Since the middle of the past century - the end of the "Little Ice Age" - the glaciarization of the European Alps has lost about 30 to 40% in glacierized surface area and around half its original ice volume. The estimated total glacier volume in the European Alps was some 130 km3 for the mid-1970s; strongly negative mass balances have caused an additional loss of about 10 to 20% of this remaining ice volume since 1980 (Haeberli & Hoelzle 1995, Maisch et al. 1997).

Periglacial permafrost in the Alps today occupies an area which is comparable in extent to the glacierized area and must have been affected as well, but its secular evolution is much less well known. Alpine permafrost is typically several decameters to more than 100 m thick and has characteristic mean annual surface temperature between the melting point and about -3°C (Vonder Mühll & Holub 1992).

Model calculations indicate that the warmest and shallowest parts of mountain permafrost probably have now begun to react of the effects of 20th century warming by raising the permafrost base. Similar calculations taking phase changes into account (Neumann solution) illustrate the strong retarding effects of latent heat exchange with respect to the heat wave penetration at depht.

In the ablation area and especially towards the snout of valley glaciers, the lowering of ice surfaces in the course of the past century can easily exceed 100 meters. This vertical loss in valley filling induces a change in the stress field inside the confining mountain walls. On slopes protected against direct solar radiation, the lowering of glacier surfaces can enable the penetration of negative temperatures (permafrost) into and the formation of ice within the rock walls originally covered by temperate ice (cf. method).

The penetration of the freezing front into previously thawed material has the potential of intensifying rock destruction through ice formation in cracks and fissures. Such ice formation, in turn, reduces the near-surface permeability of the rock walls involved and affects hydraulic pressures inside the still open (non-frozen) fissured rocks. The general lawering of water pressures in lateral rock walls accompanying the disappearance of temperate glaciers may thus be counteracted and the rock-wall stability altered.

Glacierized and perennially frozen mountain areas would be among the most heavily affected parts of the world in the event of accelerated future warming.

With continued or even accelerated atmospheric warming, large parts of Alpine glaciers could disappear within decades and extended permafrost slopes could start thawing (Hoelzle & Haeberli 1995) - first from the permafrost table downwards but later and for extended time periods also from the permafrost base (the interior of mountain slopes) upwards. Such a development would be without historical and perhaps even holocene precedence.

The evolution of permafrost may nevertheless cause unfavorable changes in hydraulic conductivity (onset of convective heat transfer in opening fissures at depht) and escapes the possibilities of simple/economic direct observation. Colder and thicker permafrost bodies are likely to react in the future with considerable delay and the adjustment to new equilibrium conditions may take centuries if not millennia.

Climate change may introduce highly complex feedback mechanisms involving surface geometry, firn accumulation, en-/subglacial temperatures and stress distribution. Long-term monitoring of ice geometries using aerial photography may help to detect unfavourable developments. Independently of such difficult attemps to forecast the time of instability, the run-out distance of potential ice avalanches can be quite realistically assessed.

Glacier shrinkage and permafrost degradation induce complex problems of slope stability in bedrock as well as in non-consolidated sediments (moraines, scree). Such problems may become more acute in the future and develop beyond existing historical experience if atmospheric warming indeed continues or even accelerates.

As a consequence of intense snowmelt and/or heavy precipitation, debris flows of highly variable size may also form at marginal permafrost sites in scree of debris cones or rock glacier fronts (Dikau et al. 1996, Rebetez et al. 1997, Zimmermann & Haeberli 1992).

(4) - Remarques générales