Press release 24.01.2006
The Finnish Forest Research Institute (Metla) conducted a study, in which models were developed for predicting probabilities of forest damage caused by various factors. The calculated damage probability is represented as the risk of damage occurring in a specific type of stand and the how the damage risk concerns different kinds of trees. The predicted probabilities can also be used to assess the impact of forest management practices on the damage risk. A forest damage risk prediction model can be integrated into different systems of forest management planning.
All damage causes monitored during the Eighth National Forest Inventory (NFI 8) conducted in 1986-1994 were used as material in the study. The significance of the damages to forest development depends on the cause, amount and level of the damage, and also the way the damage affects the tree. Damage can reduce height growth, deteriorate timber quality or eventually, kill the tree.
This research report on prediction models of forest stand and tree damage contains cause-specific prediction models for the probability of stand and tree damage incidents. The models were developed using a logistic regression analysis. First, the stand-level models predict the probability of forest damage as a function of growth site and tree-specific parameters. At the second phase, the tree-level models predict the probability of damages caused to trees, depending on the incidence of forest damage and tree characteristics. At stand level the probability can be interpreted as the proportion of the total forest area that is affected by damage. At tree-level the probability can be thought of as the proportion of trees affected by damage.
In the model, the probability of damage is divided into different classes depending on the level of damage and time of incidence. Among the most important factors influencing forest damage are the tree species and the development state of the stand or tree, expressed as dominant height, age or breast-height diameter. However, the models’ capabilities are limited in describing the immediate affects of individual forest management practices to mitigate the risks of forest damage.
The research is part of the stand development modelling in the currently ongoing research programme “Alternatives of silvicultural practices in forest management and their effects on forest production".
The aim of the programme is to produce information on the productive and economic affects of silvicultural methods to support decision-making in forestry.