Integrating the risk of wind damage into forest planning

Wind is the major abiotic risk factor in Finnish forests. Therefore, tools that help managers to assess the risk of wind damage are required. This study developed simple regression models for predicting the critical wind speed needed to uproot Scots pine, Norway spruce and birch trees at the stand edges in Finnish conditions, using the characteristics of the retained forest both downwind and upwind stands as predictors. Using information on the prevailing wind conditions in the region, the critical wind speeds were converted into probabilities of wind damage, from which a mean risk index was calculated. The mean risk index was used as an objective variable in heuristic optimisation. The results of minimizing the mean risk index were compared to other objective variables such as minimal height differences between adjacent stands. The residuals of the regression models of critical wind speeds were small, especially in Scots pine and birch. Increasing tree height of the downwind stand or area of the upwind stand (gap size) decreased the critical wind speed regardless of tree species, whereas increases in the dbh/height ratio of the downwind stand increased the critical wind speed. The shelter effect of upwind stand height was stronger in Norway spruce than in other tree species, whereas the effect of tree height of the downwind stand was larger in Scots pine and birch. Minimization of the mean risk of wind damage within forest landscapes led to smooth and non-fragmented landscape structures in terms of tree height. Incorporating even-flow constraints into the planning model led to a slight increase in the mean risk of wind damage. Of the surrogate methods for risk assessment minimization of height differences between adjacent stands performed well but not equally well as minimization of the mean risk index.