New models to predict of the increment in volume for mixed forests in the State of Durango, Mexico

Abstract

Thirty models were evaluated to predict the timber yield in 28 Permanent Silvicultural Research Plots (SPIS), in the Ejido Pueblo, Durango, Mexico. The most predominant plant association is the corresponding to mixed stands of Pinus and Pinus-Quercus, besides the vegetation associated to these forests. The models in their logarithmic form incorporate 10 competition indexes (IC) independent of the distance, 3 indexes of density (ID), besides the variable age and height at the average level of the stand. The new mathematical models relate in an indirect way the effect of the competition and the density in the growth of trees present in the stand. According to the variance analyses and the hypothesis tests, it was found that the best IC incorporated in a mathematical model for the prediction of the increment in volume was the competition index of area basal density (IABN), the competition index of diameter quadratic density (IDCN), the competition index of area basal total density (IABTN) and the competition index of maximum total height (IHMAX). The best models present a high level of accuracy with Prob > |t| = 0.0001 and values average of 0.99 in R2. The best models presented average values of R2 = 0.99 and a highly significant level of accuracy (Pr > |t| = 0.0001).