Steep, vertical or even overhanging rock walls in high-mountain areas are known to have recession rates in the order of millimeters per year. The processes involved (frost shattering, rock fall) have highly variable time and depth scales (Figure 1).  

1 - Time and depth scales of frost weathering and rock fall

2 - Rock fall in the year 1991 at Randa Mattertal, Valais, Switzerland. The rock fall occurred in two events.

The detachment of rock components and the corresponding loss of initially exposed surface within rock walls can be (statistically) related to

• the number of high-frequency (mainly daily) freeze/thaw cycles
• the degree of saturation of wet rocks, and
• the tensile strength of fractured rocks.

3 - Monte Rosa east wall (Italy) showing strong weathering and erosion processes in a partly glacierized rock wall.

Decimeter-to-meter scale:
This corresponds to the classic "volume expansion model" which explains rock destruction primarily by the volume expansion during freezing of a closed water/ice-system in fissures and cracks. Field and laboratory experiments indicate that surface conditions (wetness, snow) have a strong influence on the quantity of rock particles released.
 

Meter-to-dekameter scale:
At lower frequencies (seasons) and greater depths, the most effective process in rock destruction appears to be the formation of ice lenses in porous rocks by continued water flux towards the freezing front in an open system of cracks, fissures and pores. Optimal fracture propagation in common lithologies thereby takes place at temperatures between about -3°C and -6°C. Sustained subzero temperatures are more important than the number of high-frequency freeze/thaw cycles. Optimal conditions for fracturing require both high temperature gradients and the availability of moisture. Indeed, an activity peak does occur in springtime, when the release of latent heat from meltwater percolation through snow into the still cold, active layer over permafrost creates these ideal conditions. In contrast temperatures in permafrost areas, ground temperatures beneath thick snow in winter and spring at permafrost-free sites are known to be close to, or even at, the freezing point, and rock destruction by ice segregation would be much less intense.

This rather new "ice segregation model" helps us to understand, why (cold) walls in the shade are often more affected than (warm) walls exposed to the sun. It also helps to explain why the maximum intensity of rock fracturing must be expected in the permafrost table, where the temperature/moisture conditions are ideal, rather than at the surface.

The influence of permafrost on frost weathering and destabilization of rock walls has so far remained a virtually untouched field of research. The most important processes concern the fracturing of rocks during freezing, the change in hydraulic conductivity and pore water pressure/circulation during freezing/thawing and the change in surface geometry by major rockfalls.

4 - Predicted change in factor safety with respect to the temperature of ice in the joint (slope angle a =70°, inclination of discontinuity b = 40° (after Davies et al. 2001)

Where accompanied by glacier shrinkage, redistribution and reorientation of stresses deep inside mountain walls must be expected to accompany changes in surface temperature conditions (for instance, permafrost formation in rock walls originally covered by temperate firn/ice, photo). The stability of large rock masses in permafrost areas with temperatures close to the freezing point (production of water) is especially low (Figure 4).

29 August 2011
© ALPECOLe 2002-2007