| Summary: | A nonlinear mixed-effect modeling approach was used to model for Turkish fir [Abies nordmanniana (Stev.) subsp. bornmülleriana (Mattf.)] stands in Sinop-Ayancık District of Turkey. About 87% of the trees were randomly selected for model development and the reminder used for model validation. Based on goodness-of-fit criteria, the model including random-effects in all parameters was the best. The selected mixed-effects model showed homogeneous residual variance and autocorrelation was reduced with the inclusion of random-effects. Measures of bias and precision indicated that estimates of random-effects improved significantly the predictive capability of the taper equation when predicting upper stem diameters. Mixed models allow calibration of the model for new locations, by predicting random coefficients if additional stem form measurements are available. Diameter measurements from various stem locations were evaluated for tree-specific calibrations by predicting random-effects parameters using an approximate Bayesian estimator. It was found that an upper stem diameter at 8.3 m above ground was best suited for calibrating tree-specific predictions of diameter outside bark. The results of this study support previous findings indicating that the use of mixed-effects modeling approach increases flexibility and efficiency of taper equations for upper stem diameter prediction.
|
|---|
