Evaluating the performance of semi-distance-independent competition indices in predicting the basal area growth of individual trees

Abstract

Recent individual-tree growth models use either distance-dependent or distance-independent competition measures to predict tree increment. However, both measures have deficiencies: the latter because the effects of local variation in spacing are not represented, and the former because they cannot be calculated from normal inventory data for lack of spatial information. To overcome these shortcomings, the new class of semi-distance-independent competition indices was proposed. A semi-distance-independent competition index is a distance-independent competition measure that uses only the trees of a single small sample plot that includes the subject tree. Moreover, a semi-distance-independent competition index can be calculated in an analogous way to a distance-dependent competition index by using sample plot size, tree attributes, and intertree distances. However, many semi-distance-independent competition measures are based on simple tree attributes. Therefore, the objective of this study was to analyze if the semi-distance-independent competition indices explain the variation in measurements of tree increment more or less effectively than a set of classical distance-dependent competition indices. The results show that some of the semi-distance-independent competition indices explain at least as much variation in measurements of tree increment as any of the distance-dependent competition indices.