Sensitivity of stand dynamics to grazing in mixed Pinus sylvestris and Quercus pubescens forests: A modelling study

Abstract

Only recently, studies of forest succession have started to include the effects of browsing by wild or domestic ungulates. We aim to contribute to this topic by analysing the influence of goat grazing on the long-term coexistence of Scots pine ( Pinus sylvestris L.) and pubescent oak ( Quercus pubescens Willd.) in the low-elevation forests of an inner-Alpine dry valley. The forest gap model ForClim was first adapted to these site conditions by examining the site-dependent sensitivity of the model with regard to the species-specific parameterisation of the drought tolerance as well as the light demand of establishing and adult trees. In a second step, the behaviour of the model was investigated with respect to different grazing intensities and species-specific browsing susceptibilities. The last step was the application of a grazing scenario based on forest history, with 150 years of heavy browsing (by goats) at the beginning of the simulated forest succession, followed by less intensive grazing pressure. By comparing simulated with empirical data, it was evident that the model underestimated Scots pine basal area if succession was simulated without goat grazing. Introducing grazing led to a more realistic, higher abundance of Scots pine in the model. Regeneration of the more browsing-susceptible pubescent oak was seriously suppressed during the simulated intense grazing period, which led to a higher occurrence of Scots pine in the model. Although after 400 years of simulated succession pubescent oak reached nearly the same basal area as without historic intensive grazing, succession was delayed. The model application suggests that goat grazing allows for a much longer period of codominance of Scots pine and pubescent oak in these forests. The changes in the simulated successional pathways support the view expressed in other studies that human impacts may alter forest succession substantially. Our results allow for a first qualitative estimate, and they make clear that the representation of browsing in forest succession models needs to be improved further.