Kenneth Hewitt
Cold Regions Research Centre, Wilfrid Laurier University, 75 University
Avenue West, Waterloo, Ontario, Canada N2L 3C5
Between 1994 and 1996 catastrophic movement of the 16-km-long Chiring Glacier transferred 1-1.5 km3 of ice from its upper two thirds to its lower third, and into the main Panmah Glacier of which it is a tributary. By October 1996, a lobe of Chiring ice some 3.2 km2 in area had entered and compressed the main glacier, which was severely disturbed for 3 km above and 5 km below the junction of the glaciers (Figure 1). Ice streams and medial moraines were pushed into a series of looped or "tear-drop" forms, well-known in surging glaciers. Despite an observational record back to 1856, it was not previously recognized that this glacier surges.
Surges are relatively short-lived episodes involving a sudden increase in ice movement by at least one order of magnitude, sometimes two orders, compared to presurge, and postsurge behavior [Meier and Post, 1969; Kamb et al, 1985]. This is achieved mainly by rapid sliding at the bed. One or more pulses of sharply accelerating flow also move down the glacier accompanied by a rise and severe crevassing of the ice surface. Surge events may last from a few months to several years [Dowdeswell et al, 1991]. While some advance the glacier terminus several kilometers in two or three months (see Eos, July 20 and November 9, 1993), others dissipate before reaching the terminus, especially those of tributary glaciers such as Chiring, but may lead to glacier advance later.
Surges can be serious hazards in populated areas, engulfing occupied land, generating sudden floods, or disrupting local communications below and across glaciers. They tend to recur in cycles peculiar to each glacier involved and out of phase with general patterns of glacier advance and retreat. In regions with many surging glaciers, of which the Karakoram Himalaya is one, surges complicate the normally rather sensitive relations between glaciers and climate.
In the last 100 years, 26 sudden, rapid advances have been reported involving 17 glaciers [Table 1, Appendix One; Hewitt, 1969]. At least 12 other glaciers have features associated with surge behavior (Figure 2). Only the Alaska-Yukon ranges and the Arctic islands of Svalburd have a greater number of reported surging glaciers. Even so, Karakoram surges are certainly underreported, especially those of tributary glaciers like Chiring, and we have few previous observations of surging behavior beyond the rapid advance of the terminus.
The surge developed in the main ice stream and valley of the Chiring. By 1996, most of its upper area had collapsed 50B100 m below presurge levels, was heavily crevassed, and had impenetrable areas of seracs. Tributary glaciers were sheared off and, in turn, heavily crevassed as they collapsed toward the postsurge level of the main Chiring. The head of the surge was marked by large, arcuate crevasses concave downstream across the 4-km-wide basin where the uppermost tributaries converge. The highest crevasses were at about 5,450 m above sea level, and almost 15 km from the main glacier. The total distance over which surging occurred, or "surging length" was about 17 km, with an average fall of 86 m per km.
In its lower 5 km, the glacier was much thicker than in 1992, burying the former outwash plainC and our October 1992 base camp siteCunder 110B150 m of ice! Here were deep transverse crevasses across the central regions, 20B40 m apart, giving way in narrow lateral zones to crevasses angled sharply down glacier. However, by late 1996, the active ice was well below the full height of the surge.
Nine km along both flanks of the lower Chiring, huge masses of dead ice recorded the higher levels of the surge's passage. In places, the surge "trim-lines" rose 150 m above the postsurge active ice (Figure 3). Their height varied greatly and recorded thickening at constrictions and bends in the valley. This profile differed considerably from the surface slope of the glacier before and after the surge, which mainly reflects valley floor slope. Along the left/south flank the dead ice was plastered against the rockwalls, filled chutes, and embayments. Along the right/north flank it covered the lateral moraine ridges.
Particularly striking was a smooth wall of dirt-veneered ice, 15B25 m high and capping the highest morainic ridge at Skinmang (see Figure 1). It extended, almost unbroken, for 3.5 km. On closer inspection, this smooth, dense, unfractured ice proved to contain intensely folded and microfaulted, finely foliated sections, recording great strain in the surge. These remnants represent 2B3 x 107 m3 of surge ice abandoned by the subsiding ice stream and separated from active ice by marginal shear lines typical of Karakoram glaciers. Large masses of postsurge stagnant ice often occur in over-extended terminal areas. There seem to be no other reports of the same along the lateral margins.
