Fitting Nonlinear Mixed Models with the NLME Procedure in S-Plus

Height-diameter models are used in forest inventories and in stand projection systems mainly because tree height measurements are expensive and time-consuming. An individual tree height-diameter model was developed for dahurian larch (Larix gmelinii. Rupr.) in northeastern China based on Chapman-Richards growth model using nonlinear mixed-effects modeling approach. Nonlinear mixed-effects models involve both fixed effects and random effects. The processes of model building include which parameters should be considered to be random and which should be purely fixed, as well as procedures for modeling the dependence of parameters on individual-specific covariates. Information criterion statistics are used to compare different structures of the random effects components. The Chapman-Richards model with two random parameters showed the best performance.