Sean M. Garber

Modeling stem taper of three central Oregon species using nonlinear mixed effects models and autoregressive error structures

Variable exponent taper models were developed for three species in the central Oregon Cascades using nonlinear mixed effects models. Introduction of two random effects partially reduced autocorrelation, but were not sufficient to completely eliminate it. A first-order continuous autoregressive (CAR(1)) error process was incorporated into the modeling to provide valid tests of significance on model parameter estimates. The variable exponent model captured variation in the stem form across a wide range of spacings and several species mixes. Functions of diameter: height ratio were important variables for distinguishing among tree forms in these plots of varying structure. Volumes estimated from the equations developed for Pinus ponderosa Dougl. ex Laws. in this study were similar to volumes estimated from previously published equations developed for the east slopes of the Washington and Oregon Cascades. However, estimates differed for both Abies grandis (Dougl. ex D. Don) Lindl. and Pinus contorta Dougl. ex Loud. underscoring the importance of developing site specific volume and taper equations when assessing tree responses to silvicultural treatments.