A mixed-effects model for the dbh-height relationship of shortleaf pine (Pinus echinata Mill.)

Individual tree measurements were available from over 200 permanent plots established during 1985-1987 and later remeasured in naturally regenerated even-aged stands of shortleaf pine (Pinus echinata Mill.) in western Arkansas and eastern Oklahoma. The objective of this study was to model shortleaf pine growth in natural stands for the region. As a major component of the shortleaf modeling effort, an individual tree-level dbh-total height model was developed in which plot-specific random parameters were fitted using maximum-likelihood methods. The model predicts tree height on the basis of dbh and dominant stand height (which could be obtained from a site-index model). The mixed-effects model approach was found to predict the total height better than the similar models developed previously for this species using ordinary least-squares methods. Moreover, such a model has the appeal of generalization of the results over a region from which the plots were sampled; and also of calibration of parameters for newly sampled stands with minimal measurements.