Case Histories of glacierpermafrost Interaction

Combined glacier-permafrost effects at the Gruben Lakes, Switzerland

In the Gruben cirque, a polythermal glacier and the relatively warm permafrost of a large, active rock glacier are juxtaposed. Long-term research in this area has accompanied and assisted successful flood prevention work from ice-marginal as well as thermokarst lakes after two major lake outbursts and debris flows in 1968 and 1970 (Haeberli et al. 2001). Figure 5 presents the site and its most critical aspects. The highest reaches of the accumulation area of Gruben glacier (1 on Fig. 5) are cold and frozen to the bedrock beneath (Haeberli 1976; Suter et al. 2001). Such cold firn areas are likely to be warming with rising air temperatures, inducing faster flow (enhanced deformation and sliding) of the uppermost part of the glacier. A recent rock fall (2 on Fig. 5) started from perennially frozen rock at about 3600 m (a.s.l.; above sea level; Noetzli et al. 2003; cf. Davies et al. 2001) and travelled along the centreline of the glacier surface to an altitude of about 3100 m a.s.l. Such mass movements across glaciers can have especially long flow paths and a major rockfall could possibly reach the area of the ice-marginal lakes and the flood protection structures. This is

Kolka Karmadon Rock Ice Slide

Fig. 4. Lake 3 at Gruben before and after the construction of an overflow channel (circle in second image). Floating of the ice dam was probably prevented by the fact that the glacier margin is frozen to sub-glacial permafrost and water cannot penetrate directly to the glacier bed. Artificial lowering of the ice-marginal lake was necessary, because the retention capacity designed earlier for proglacial Lake 1 was not sufficient for combinations of events such as heavy precipitation or a subglacial water-pocket rupture combined with an outburst of Lake 3. Photographs by Andreas Kaab, 29 July and 7 August 2003

Fig. 4. Lake 3 at Gruben before and after the construction of an overflow channel (circle in second image). Floating of the ice dam was probably prevented by the fact that the glacier margin is frozen to sub-glacial permafrost and water cannot penetrate directly to the glacier bed. Artificial lowering of the ice-marginal lake was necessary, because the retention capacity designed earlier for proglacial Lake 1 was not sufficient for combinations of events such as heavy precipitation or a subglacial water-pocket rupture combined with an outburst of Lake 3. Photographs by Andreas Kaab, 29 July and 7 August 2003

especially true for the permafrost rock wall of the Inner Rothorn (3 on Fig. 5), where intense rock fall activity provides larger blocks (metre to decametre scale), as must be expected from seasonal frost weathering and permafrost degradation (Matsuoka et al. 1998). Lowering of the glacier surface at this site introduces stress redistribution within the rock face combined with greater frost penetration and intensified rock destruction (cf. Haeberli et al. 1997). Thus, this is a location with rapidly changing stability conditions. The now-regulated so-called 'Lake 3' (4 on Fig. 5) continues to enlarge towards the ice dam of the thinning glacier, and in 2003 it reached the floatation level of the ice dam, which is cold and still frozen to the subglacial permafrost at the margin of the polythermal glacier. An artificial cut in the ice dam helped to accelerate and enhance overflow. The lake did not drain completely and must be kept under continuous

Fig. 5. Overview of the situation at Gruben where permafrost-glacier interactions determine hazards induced by climate change (numbers refer to the explanation in the text; the moraine dam at Lake 1 with the breach of earlier outbursts is partially hidden below the wing of the aeroplane). Photograph (taken from the air) by Christine Rothenbiihler October 2003

Fig. 5. Overview of the situation at Gruben where permafrost-glacier interactions determine hazards induced by climate change (numbers refer to the explanation in the text; the moraine dam at Lake 1 with the breach of earlier outbursts is partially hidden below the wing of the aeroplane). Photograph (taken from the air) by Christine Rothenbiihler October 2003

observation, especially if it grows towards the vanishing glacier tongue and the rock-fall zone. The permafrost of the rock glacier (5 on Fig. 5) was in contact with the glacier during earlier (Holocene and historical) advances. It still contains buried massive ice, which is a favourable precondition for the exponential growth of potentially dangerous thermokarst lakes (Kaab & Haeberli 2001). The moraine dam at 'Lake 1' (6 on Fig. 5) is now protected by a boulder dam, a specially designed outlet structure and concrete injections. At this location, breaching in non-consolidated material with large cavities from former subglacial permafrost took place during two outburst floods in 1968 and 1970.

