Superficial Avalanches may be defined as avalanches in which a layer of snow, more or less deep, slides off an underlying layer of snow or ice.
Before proceeding to discuss avalanches in general, and to analyse more exactly the various subdivisions of these two principal categories, it is essential to analyse the primary conditions that produce avalanches. Primary conditions may be defined as those which exist before the snow has covered a slope, originally bare of snow. These primary conditions are the contour and gradient of the slope and the nature of the surface that underlies the snow.
No avalanche has yet been observed on slopes whose gradient is less than 23 degrees, though, of course, even level slopes have often been overwhelmed by avalanches falling from steep slopes above.
Other things being equal, the stability of a snow slope depends not only on its gradient, but also on the gradient of the slopes just below and the slopes just above.
A concave slope, for instance, which has an even outflow so that there is no sudden change of gradient and so that the steeper slopes merge gradually into gentle slopes and these gentle slopes into a level outrun, is infinitely safer than a convex slope the higher portions of which are more gradual than the lower portions below. Slopes that steepen suddenly below a comparatively safe gradient should always be treated with great respect.
A slope, whose gradient would be perfectly safe if the slope petered out gradually, may be highly dangerous if the gentler slope ends suddenly in a steep slope, for the snow on the gentle slope is, so to speak, ‘in the air.’ It has lost the natural support which is afforded by a gradual concave base leading out on to the level, and there is a reasonable chance of the weight of the snow on the safer slope proving just too much to stand the strain at the point where the slope steepens. In general concave slopes are safer than convex slopes, and slopes where the gradient steadily diminishes towards the base are safer than those in which the gradient increases before the base is reached. Of course, any slope overhanging a cliff is always to be treated with very great care, as even a superficial snow slide which would be quite innocuous if the slope ended on easy safe ground may prove fatal if it carries the ski-runner to the edge of a cliff below.
The chance of surviving an avalanche depends greatly on the nature of the ground where the avalanche comes to rest. Many ski-runners have escaped unhurt after being carried down several hundred feet because they have managed to keep on the surface of the avalanche and because the avalanche has gradually spread out fan-shaped on open, gentle slopes. But an avalanche falling into the bed of a narrow V-shaped valley with steep sides is almost certain to prove fatal, for the victim of this avalanche will be buried by the snow falling above, and this snow will fill up the narrow bed of the valley and freeze solid instantaneously by pressure. Thus all narrow valleys such as the Urbachthal, or the upper Rhone valley between Gletsch and Oberwald, should only be ascended when the snow is thoroughly safe.
An analogous case is where a tributary ridge runs across a hillside. An avalanche falling down this hillside will pile itself up against the tributary ridge and a ski-runner will probably be crushed below the avalanche, squeezed in between the tributary ridge and the main slope. Often a large moraine fulfils these conditions, so that an avalanche falling from a neighbouring slope is arrested at the moraine and piled up against it.
Similarly, if you are caught by an avalanche while ascending a gully, your chance of escape is much greater if the gully widens below the point where the avalanche overwhelms you. If it contracts, the pressure of the snow forcing its way through a narrow space may crush you to death. Compare the account of the avalanche that killed Bennen quoted in Scrambles in the Alps.
The bottom of a valley is not only dangerous for reasons just stated, but also because the stream at the bottom of a valley often exercises, especially in spring, an undercutting effect on the snow slopes that end in the stream bed.