Markov chain models for forest successions in the Erzgebirge, Germany

Markov chain models have been developed to describe hypothetical and observed forest successions in the Erzgebirge (Ore Mountains), Germany. The first model is time-homogeneous and it is applied to the whole Erzgebirge area (altitude up to 900 m). The natural course of succession in this area is hypothetical since forest management over the last 200 years has been depriving the forest vegetation of any chance to develop in its natural way. The second model is time-inhomogeneous and it is applied to an area where secondary succession has really begun after the devastation of spruce stand due to strong SO2-pollution (especially in the 1980s) and removal of spruce from the forest. Taking possible human impact into account, the model can predict vegetation development in the long-term under particular scenarios of forest management and other impacts. Both examples demonstrate that Markov chain models represent an efficient tool to organise the existing knowledge of forest succession into a system of quantitative predictions.