Evaluating published approaches for modelling diameter at breast height from stump dimensions

Abstract

We aggregate and summarize published equation forms for predicting diameter at breast height (d.b.h.) from stump diameters, which have rarely been compared within and between regions. Equations differ by the number and type of predictors and forms of conditioning such that predictions match expectations as stump height approaches breast height. We select four equation forms published for species common to North American coolerate tolerant hardwood forests and test them, using published and localized coefficients with a large independent dataset. We compare the predictive strengths of different forms using regression-based equivalence testing and the financial ramifications of selecting one form over another. Predictions of d.b.h., volume and value are made for four harvested stands using each equation with published and localized coefficients. We find that most equations that predict d.b.h. from stump diameter do so quite accurately. However, equations that include stump height as a predictor are superior. There are significant financial differences between valuation of predictions of removals made with different equation forms and predictions made with published and localized coefficients. Further gains in predicting breast height from stump dimensions may come through application of a neiloid form and a whole-tree taper equation.