Combined glacier-permafrost effects in the starting zone of the Kolka-Karmadon rock/ice avalanche, Osetian Caucasus, Russia

The large rock/ice slide of 20 September 2002 from the north-northeast wall of the summit of Dzhimarai-khokh, Kazbek massif, Northern Osetia, Russian Caucasus, killed more than 140 people and destroyed the access road through the Giseldon Valley and Genaldon gorge - a primary tourist attraction of the region (Kaab et al. 2003; Popovnin et al. 2003). The avalanche starting zone as depicted in Fig. 6 was approximately 1 km wide and located between about 4300 and 3500 m a.s.l. A rough estimation of the volume of steeply inclined metamorphic rock layers detached in the slide is some 4 million m3, with failure extending to a depth of about 40 m. A similar thickness/volume of snow, firn and glacier ice was also entrained in the slide. The primary cause of the instability must therefore have been within bedrock rather than surface ice. In view of the fact that bedrock stability in cold mountain areas can be especially low in warm or degrading permafrost (Davies et al. 2001), thermal conditions affecting ice and water within rock joints are likely to have exerted a major and detrimental influence.

The situation after the event necessitated an immediate assessment of the potential for a repetition of similar or even larger accidents from the mountain slope. This involved interpretation of photographs collected by the authorities during reconnaissance flights by helicopter together with some best guesses about the thermal condition of firn, ice and permafrost in the starting zone (Haeberli et al. 2003). Bedrock surface temperatures in the detachment zone

Fig. 6. Upper avalanche path of the Kolka-Karmadon rock/ice slide. The summit of Dzhimarai-khokh and the starting zone (S) are in the background. An active talus-derived rock glacier (R) is seen in the lower right corner. Note the steam/dust (?) cloud at the foot of the slope where the Kolka glacier (K) has been sheared off. Two flow parts (Fl, F2) of the fast travelling rock/ice/water mass can be discriminated. The Kazbek volcano would be to the left. Photograph by Igor Galushkin 25 September 2002

Fig. 6. Upper avalanche path of the Kolka-Karmadon rock/ice slide. The summit of Dzhimarai-khokh and the starting zone (S) are in the background. An active talus-derived rock glacier (R) is seen in the lower right corner. Note the steam/dust (?) cloud at the foot of the slope where the Kolka glacier (K) has been sheared off. Two flow parts (Fl, F2) of the fast travelling rock/ice/water mass can be discriminated. The Kazbek volcano would be to the left. Photograph by Igor Galushkin 25 September 2002

were estimated to be about —5 to — 10°C, i.e., conditions of cold permafrost. Before the event, the steep impermeable lower rock slope had been covered by hanging glaciers. Such hanging glaciers have two thermally different parts: (1) cold ice frozen to bedrock forms the vertical/impermeable ice cliffs where meltwater runs off immediately and ice lamellas break off as an important ablation process of these ice bodies, while percolation and refreezing of melt-water cause the existence of (2) much warmer or even temperate firn and ice below the less steep upper surfaces where snow accumulation predominates. A polythermal structure is likely to have existed in the hanging glaciers, at least in the lower parts of the wall (cf. Haeberli et al. 1997). The detachment zone at Dzhimarai-khokh had therefore probably been in a complex condition of relatively cold/thick permafrost combined with warm if not unfrozen parts with meltwater flow in very steeply inclined materials with heterogenous permeability. This situation favoured high and locally variable water pressures and was further complicated by the fact that hot springs are known to occur in this volcanic region of the Kazbek massif. The event itself removed hanging glaciers with warm firn areas, reducing the load on the slope and eliminating the main meltwater source. Moreover, the exposed bedrock is now subject to strong cooling and deep freezing. These two facts led to the conclusion that the threat of a similar or even bigger event at the same site in the immediate future could be considered minimal but that instabilities in the remaining parts of the slope could continue and should be observed accordingly. Effects of potential future atmospheric warming would also have to be taken into account over the coming decades.

Was this article helpful?

0 0

Post a comment