At the head of the Chiring is the New Mustagh Pass (5,800 m), an ancient route to central Asia. The discovery that Chiring is a surging glacier gives a new slant to an old debate about the role of glacier fluctuations in historic closings of this and other glacier passes to Inner Asia [summarized by Shipton, 1938]. Maps, drawings, and photographs from 1856, 1861, 1929 and 1937 show the lower Chiring was easily crossed by travelers [Godwin-Austen, 1864; Desio, 1929; Shipton, 1938; and Kick, 1993]. Although altitude and bad weather posed problems, the upper glacier also offered a relatively straightforward traverse to the pass.
However, in 1887 a British explorer, Francis Younghusband, coming from the Chinese side, found the pass closed. After crossing by another route, he attempted to ascend the Chiring but found it impassable because of "... an immense ice-slip on to the glacier and gigantic blocks of ice... tumbled about on top of one another..." [Younghusband, 1896, 205B6]. His descriptions accord with the effects of a surge and strongly suggest that the Chiring last surged in 1885B87, giving a surge cycle of about 110 years.
There is evidence that the Chiring is not the only surging tributary of Panmah. Satellite imagery and earlier expedition reports show the Drenmang, immediately upstream, had surged twice in this century. Major changes occurred in the Maedan between 1856 and 1861, suggesting that a surge took place (Godwin-Austen, op cit). This tributary of the Chiring, just before it enters the main glacier, advanced 1.5 km between 1993 and 1996 (see Figure 2).
There is a consensus that, whatever the controlling factors and exact mechanisms, the key to surging lies in conditions that promote large, episodic instability at the glacier bed. Proposed trigger mechanisms include fluctuations in thermal or hydrological conditions or in deformable subglacial sediment, acting alone or in combination [Clarke et al, 1984; Kamb, 1987; Raymond, 1987].
Nevertheless, the geography of surges is highly uneven. There are large numbers in just a few regions, while none have been recognized in most glacierized areas. This suggests there are special but varying combinations of environmental conditions that promote or suppress surging. It is in relation to these questions that the Karakoram glaciers and the kinds of evidence available for them are of broadest scientific interest (Table 2).
These glaciers lie between 3,000 and 7,500 m above sea level, much higher than the more intensively studied examples of the Alaska-Yukon ranges, Svalbard, or Iceland. They lie in subtropical latitudes similar to examples in Andean Argentina and have an extreme continental location comparable to the nearby Pamir surging glaciers. However, there is heavy snowfall and year-round avalanching at high elevations, which promotes rates of flow and throughput of ice comparable to more humid conditions and maritime glaciers [Mercer, 1975; ed. Hewitt, 1990; Hewitt et al, 1989].
As with the Chiring, the glaciers are surrounded by precipitous rock walls of enormous extent and elevation range. This relates to, perhaps, their most distinctive feature. Many Karakoram glaciers, and all of those known to surge, are predominantly or wholly avalanche fed [von Klebelsberg, 1925B6]. The highest precipitation occurs in the perennial ice climate zone between 5,000 and 7,000 m [Hewitt, 1993]. Avalanches carry this more abundant snow directly to the glaciers. Much of it accumulates at or below the regional snow and firn limits, which are at about 5,000B5,500 m. The succession of relatively warm and dirty summer avalanches and cold winter ones can result in complex thermal layering and debris-rich horizons in the ice.
Avalanche-derived ice tends to be heavily freighted with debris. This relatively dirty ice contributes to higher melting rates in the upper and middle ablation zones, while thick supraglacial debris suppresses melting in the lower ablation zones. Enormous ramps of debris develop and build outward beside and beneath the ablation zones of these avalanche-fed glaciers [Goudie et al, 1984; Hewitt, 1993]. Surging may be influenced by an unusual buildup of deformable sediment beneath these zones and/or by unstable transitions from frozen to unfrozen bed conditions.
Karakoram surges occur in a highly active tectonic zone with globally extreme rates of uplift and denudation [Searle, 1991]. The glaciers drape the highest parts of the range, where a series of steeply inclined lithospheric thrust faults occur. However, structures and rock types are complex and poorly known where blanketed by snow and ice. Most surging glaciers cross two or more major formations. No specific or distinctive relationship of surging to lithology, indicated in some other regions, has yet been found. Hot springs are widespread across the region and it has been suggested they, or the geothermal heat flow implied, could be a factor in surges.
