasus and Central Asia), and North America (mainly the Canadian Rockies). Source: IPCC, Climate Change 2007: Fourth Assessment Report (Cambridge: Cambridge University Press, 2007)
On the whole, the Alpine glaciers are retreating; this is also true, at least over the past fifty years, for other glaciers on the planet. The most marked shrinking has been seen in Patagonia and North America, then in Asia and in Arctic regions. This diminution is smaller and practically nonexistent in Europe, which is surprising after the picture we have just sketched of the Alps. In fact, the European assessment accounts for both the Alps and Scandinavia. The glaciers in northern Europe are distinguished by surprising new advances in the 1970s, a tendency that accelerated at the end of the 1980s. This advancing, limited to glaciers near the coasts, can be attributed to an increase in winter precipitation and in the amount of snow, all the more since a portion of the autumn rains have moved to the winter months.
Let’s now look at the situation that threatens the tropical glaciers, those located in the northern part of the Andes, in East Africa, and in New Guinea. Looking at the Rwenzori Mountains, on the border of Congo and Uganda, the area of the glaciers there, now less than 1 km2, is one-seventh what it was in 1906, and we expect them to disappear completely during the next two decades. The same diagnosis has been made by glaciologists regarding the mythical Kilimanjaro Glacier, but the causes of its marked shrinking are hotly debated: rather than a direct consequence of climate warming, the glacier’s reduced size could be linked, at least up to a recent period, to a drier climate. The glaciers of the Andes are evolving similarly to those of East Africa: the cumulative losses over the recent period have been as much as one meter per year. Some small glaciers, like the Chacaltaya Glacier in Bolivia, have recently disappeared. The larger ones resist better, but for how long?
The Arctic Ocean in the Time of the Explorers
Beginning in the sixteenth century European rulers and merchants were looking for another route to China by going north. By boat and on land, passages were discovered, but the northwest routes, going around America, and the northeast ones, going along the coast of Siberia, were not easy to uncover in the middle of the ice. In his quest for a new route the explorer John Franklin set out from England in 1845 with 129 men and two ships, the Erebus and the Terror. Not one person returned alive. Their remains were discovered almost fifteen years later, skeletons of men walking in search of their salvation. It was the Norwegian Roald Amundsen, one of the great legends in Arctic exploration, who opened the way in 1905 after an expedition that lasted three years. After spending two winters in the Arctic, he was able to locate the surface magnetic pole and to study the Inuit way of life. The site of so many sacrifices, the Northwest Passage would remain unused for a long time.
The Finn Nils Nordenskjold opened the northeast route from Sweden to Japan in 1878–79, combining scientific agendas and economic motivations. But the use of the passages he discovered, which depends on the variable presence of ice, is difficult even today despite the shrinking sea ice.
Let’s look again at the nineteenth century and the first explorations of the sea ice, which extends into the seas of the Arctic regions and, at least in the winter, occupies a large part of them. In 1879 the New York Herald wanted to offer its readers a report on the mysteries of the North Pole: Was it a sea or land? To answer that the newspaper sent the Jeannette to the north of the Bering Strait. It was crushed by the ice in June 1881 at 77° 15´ N. Three years later, Eskimos found fragments of the wreck on the southwest coast of Greenland. Carried by the sea ice, the debris had thus floated more than 5,000 kilometers over the entire Arctic Ocean, at an average speed of five kilometers per day. That journey inspired a young scientist.
On June 24, 1893, the Norwegian Fridtjof Nansen set off from the port of Bergen. He had had a boat built, the Fram (“Forward”), and he planned to explore the polar basin—not by navigating but by allowing the vessel to freeze into the ice and then floating with it to the North Pole. At the end of September he reached the edge of the pack ice at 77° 14´ N, beyond the mouth of the Lena. He thought he would get close to the pole, but the Fram drifted first toward the southeast, away from its objective. In December the direction of the drifting was reversed: the ship ended up at the same latitude as two months earlier. The crossing of the Arctic Ocean then began. One year later, the Fram had traveled approximately 50 kilometers closer to the North Pole, but it is probable that the trajectory did not go beyond 85° N. Nansen and his companion, Hjalmar Johansen, left the ship in March 1895 to reach the pole with dogs, sleds, and kayaks. They had to give up: hummocks—true sentinels of jagged ice caused by internal pressure to which the sea ice is subjected during its formation and movement—made the surface chaotic and extremely difficult to traverse. In the spring, the canals of flowing water increased, and the “moving carpet” of the sea ice did not go in the direction of the pole. On July 24, more than two years after their departure, they reached Franz Josef Land. They finally reached Spitzberg after spending another winter, surviving by killing bear and walrus like true Eskimos. On June 17, 1896, more than three years after the beginning of this exceptional adventure, everyone ended up, including the boat, at Tromsoe, in Norway. The Fram, which has sometimes been presented as the most resilient ship in the world, can still be seen today in the Oslo museum dedicated to it.
The Arctic Ocean: Vulnerable Ice
Since that time, satellite observations of markers placed on the sea ice and drifting stations occupied mainly by Russians and later by Americans have allowed scientists to describe the drifting of the Arctic ice pack. Two main currents control the ice. One is transpolar: it starts from the Bering Strait, goes along northern Russia, then goes toward northern Greenland and follows its east coast, bringing Arctic ice to the North Atlantic. The average speed of the ice is on the order of five kilometers per day; it is slower in the Arctic basin where the ice is thicker (two kilometers per day) but much faster (20 kilometers per day) in the Greenland Sea where the ice pack breaks apart. A branch of this current goes up toward northern Canada, and the Beaufort Sea feeds a circular current in the heart of the Arctic Ocean. This is where we find the oldest ice (more than ten years old) and the thickest (seven to eight meters thick). This thin layer, on an ocean thousands of meters deep, can appear quite vulnerable.
