Simulation Of Height Curve Model For Larix Gmelinii Based On Mixed Effects

The present study focuses on the Larix gmelinii Rupr. in Dailing regions in Heilongjiang province, with variables of forest factors, such as density and site conditions. Based on data of research on41standard sample area and the selected ten basic equations of curve model on tree height, SAS software was employed to fit the selected ten basic models. By analyzing and comparing the fitting results, Richards model was chosen to be the best growth model of heights of Larix gmelinii Rupr.Height-diameter model with stand variables was developed based on sample data from Richards’s model and its fixed and random parameters. A mixed-effect modeling approach was applied in fitting Richards’s model and the same model with quadratic mean diametery Results showed that Richards model with inclusion of random parameters b1、b2was the best when fitting mixed-effects model of Richards; Inclusion of random parameters b2、b4was the best when fitting Richards model with quadratic mean diameter. Model validation indicated that mixed model showed lower error than fixed-effects model when randomly selected independent samples. However, using calibration with randomly selected four trees, mixed model reduced the bias and RMSE of the prediction by almost71.8%and42.1%, respectively.