Predicting past and future diameter growth for trees in the northeastern United States

Abstract

Tree diameter growth models are widely used in forestry applications, often to predict tree size at a future point in time. Also, there are instances where projections of past diameters are needed. A relative diameter growth model was developed to allow prediction of both future and past growth rates. Coefficients were estimated for 15 species groups that cover most tree species in the northeastern United States. Application of the model to independent data generally showed slight underprediction of growth, although the bias was negligible. Correlated observations were accounted for via a mixed-effects modeling approach, and an error function was specified to address heterogeneous variance. The models use a minimum amount of field-collected data, thus keeping data acquisition costs low and facilitating use in many forest growth applications.

Citation

Westfall, James A. 2006. Predicting past and future diameter growth for trees in the northeastern United States. Canadian Journal of Forest Research 36:1551-1562