Growth and yield models are critically important for forest management planning. Biophysical factors such as light, temperature, soil water, and nutrient conditions are known to have major impacts on tree growth. However, it is difficult to incorporate these biophysical variables into growth and yield models due to large variation and complex nonlinear relationships between variables. In this study, artificial intelligence technology was used to develop individual-tree-based basal area (BA) and volume increment models. The models successfully account for the effects of incident solar radiation, growing degree days, and indices of soil water and nutrient availability on BA and volume increments of over 40 species at 5-year intervals. The models were developed using data from over 3000 permanent sample plots across the province of Nova Scotia, Canada. Model validation with independent field data produced model efficiencies of 0.38 and 0.60 for the predictions of BA and volume increments, respectively. The models are applicable to predict tree growth in mixed species, even- or uneven-aged forests in Nova Scotia but can easily be calibrated for other climatic and geographic regions. Artificial neural network models demonstrated better prediction accuracy than conventional regression-based approaches. Artificial intelligence techniques have considerable potential in forest growth and yield modelling.

Global climate is changing due to increasing anthropogenic emissions of greenhouse gases. Forest managers need growth and yield models that can be used to predict future forest dynamics during the transition period of present-day forests under a changing climatic regime. In this study, we developed a forest growth and yield model that can be used to predict individual-tree growth under current and projected future climatic conditions. The model was constructed by integrating historical tree growth records with predictions from an ecological process-based model using neural networks. The new model predicts basal area (BA) and volume growth for individual trees in pure or mixed species forests. For model development, tree-growth data under current climatic conditions were obtained using over 3000 permanent sample plots from the Province of Nova Scotia, Canada. Data to reflect tree growth under a changing climatic regime were projected with JABOWA-3 (an ecological process-based model). Model validation with designated data produced model efficiencies of 0.82 and 0.89 in predicting individual-tree BA and volume growth. Model efficiency is a relative index of model performance, where 1 indicates an ideal fit, while values lower than zero means the predictions are no better than the average of the observations. Overall mean prediction error (BIAS) of basal area and volume growth predictions was nominal (i.e., for BA: -0.0177 cm2 5-year-1 and volume: 0.0008 m3 5-year-1). Model variability described by root mean squared error (RMSE) in basal area prediction was 40.53 cm2 5-year-1 and 0.0393 m3 5-year-1 in volume prediction. The new modelling approach has potential to reduce uncertainties in growth and yield predictions under different climate change scenarios. This novel approach provides an avenue for forest managers to generate required information for the management of forests in transitional periods of climate change. Artificial intelligence technology has substantial potential in forest modelling.

No abstract