DEVELOPMENT OF DYNAMIC SITE INDEX CURVES FOR Pinus cooperi BLANCO

Abstract

Two fitting procedures for models expressed in generalized algebraic differences are compared to develop dynamic site index curves from stem analysis data corresponding to dominant trees of Pinus cooperi Blanco. The main advantage of the generalized algebraic difference approach (GADA) is that the base equation can be expanded according to various theories of growth (e.g., asymptote and growth rate), thus enabling more than one parameter to be site-specific, allowing the derivation of more flexible dynamic equations, and obtaining polymorphic site-index curves with multiple asymptotes. The objective is to obtain base-age invariant curves that directly estimate height and site index from any height and age. A second-order autoregressive error structure was used in the fitting process to correct the serial correlation of the longitudinal data used in this study. The resulting equation, derived from the Chapman-Richards model, is very flexible and can be utilized for any reference age without affecting the dominant height or site index predictions.