Studies in the Yukon Territory of Canada and Svalbard found that surge-type glaciers tend to be longer, wider, and of lower gradient than normal glaciers. We lack the data for a comparable analysis for the Karakoram, but it may be worth pointing out that its known and suspected surging glaciers are of short or intermediate length for the regionC6 km to 21 kmCand relatively steep. Meanwhile, no surges are recorded for more than 30 glaciers that are longer. Among the longest, widest, and lowest gradient glaciers, Siachen (75 km long), Biafo (68 km), Batura (60 km), Chogo Lungma (47 km), and Chiantar (35 km) have exhibited normal advance and retreat over the past one or two centuries [Mercer, 1975]. As with Panmah-Chiring, main glaciers not known to surge are much longer, of gentler slope, and wider than their surging tributaries.
The orientation of the watershed and ice stream seems to be unimportant or incidental to surging in Alaska-Yukon [Clarke, 1991]. However, two thirds of Karakoram surging ice streams originate mainly or wholly on slopes with a northerly aspect and most flow in a northerly direction. The one fifth with southerly orientation include the more extreme, high-elevation watersheds with steep-walled, avalanche-fed glaciers.
It has been argued that nineteenth century "little-ice-age" behavior in the Karakoram is poorly correlated with the European Alps [Kick, 1989]. The favored explanation is that widespread, heavy debris covers on the ablation zones of Karakoram glaciers buffer them against climate change. However, while recognized, surging was not previously known to affect the very glaciers believed to be out of phase with the Alps. They include Panmah, Baltoro, Hispar, and Bualtar-Barpu. If surging had been as widespread as now appears, it would complicate the interpretation of all neoglacial and possibly Pleistocene glaciations in the region.
The Karakoram is of unusual interest and perhaps sensitive to climate change, since its glaciers lie within the variable influence of three major weather systems: the sub-Mediterranean regime of mainly winter, westerly storms; the summer monsoon; and the Tibetan anticyclone. Winter storms dominate glacier nourishment at present. However, nearly one third of the high-elevation snow accumulation we have measured occurs in summer [ed. Hewitt, 1990]. It has been argued that general patterns of advance and retreat in the region relate to changing vigor of the summer monsoon [Mayewski et al., 1980]. The possibility of such large shifts in the atmospheric sources, regime, and seasonal occurrence of glacier nourishment, does not seem to be a factor in other regions with surging glaciers. This seems to be a further reason to give more attention to surging glaciers in a relatively neglected region.
Fig. 1. The junction of the Chiring with the main
glacier in October 1996 showing the results of the surge. The view looks
southwest from a peak at 5,600 m above sea level. A lobe of Chiring ice
has compressed the main glacier and created looped or "tear-drop" forms
in its ice streams. Arrows in the left, near-ground identify "trim-lines"
of stagnant ice left by the passage of the surge (see text). Here, the
Chiring is roughly 1 km wide, the main glacier, 2.5 km wide (photo by author,
September 27, 1996).
Fig. 2. The Karakoram Himalaya and its glacier cover,
identifying known and suspected surging glaciers.
Fig. 3. View looking due west down the lower Chiring
glacier. It shows the trim-line of sheared-off ice, left by the surge along
the south/left flank of the glacier. The difference between present ice
surface and trim-line exceeds 150 m in places. The ice stream is roughly
1 km wide in this section. Disturbed ice of the main glacier is just visible
in the background, approximately 8 km away from the foreground (Photo by
author, September 29, 1996).
