Adjustments of individual-tree survival and diameter-growth equations to match whole-stand attributesThis article is part of a larger document. View the larger document here.
Individual-tree models are flexible and can perform well in predicting tree survival and diameter growth for a certain growing period. However, the resulting stand-level outputs often suffer from accumulation of errors and subsequently cannot compete with predictions from whole-stand models, especially when the projection period lengthens. Evaluated in this study were five methods for adjusting the tree-survival equation to match observed number of trees per acre and three methods for adjusting the diameter-growth function to match observed basal area per acre. The evaluation procedures were repeated for adjustments using outputs from a whole-stand model. The different methods for adjusting tree survival probability and diameter growth were found to produce similar results. The methods selected (one for survival probability and another for diameter growth) both involved direct computation of the adjustment factor. Use of observed stand attributes for adjustment resulted in improvement of tree-level predictions of the individual-tree model. On the other hand, when predicted stand densities were used for adjustment, the quality of the tree-level predictions depended on the reliability of the stand predictions.