Using Nonlinear Mixed Model Technique to Determine the Optimal Tree Height Prediction Model for Black Spruce

  •  Shongming Huang    
  •  Shawn X. Meng    
  •  Yuqing Yang    


Based on the nonlinear mixed model technique, four base height-diameter models were evaluated for black spruce. The Chapman-Richard model was chosen. Top height and basal area were incorporated into the base model. Comparison of the base and expanded models showed that, although the goodness-of-fit measures on the modelling data were improved with the inclusion of top height and basal area, the predictive accuracy of the expanded models at the subject-specific level where the predominant interest of nonlinear mixed models lies, was reduced when tested on the model validation data. This has important practical implications because more accurate individual tree height predictions can be better achieved using the base height-diameter model without requiring the addition of other variables. It also reaffirms that determining the adequacy of a model on model fitting statistics alone can be misleading. A fitted model is best judged on separate validation data.

This work is licensed under a Creative Commons Attribution 4.0 License.