A calibrateable site index model for Pinus kesiya plantations in southeastern Africa

Abstract

The aim of the study was to develop a site index model for Pinus kesiya Royle ex Gordon plantations in southeastern Africa based on the relationship between the dominant height and stand age. Conversely, analysis of dominant height and age data showed that the growth patterns of plantations were different. In addition, the asymptotes and forms of standwise dominant height curves varied within plantations. In developing a common site index model, instead of using the more common approach of estimating separate dominant height–age models for different plantations or sites, a mean curve approach based on a linear random parameter model with fixed and random parameters was applied. The random parameter model of this study was calibrated by predicting random parameters for the plantation and stand effects, in accordance with the standard linear prediction theory. The analyses showed that the calibration of the dominant height model was an efficient method to obtain reliable dominant height predictions of a stand, particularly when several dominant height–age observations from different stands of a plantation and at least one measured dominant height and stand age of a target stand are available. This is the case in many forest inventories based on temporary samples, i.e., cross-sectional data. The new site index model is a useful tool for use in different mensurational applications, and its properties can efficiently be utilized for example in forest inventories of P. kesiya plantations in southeastern Africa.