The Earth's climate undergoes fluctuations and for the past thousand years has experienced periods of warming and cooling. In the seventeenth century, severe and long-standing winters, known as the Little Ice Age, recurred in the south of Europe. One can see frozen channels and snow-covered Holland in the paintings of famous Flemish artists.

Little Ice Age

The cold epoch ended in the middle of the nineteenth century and the climate began to warm in the last century and a half, quicker in the first half especially in the Polar regions. In the decade starting 1920, average annual temperatures in the Arctic had risen by 2° to 4°C, along with frequent cyclonic activity in the northern latitudes. Glaciers retreated in mountains almost everywhere: By 1950, glacier area shrunk by 25 per cent in Switzerland and by 15 per cent in the Caucasus. Although mountain glaciers sporadically advanced in the 1920s and 1960s, these cold periods were short and limited in scope and, by 2000, the steady rise in temperature became global.

Global warming has caused serious concern and has been the subject of discussions amongst scientists and decision makers. The UN Intergovernmental Panel on Climate Change (IPCC) assessment reports verify that surface air temperatures are rising. Undoubtedly, this warming is one of the factors for faster melting of mountain glaciers and ice sheets, with the exception of the Antarctic ice sheet that stores about 90 per cent of global ice. And studies of the Antarctic ice sheet over the past 50 years show that, not withstanding measurement errors, the ice mass did not diminish, which demonstrates its stability. However, the slow rise in sea levels caused by thermal expansion of oceanic water and by the melting of the Greenland ice sheet and other polar and mountain glaciers is of great importance to coastal populations.

Against this background of climate change, I will try to explain how glaciers become unstable, or surge into surrounding areas threatening human populations, and how to monitor against such surges.

Special Class of
Surging Glaciers

On occasion, individual glaciers advance quickly against a general shortening of glaciation -- the term for a glacier's growth and maturity. For instance, in 1963, the glacier Medvezhiy, "awoke" and surged down the western slope of the Academia Nauk mountain range, the highest of the Pamirs in Central Asia. The glacier's usual velocity had been 200-400 metres per year, no more than 1 metre per day. But in April 1963, its velocity suddenly shot up by a hundredfold, and it surged down the valley at 100 metres a day. Within a month, Medvezhiy's tongue measured almost 2 kilometres (km) as it split a valley into two, forming an 8-metre-deep lake. The lake's pressure eventually fractured the ice coffer dam and soon water flooded the Vanch River at 1000 cubic metres per second, ferrying large chunks of ice and stone. (Fig. 2)

Medvezhiy's surge lent impetus to a study of surging glaciers in the former Soviet Union. Starting in 1963, a special expedition regularly explored the glacier, recording thorough observations which formed the basis for predicting the next surge. Following several surveys of Medvezhiy, it was possible to predict the time and scale of the next surge, which happened in the summer of 1973. It was the first-ever scientific prediction of a glacial calamity. A few months later, in the spring of 1973, Medvezhiy again began to advance. Within two months its tongue lengthened 1.8 km and lapped up the remains of its surge 10 years ago. The glacier again dammed the tributary in the valley, and a lake again welled up behind the glacial barrier, which broke twice, flooding the surroundings with a discharge of 1000 cubic metres per second.

Since the 1960s and 1970s, studies of surging glaciers, particularly of Medvezhiy, painted a scientific picture of the structure and nature of glaciers. It was found that a glacier repeated its surge in almost equal time intervals provided that external conditions did not change. Yet even under similar geographic conditions, different glaciers reacted differently. Medvezhiy surged every 9-17 years. According to indirect data and local sightings, the glacier advanced in 1937 and 1951, while direct observations noted surges in 1963, 1972, and 1989.

The sudden surging of glaciers is not related to climatic fluctuations, and surges can take place even at times when glaciers retreat. This is the usual behaviour of some glaciers and can not be evidence of an impending surge. A time interval between the start of a surge and its end is called a pulsation and, in fact, periodic pulsations arise due to instability within the glacier. The frictional force at the glacier's bed breaks the ice. Hundreds of surging glaciers are now known in many glacial regions, with the largest numbers in Alaska, Iceland, the Spitsbergen -- the largest island of the Svalbard archipelago in the Arctic Ocean -- as well as in the mountains of Central Asia and in the Pamirs.

A glacier's pulsation consists of two main stages: a surge and a regeneration. During a surge, a glacier releases accumulated tension from its preceding stage of regeneration. The glacier splits, its velocity shoots up, and ice mass from the top of the glacier is displaced into its middle and lower zones. As this happens, the glacier's upper and lower zones come closer, forcing the glacier's tongue to lengthen. At the end of a surge, the regeneration begins when ice accumulates in the upper pulsating zone, while the front end of the activated area gradually moves.

But there is an unsolved question: what is the difference between a "normal" and a surging glacier, and can a normal glacier turn into a surging one? In any case, as mentioned earlier, a glacier's activation does not always result in a surge. But the glacier's movement does disrupt the structure and regime of the glacier as a whole: as ice accelerates, cracks appear in it that are filled with stone and new morainic deposits.

Causes and mechanisms of glacier surges

Generally, a glacier surges when narrow mountain valleys or morainic cover holds back the discharge of ice, creating instability. Unfortunately, direct observations of a change in the movement of a glacier at the onset of a surge are still very rare, and the causes for surges are not yet clear. At present, some hypotheses are being proposed in an attempt to explain the mechanism of surges.

