Evaluation of methods for calibrating a tree taper equation

With recent advances in laser technology, it is now more and more affordable to measure upper-stem diameters accurately. In this paper, diameter measurement at a point halfway between the tree tip and breast height was used to calibrate a taper equation to improve predictions of diameters along the tree bole. Different calibration methods were investigated, including the use of fixed- and mixed-effects models, as well as quantile regression models. Felled-tree data collected in a loblolly pine (Pinus taeda L.) plantation showed that all calibration methods yielded improved results as compared to those from the uncalibrated taper model. Furthermore, the mixed-model approach and the quantile regression method based on five quantiles performed slightly better than the other calibration methods, based on two evaluation statistics. The methods presented in this paper should be applicable to data sets of loblolly pine or of other conifer species as well.