This is the case at least in the Arctic, where one of the sources of data are measurements taken by submarines equipped with sonar. Kept secret for a long time, they have recently been made available to the scientific community by the Russian and American militaries. The results of these observations are revealing: they indicate as much as a 40% decrease in the thickness of the ice. But since they are limited to the routes followed by the submarines, they do not provide a reliable estimate for the entire Arctic Ocean. However, combined with the results of models, the available data indicate a thinning that, since the end of the 1980s, might have reached one meter in the central regions of the Arctic. We expect a great deal of data to be made available in the near future from satellite methods based on radar or altimetry observations using lasers.
The fragility of the sea ice in the Arctic is also illustrated by the decrease in its surface area since 1978; we know this thanks to microwave measurements from satellites (Figure 2.2). This area has decreased notably—close to 3% every ten years—and even more quickly in the summer—more than 7% every ten years. The connection between this decrease—which accelerated abruptly in 2007, then was followed by an increase in 2008 and 2009, while maintaining levels that were greatly inferior to those at the end of the 1970s, but decreased again in 2011 to a value close to the minimal value of 2007—and climate warming requires investigation; we will return to this in part 3.
In the south, around Antarctica, the sea ice can be traversed only in the short summer season by ships transporting provisions to the bases there. Its area increases by a factor of five in August and September; between summer and winter, it goes from four to 20 million km2. The sea ice can then extend more than 2,000 kilometers from the coast. The rapid breaking up of the ice in November and December is facilitated by the ocean current that circulates around Antarctica and disperses the ice that will melt farther to the north in warmer waters. The ice is generally less than 50 centimeters thick and, in the summer, a large part of the coast is free of ice. In certain zones that are little affected by the marine currents (notably near the Ross, Bellingshausen, Amundsen, and Weddell seas), the sea ice can survive for several years and reach a thickness of two meters. Unlike what is observed in the Arctic, the surface area around Antarctica has not varied significantly over the last few decades; as for its thickness, the data are insufficient to provide any estimates.
While the Arctic sea ice melts first on its surface, creating pools, that of the south melts more rapidly from its base and on its sides. From the action of the wind and the marine currents, the ice can break up and liberate vast expanses of water, like the polynya (a stretch of open water surrounded by sea ice) of 350,000 km2 often present in the Weddell Sea and detected by satellite observations. It was in one of these zones in the Weddell Sea that the Endurance, the ship of Ernest Shackleton, became imprisoned by the ice pack on January 20, 1915. After drifting for ten months, the ship had to be abandoned; for five months the crew survived on a floe before setting out on three small boats, surviving encounters with southern storms and finally reaching Elephant Island in southern Georgia. This epic journey concluded with the rescue of the entire crew in August 1916 by a Chilean ship.
Figure 2.2. Variation in millions of square kilometers in the minimum surface area of sea ice in the Arctic Ocean between 1979 and 2009 with (straight line) the linear tendency. In 2011 the average monthly Arctic sea ice extent decreased again to a value close to the minimal value of 2007. Source: National Snow and Ice Data Center, Boulder, Colorado.
The evolution of sea ice is very dependent on the currents that circulate in shallow ocean waters. In the past few years, scientists have discovered another aspect of the interaction between ice and oceans: the formation and the presence of sea ice play a major role in ocean circulation and in the evolution of the climate. Cooling and the fact that the ice rejects a large portion of marine salt during its formation lead to denser surface waters, which sink to the bottom. They are replaced by warmer water coming from lower latitudes. This is a major feature of global ocean circulation through two zones of formation of deep water—in the north in the Norwegian Sea and in the south around the Weddell Sea. These zones thus play a key role in the “conveyer belt” that describes ocean circulation and whose behavior and intensity are important for the evolution of the climate, including that of the continents.
Greenland: An Island Inhabited for Millennia
Unlike Antarctica, Greenland has been inhabited for thousands of years. The first people to set foot on this huge, frozen, snow-covered island arrived more than 10,000 years ago from the North American continent, taking advantage of the low level of the sea. The harsh, cold climate did not prevent successive migratory flows, which led to the development of several civilizations during the last 4,500 years. The best known and the most extended was that of Thule (1000–1500 AD). Can we attribute the beginning of the Scandinavian colonization of Greenland in the tenth century to the violent character of Eric the Red? It is quite probable. Banished first from his native land of Norway then from Iceland, he continued his journey, traveling along the eastern coast of Greenland, and landed in the south, the more inhabitable part of this island. Was there no snow or ice visible to the Vikings at that time? Or did the island seem at least more favorable than Iceland, which appeared to them as an island of ice? Perhaps, like an early tourist agent, he was trying to attract Vikings
to colonize this new land. Whatever the case may have been, he managed to persuade hundreds of people to follow him. Norway annexed Greenland in the thirteenth century and Denmark in the seventeenth century. Inuit populations and the Scandinavian colonists remained concentrated in villages confined to the coasts; as for the exploration of the inland ice sheet, that did not begin until the twentieth century.