A glacier's velocity increases abruptly either due to intensification of forces that drive it (mainly gravity) or a more problematic cause is the weakening friction inside the glacier and on the mountain bed. Both these processes are connected. The flow of ice increases suddenly after overcoming a threshold load, which happens as a result of accumulation of ice over the years in the glacier's reservoir.
One of the factors that increases the glacier's slide along the bed is the formation of a film of liquid water. Small obstacles (several centimetres in size) on the rock bed resist the flow of ice until it melts into a film of water that equals the volume of the obstacles. Glaciers also surge when a layer of snow, say 25 to 50 metres thick, accumulates over time on the surface, exceeding the critical mass of a glacier and causing it to slide on a layer of water. The thicker the glacier, the more difficult it is for the cold flux to penetrate into the lower zones, while the heat flux from the Earth remains unchanged. A surge also results in a decrease in its thickness, and temperatures at its bed fall below zero, which slows down its movement.

Acceleration of glaciers can also be explained by the appearance of hollow, water-filled areas, between the glacier's ice and the bed, that are under great pressure; the ice moves above the mountain ledges and along the bed's irregularities owing to the melting and repeated freezing of the melt water, as well as to plastic deformations of the ice.

Catastrophe in the Caucasus

A small glacier, Kolka, 3 km long, located on the north slope of the Kazbek mountain, one of the highest peaks in the Caucasus, belongs to the class of surging glaciers. Kolka surged in 1902 and again in 1969 when it grew to almost twice its length and a thick mud stream with ice stones gushed out. Kolka had been predicted to surge in the 2030s. But it happened much earlier.

On 20 September 2002, huge masses of ice, water and stones rushed along the valley and stopped before a ridge in the Skalisty range, forming a lake and a dam 100 m thick and 4 km long (Fig. 3). Around 110 million cubic metres of ice got stuck in the valley's narrow gorge. This mass of ice soon broke the dam and rushed down the valley, forming "waves" on slopes (Fig. 3) and depositing blocks of ice and stone 100-140 m higher than the river bed. Further along the gorge, a heavy mudflow with ice blocks traveled a distance of 12 km causing further destruction and killing 130 people.1

Kolka's unexpected behaviour opened an "eye" in the glacier cirque (a bowl-shaped depression on the side of mountains). It turned out that the glacier had "moved away" from its bed completely. Nowhere in the world had a similar event ever occurred, especially noteworthy, as Kolka was not a hanging glacier: it was located in a cirque and had a small slope on its surface of an angle of 7o to 9o. A similar event could happen only due to the accumulation of a great volume of water under the glacier, caused by anomalous melting of ice and snow in the Alpine zone of the Caucasus in the preceding four years. An abundance of water on surrounding slopes and the thickness of the glacier itself had prepared the glacier for further catastrophe, which included an avalanche of ice and rocks coupled with small volcanic and seismic events.

This catastrophe of the Kolka Glacier had demonstrated a necessity to permanently monitor and observe surging glaciers.

Monitoring Surges

Glaciers about to activate and surge show some changes to their outlines: the tongue shapes itself into a tear drop; fault lines and zones of crushed ice appear on the edges; a number of crevasses open up; the tongue thrusts over other glaciers; and mountain slopes and lakes take shape. At the end of a surge event, the tip of the glacier's tongue is a chaotic pile of ice blocks.

Repeated satellite photos and aerial and ground observations are especially valuable to discover surging glaciers and study their behaviour. I first gained experience in the Pamirs where many glaciers surged between 1972 and 1977. Using satellite photos and orbital stations, more than 20 large surges were discovered in the Central Pamirs between the 1960s and the 1990s.

An inventory of surging glaciers in the Pamirs was published 10 years ago.2 Cataloguing this large region had been possible due only to Russian satellite surveys of the Pamirs conducted between 1972 and 1991. In total, 630 glaciers with signs of instability were discovered in the Pamirs, including 51 with fixed large surges, 215 with signs of dynamic activity, and 464 with signs of past surges or unstable activity.

I propose the establishment of a system of regular monitoring of changes to sizes and forms of surging glaciers and of their dynamic behaviour through ground, aerial and space observations.3 Ground observations include setting up regular photogeodetic measurements of glacier fluctuations, permanent glacial-meteorological stations, and conducting field studies. Aerial observations include regular aero-visual monitoring and periodic distant aerophotogrammetric surveys. Space observations include continual satellite photography, with a resolution of 15 m to 20 m. Currently, such information is available from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), a cooperative effort between NASA and Japan, flying on the Terra satellite; from Landsat 7; as well as from the Russian section of the International Space Station, where cosmonauts photograph glaciers.

During global warming, solutions to surging glaciers and their unpredictable behaviour are still far from being found and demands an organized national and international research.

It should be emphasized that the problem of climate change is extremely difficult to understand, and it has still not been possible to know what factors in the past decades -- natural or anthropogenic -- have caused the warming. There are still many uncertainties in solving this problem. IPCC estimates are rather wide in their range of accuracy and, therefore, cannot predict with confidence the emergence of an ice age on Earth -- at least not in the coming decades and centuries.

Notes 1 Kotlyakov V.M., Rototaeva O.V., Desinov L.V., Osokin N.I. Causes and effects of a catastrophic surge of Kolka Glacier in the Central Caucasus/Zeitschrift für Gletscherkunde und Glazialgrologie, 2004, Bd. 38, Ht. 2, s. 117-128.
2 Osipova G.B., Tsvetkov D.G., Rudak M.S. Inventory of the Pamirs surging glaciers/Data of Glaciological Studies, 1998, issue 85, pp. 3-136 [in Russian].
3 Kotlyakov V.M., Osipova G.V., Tsvetkov D.G. Monitoring surging glaciers of the Pamirs, Centrla Asia from space/Annals of Glaciology, 2008, vol. 48, pp. 125-133.