Allocation of oaks to Kraft classes based on linear and nonlinear kernel discriminant variables

A method of discriminant variable determination was used to visualize the division of oak trees into Kraft classes. Usual discriminant variables and several types of kernel discriminant variables were studied. For this purpose the traits of oak (Quercus L.) trees, measured on standing trees, were used. These traits included height of tree, breast height diameter and crown projection area. The use of the Gaussian kernel and modified Gaussian kernel enabled the clearest division into Kraft classes. In particular, the latter method proved to be the most effective.