Evaluation of alternative approaches for predicting individual tree volume increment.

Abstract

The volume increment of individual trees is often inferred from a volume or taper equation and predicted or observed diameter and height increments. Prediction errors can be compounded with this type of approach because of the array of equations used and differences in their accuracy. The consequences of several alternative approaches for indirectly or directly estimating individual tree volume increment were examined using an extensive stem analysis data set of Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) in southwest Oregon. The data were used to construct new stem volume, taper, and volume increment equations, which were then used to compare predicted and observed 5-year volume increments. The results of this analysis suggest that the indirect prediction of volume increment is sensitive to both the approach used for estimating stem volume and the use of actual versus predicted diameter and height increment, especially diameter increment. In addition, using the indirect method of volume and taper equations was found to have a slightly lower level of accuracy in predicting stem volume increment than the direct method. It was found that the use of local calibration procedures could help to mitigate possible problems with the bias incurred by using predicted rather than actual diameter increment.