Diameter and height increment and mortality functions for loblolly pine trees in thinned and unthinned plantations
Although there remains some controversy as to the benefits of individual tree growth and yield prediction models over stand-level models, individual tree models still have wide acceptance. A generalized thinning response variable which can account for the intensity of thinning, as well as the age of the stand at the time of thinning and the time elapsed since thinning, was applied to two existing models for loblolly pine (Pinus taeda) in cutover site-prepared plantations. A site index equation for predicting mean total height of dominant and co-dominant trees and a diameter increment model were developed to incorporate the thinning response variable. New fits of height increment and mortality functions to the available data were also completed. Separate mortality functions were fit to data from unthinned and thinned stands.
The base models for this analysis were from the individual tree growth simulation model PTAEDA2. Evaluations for predictive ability of these models were done in a reduced version of the growth simulator which was modified to accept external data. The mean total height model had improved predictive ability over the original PTAEDA2 model for this variable. The diameter increment model produced no significant improvement in simulation comparisons. Fitting the two mortality functions to the multiple observation data resulted in the reduced predictive ability of the simulator compared to the original mortality model from PTAEDA2 which was fit to data from unthinned stands only.