Date Glacier (valley) Surge Sources Location (see Fig. 2) 1860s (?) Karambar Probable surge. Hayward (1871) 1 (Gilgit-Ishkoman) Damaging floods Kreutzmann (1994) on Gilgit R. 1861 attrib. Karambar ice dam burst. 1860-61 "Maedan" Probable surge pushing Schlagentweit (1866) VII Panmah tributary aside Chiring and Godwin-Austen (1864) (Braldu) Nobande Sobande, overriding flanks and draining lakes. 1868-69 Aktash Teminus advanced Shaw (1871) 16 1,600 m in three months 1886-87 Chiring, Panmah "Immense ice-slip Younghusband (1892) 12 tributary (Braldu) on glacier and Shipton (1938) gigantic blocks of ice..." prevented crossing to Skinmang and access to 'New' Mustagh Pass late 19th Garumbar, Approx. 2.5 km Mason (1931) 7 century Hispar tributary advance. late 1800s (?) Sumaiyar Advanced to join Conway (1894) V Bur-Burpu Bualtar. (Hispar-Nagyr) 1890-92 Pumarikish tributary Overriding lateral Conway (1894) 8 (Hispar) maraines and pushing main glacier aside. 1892-93 Minapin (Hunza) 370 m sudden Mason (1930; 1935) 3 advance. 1,200 m advance in all. 1893-95 Hassanabad (Hunza) 9.7 km in 2.5 Hayden (1907) 2 months. "...2 miles Workman (1911) in present summer Neve (1907) (1895)...then stopped..." 1895-1905 Karabar (Ishkoman) Surge dammed river. Todd (1930) 1 1905 glacier lake Kreutzmann (1994) outburst and largest flood disaster on Gilgit River 1901-02 Yengutz Har "2,600 m in spring" Mason (1931) 6 (Hispar) "...2 miles...in 8 days" 1902-03 Aktash (Upper Sudden, rapid Longstaff (1910) 6 Shyok) advance. 1902-03 Chogo Lungma "...several miles..." Workman and Workman 10 tributary (Shigar- (1908) Basha) 1929-30 Bualtar (=Hopar) 550 m and further Mason (1931) 4 (Hispar-Nagyr) 150 m during summer. 1930 Karambar "...100 paces (in.) Mason (1931) 1 (Gilgit-Ishkoman) three weeks of March..." 1930 Sultan Chhussky "...enormous push Visser and Visser-Hooft 17 (Upper Shyok) forward...200-300 (1935-1938) million meters cubed of ice..." c. 1931 Drenmang/Panmah Reconstructed from Shipton (1938 and 11 tributary (Braldu) location of surge lobe unpublished) in Nobande Sobande in 1937. 1935-36 Aktash (Upper 2.5 km in 7 months Lyall-Grant and Mason 16 Shyok) (1940) 1953 Kutiah (Stak Valley) 12 km in 2 months Desio (1954) 9 March to early May (Desio et al, 1961) 1955 Karambar Surge blocked Hewitt (unpubl. notes) 1 (Gilgit-Ishkoman) valley but Karambar R. maintained tunnel under the ice. 1958 (?) Aktash Rapid advance. Mercer (1975) 16 (Upper Shyok) 1974-77 Balt Bare (Hunza) 2 km rapid advance Wang et al (1984) 5 in 1976-77, preceded by huge debris flow in 1974 1977-78 Drenmang/Panmah Rapid advance Hewitt (unpubl. from 11 tributary (Braldu) oflobe into satellite imagery) Nobande Sobande. 1986-90 Bualtar (=Hopar) Two surges, one Gardner and Hewitt 4 (Hispar-Nagyr) following rock (1990) avalanche (1986-87, Hewitt (unpubl. field and second in 1989- notes) 90, 2 km advance. 1988-89 Pumarikish/Hispar 1.5 km surge Wake and Searle (1993) 8 tributary pushing main (Hispar) glacier aside. 1989-93 Lokpar/Alling Surge of Lokpar Hewitt (unpubl. field 13 tributary tributary followed notes) (Shyok-Hushe) by steepening and 1.5 km+ advance of main terminus. Dam burst flood from ice margin lakes (1990). 1992 (?) Sumaiyar Bar/ Report of sudden, Hewitt (unpubl. field V Barpu tributary. massive thickening notes) (Hunza-Hispar) and other surge-like behavior. 1993 Karambar Surge began in April Hewitt (unpubl. field 1 (Gilgit-Ishkoman) and glacier notes advancing 7-10 m per day in June. 1993 (?) Masherbrum Gl. Thickening and Hewitt (unpubl. field XI* (Hushe) surge-like behavior notes) of upper glacier. 1994-96 Chiring/Panmah Surge advanced Hewitt (unpubl. field 12 tributary 2.5 km, pushing notes) (Braldu) aside ice of main glacier. 1990s Liligo tributary 2 km rapid advance Hewitt (unpubl. from 14 of Baltoro to reach main glacier satellite imagery) (Braldu) 1990s Moni tributary & 2 km rapid advance Hewitt (unpubl. from 15 Sarpo Laggo across and down satellite imagery) (Shaksgaur) main glacier.
Clarke, D. K. C., S. G. Collins, and D. E. Thompson, Flow, thermal structure, and subglacial conditions of a surge-type glacier, Can. J. Earth Sci., 21, 232B40, 1984.
Clarke, G. K. C., Length, width and slope influences on glacier surging, J. Glaciol., 37, 236B46, 1991.
Conway, W. M., Climbing and Exploration in the Karakoram Himalaya, 3 vols, Unwin, London, 1894.
Desio, A., and A. di Savoia-Aosta, La Spedizione Geografica Italiana al Karakoram. Bertarelli, Milan, 1929.
