Segmented Polynomial Regression Applied to Taper Equations 论文

摘要

Abstract The shape of various segments of tree boles approximate different geometric solids. The lower bole portion is generally assumed to be a neiloid frustum, the middle a paraboloid frustum, and the upper portion a cone. To describe stem taper throughout the bole, a set of submodels can be used. When these submodels are taken to be polynomial models, the techniques of segmented polynomial regression can be applied. Data from plantation and natural stands of loblolly pine were used as examples to compare three segmented polynomial regression models and a previously proposed single quadratic model for ability to describe tree taper. Results of hypothesis tests and analyses with an independent data set showed the segmented polynomial regression models with three submodels to be the best model tested for predicting taper in loblolly pines. Forest Sci. 22:283-289.