Kozak's variable-exponent taper equation regionalized for white spruce in Alberta

Kozak's (1988) variable-exponent taper equation was estimated for white spruce by ecoregion of Alberta to reflect stem form variability among different ecoregions. Inspection of fit statistics and residual plots showed that the taper equation fitted the data quite well. Since the relative height constraint p had little impact on the overall performance of the taper equation, the optimum p value was estimated as a part of the nonlinear least squares procedures. Regional differences of the taper equation were examined and tested using the nonlinear extra sum of squares method. Ecoregions of similar taper relationships were combined to provide a composite equation. Performance of the taper equation in predicting diameter inside bark, total volume, and merchantable height was evaluated. Results indicated that the biases in predictions were small, both across different parts of the stem and for various tree sizes.