Predicting the effect of thinning on growth of dense balsam fir stands using a process-based tree growth model

A tree-level process-based model of forest growth is used to investigate the effects of thinning on the growth of balsam fir (Abies balsamea (L.) Mill.) in stands that have almost reached commercial maturity but that have never been thinned. The model is applied to predict the 20-year growth of a stand following a recently established thinning experiment in which four thinning treatments were tested. The combination of stand properties and treatment type is quite particular and the resulting long-term effect on growth cannot be evaluated based on past experiments. The objectives of the study are to provide estimates of treatment outcome and of their errors over the appropriate time frame for decision making. This is achieved by representing growth processes through functions empirically adjusted to field observations while limiting the inputs of the model to what are usually available through regular forest inventory. Simulations suggest that 20-year growth of individual trees from the smaller diameter classes is improved by the treatments, but the growth of larger trees (>0.1 m(3)) is left unchanged. When the model error is not taken into account, the results after 20 years suggest, with a confidence level greater than 95%, that the merchantable volume of the treated plots does not recover to the level found in the untreated control plots, a result contrary to the initially expected effect of such thinning. By including modelling uncertainty, however, the confidence level associated with such a result is reduced to 70%. Such an inclusion prevents the misuse of the model predictions too far into the future.