Dowdeswell, J. A., G. S. Hamilton, and J. O. Hagen, The duration of the active phase on surge-type glaciers: contrasts between Svalbard and other regions, J. Glaciol., 37, 388B400, 1991.
Godwin-Austen, H. H., The glaciers of the Muztagh Range. Proceedings of the Royal Geographic Society, 34, 19-56, 1864.
Goudie, A. S., D. K. C. Jones, and D. Brunsden, Recent fluctuations in some glaciers of the Western Karakoram mountains, Hunza, Pakistan, in The International Karakoram Project, 2, 411B55 edited by K.J. Miller, 1984.
Hewitt, K., Altitudinal organisation of Karakloram geomorphic processes and depositional environments, chapter 7, in Himalaya to the sea: geology, geomorphology and the Quaternary, edited by J. F. Shroder, Jr., Routledge, London, 159B83, 1993.
Hewitt, K., Glacier surges in the Karakoram Himalaya (Central Asia), Canadian J. Earth Sci. 6, 1,009B1,018, 1969.
Hewitt, K., C. P. Wake, G. J. Young, and C. David, Hydrological investigations at Biafo Glacier, Karakoram Himalaya: an important source of water for the Indus River. Ann. Glaciol., 13, 103B108, 1989.
Hewitt, K. (Ed.), Overall report: snow and ice hydrology project, Upper Indus Basin SIHP, Cold Regions Research Centre, Wilfrid Laurier University, 179 p, 1990.
Kamb, W. B. et al, Glacier surge mechanism: the 1982B83 surge of Variegated Glacier, Alaska, Science, 227, 469B79, 1985.
Kamb, W. B., Glacier surge mechanism based on linked cavity configuration of the basal water conduit system, J. Geophys. Res., 92(B9), 9,083B9,100, 1987.
Kick, W., Adolf Schlagintweits Karakorum-Forschungsreise 1856 Forschungsberrichte des Deutschen Alpenvereins, Bd. 6, Munchen, 1993.
Kick, W., Exceptional glacier advances in the Karakorum. J. Glaciol., 3(23), 229, 1958.
Kick, W., The decline of the last Little Ice Age in High Asia compared with that of the Alps, in Glacier Fluctuations and Climatic Change, 129B42, edited by J. Oerlemans, Kluwer, Dordrecht, 1989.
Mason, K., Expedition notes: tours of the Gilgit Agency, Himalayan J., 3, 110B115, 1931.
Mayewski, P., G. P. Pergeant, P. A. Jeschke, and N. Ahemad, Himalayan and transhimalayan glacier fluctuations and the S-Asian Monsoon record, Arctic and Alpine Research, 12/1, 171B82, 1980.
Meier, M., and A. S. Post, What are glacier surges? Can. J. Earth Sci., 6, 807B18, 1969.
Menzies, J., The dynamics of ice flow, chapter 5, in Modern Glacial Environments: Processes, Dynamics and Sediments, 139B96, edited by J. Menzies, Butterworth-Heineman, Oxford, 1995.
Mercer, J. H., Glaciers of the Karakoram, in Mountain Glaciers of the Northern Hemisphere 1, Chapter 3, 371B409, edited by O. W. Field, 1975.
Raymond, C. F., How do glaciers surge? A review. J. Geophys. Res., 92(B9), 9,121B34, 1987.
Schlagintweit, H. A. and R., Results of a Scientific Mission to India and High Asia, Undertaken Between the Years 1854 and 1858, 4 vols, Trubner, London, 1861B66.
Searle, M. P., Geology and Tectonics of the Karakoram Mountains, John Wiley, New York, 1991.
Sharp, M. J., Surging glaciers: behaviour and mechanisms, Progressive Physical Geography, 12, 349B70, 1988.
Shaw, R. B., High Tartary, Yarkand and Kashgar, a Return Journey Over the Karakoram Pass. Constable, London, 1871.
Shipton, E. E., Blank of the Map, Hodder and Stoughton, London, 1938.
Visser Ph.C., and J. Visser-Hooft (Eds.), Karakoram: Wissenschaftliche Ergebnisse der Neiderlandischen Expeditionen...etc. 3 volumes, Leiden, 1935B38.
von Klebelsberg, Der Turkestanische Gletschertypus, Zeitschrift für Gletscherkunde, 14, 193B209, 1925B26.
Wake, C.P., and M. P. Searle, Rapid advance of the Pumarikish Glacier, Karakoram Himalaya. Correspondence, J. Glaciol., 39, 1993.
Younghusband, F. E., The Heart of a Continent, John Murray, London, 1896.