Individual Tree-Based Diameter Growth Model of Slash Pine in Florida Using Nonlinear Mixed Modeling

摘要

Field trials have been installed in the southeast United States in an effort to better understand the best harvesting techniques for the conversion of even-aged stands into uneven-aged stands. Another approach is to use simulation modeling of forest growth and harvest, which provides projected estimates on the best techniques. The objective of this study was to develop a spatially explicit diameter growth model that can be used in a forest growth simulator for testing different harvest regimes for stand conversion of slash pine (Pinus elliottii Engelm.). We fit various linear and nonlinear models of growth and select the best model based on several fit statistics. The best model was further improved by including a random effect in a nonlinear mixedmodeling approach. Predictions from models were validated using a leave-one-out cross-validation approach. Our best model did not include stand density measures such as total basal area per ha, and after a random effect was included in the model, the term explaining the stand-level effect of site index was insignificant. The final model included effects for initial height, diameter, and the modified Hegyi index for diameter. This model will be useful for simulating the dynamics of an uneven-aged stand.