Behavior of Traditional and Generic Hypsometric Models for Pinus oocarpa Plantations for Different Treatments

Abstract

The objectives of this research were: a) to test several traditional and generic hypsometric models; and b) to analyse their adjustments and performances according to age and number of thinnings. The data set came from measurements of diameter and height of 1099 trees distributed equitatively over 11 treatments with age ranging from 5 to 25 years, and number of thinnings varying from 0 (zero) to 6. Each treatment considered consists on the combination of different ages with different number of thinnings. The Pinus oocarpa stands belongs to Duraflora S/A, located in Agudos, State of São Paulo, Brazil. The selection of the best hypsometric model was based on adjusted coefficient of determination (R²adj), standard error of estimate and graphic analysis of residuals. Due to the small variation of model statistics for each treatment, a ranking of the models was performed. In a general way, the Curtis traditional model provided the best fit to the data. The models number 6, 7 and 8 had the best performance among the generic ones. The hypothesis that selective thinnings dissociat the hypsometric relationship was confirmed by the gradative reduction of the R² as the number of thinnings increased. As the stands became older, and the number of thinnings increased, the estimation of heights for these cases was obtained by using the equation hˆ = b₀, b₀, being equal to the average height.