Predicting future windthrow susceptibility coupling forest growth and wind risk models: an application to a study area in the eastern Alps

Climate change will increase the frequency and magnitude of extreme meteorological events. Therefore, also windstorm events are expected to increase the amount of damages to natural and human resources. Regarding forests, windstorms are already the first cause of timber damages in Europe. In such sense, it is crucial to predict the vulnerability of forest stands and in this direction, different stochastic and statistical models have been developed and successfully tested on real events. To improve the resistance of forests against windstorms it is also necessary to forecast the future probability of damage (i.e. 100 years scenario) taking into account the changes in the characteristics of the forest cover as species composition, density, stand height, etc. In this way, forest managers can delineate specific silviculture operations in order to obtain more wind-resistant forests in a climate change context. In this study, we simulated the changes in the forest cover due to climate change and past natural events that affected the forest structure through the use of a dynamic vegetation model (e.g. TreeMig). The outcomes of the model are directly transferred to the wind susceptibility model in order to compute the probable critical wind speeds of overturning and breakage. The proposed methodology is applied to the municipality area of Rocca Pietore (74 km² of which half covered by forest), north eastern Italian Alps. The input data for the forest model are semi-automatically generated by a LiDAR survey while for the climate data high resolution CHELSA dataset has been used. The outputs are post-processed and passed to the wind risk model ForestGALES. The outputs (derived at a fine scale) show the variation in wind vulnerability in accordance with the changes of the forest characteristics. The methodology proposed in this study produce future windthrow vulnerability maps. Consequently, forest managers could quickly identify areas that can be at high risk and test alternative scenarios to increase the stand resistance.