A generalized interregional nonlinear mixed-effects crown width model for Prince Rupprecht larch in northern China

Abstract

Crown width is a tree variable that is commonly used as an important predictor in forest growth and yield models that serve as decision-support tools in forestry. Here, we developed a generalized interregional nonlinear mixed-effects individual tree crown width model using data from 3369 Prince Rupprecht larch (Larix principis-rupprechtii Mayr.) trees on 116 sample plots that were distributed across the two most important regions in northern China. Because measurements from the same sample plot were highly correlated with each other, random effects at the levels of both sample plots and stands with different site conditions (blocks) were used to develop a two-level nonlinear mixed-effects crown width model. To describe the interregional variability of crown width between the regions, a dummy variable, which accounts for region-specific differences, was introduced into the model. The results showed that the two-level interregional nonlinear mixed-effects crown width model accurately described the regional variability of crown width for Prince Rupprecht larch in northern China. Modeling the random effects at the block level alone led to significantly high correlations among the residuals. However, these correlations decreased significantly when the random effects were modeled at both the block and sample plot levels. Measuring the crown width of four randomly selected Prince Rupprecht larch trees per sample plot is recommended for localizing the mixed-effects crown width model and precisely predicting the crown widths of the remaining trees on each plot.