Mountain forests of the Alps fulfil important protection functions against avalanches and other natural hazards. The majority of the subalpine forests in the Swiss Alps are dominated by spruce (Picea abies Karst.). The protection effectivity of these forests is highly variable in space and time and depends on different circumstances such as topographical settings, predominant disturbance regimes and management history.
Structures and protection effect of mountain forests are permanently changing. After centuries of overexploitation, density of most remaining forests in the Alps were very low in the middle of the 19^th century, leading to increasing concerns about their protection function (Kasthofer 1822, Landolt 1960). Since then, there is a high effort to increase and permanently maintain the protection function of mountain forests. Consequently, forest-cover and forest density drastically increased during the 20^th century, generally leading to improved protection against natural hazards, but often also to a dominance of more or less even aged (~100-150 years old) protection forests characterised by high densities and monotonous structures with very low regeneration. The successful reforestation efforts and regeneration problems in even-aged forest stands supported the notion that "only maintained protection forests provide enough and sustainable protection against natural hazards". There is, however, only little knowledge about natural dynamics of subalpine, spruce dominated forests in the Swiss Alps. The goal of a new international research project is thus to determine and analyse the relevant processes occurring in subalpine spruce dominated forests without human impact.
In this paper we give a short overview on the current state of knowledge on dynamic processes in unmanaged subalpine avalanche protection forests in the Alps and we present and discuss methodical approaches to explain the natural development of unmanaged forests in the subalpine altitudinal belt with the help of newly available inventory data series and tree-ring data.