Pre harvesting modeling of the diametric structure in stands of Pinus spp. using probability density functions

Abstract

The objective was to study the behavior of the distribution curves diamétric of plantations Pinus spp. in the municipality of Palmas - PR. Data are from a stand of 87 ha, with 18 years of age. The probability functions Normal, Log-normal, Gamma, Beta, Johnson SB, Weibull 2P, 3P Weibull and Weber were adjusted to estimate parameters using the Method of Moments, Method of maximum likelihood and least squares method. The quality of the fits of the distributions was assessed using the χ2 test, with 5% probability and the relative deviation between the observed and estimated sums of the diameters. There was overlap of the curves in some classes, it was not possible to discriminate the best distribution only observing the histograms. The analytical curves were compared using the chisquare (χ2), which revealed significant differences between the observed and estimated curves. There has been analysis of the relative deviations between the observed and estimated sums of the diameters, only for the functions that were adherent in the χ2 test. Among the functions tested, the Weibull 3P function showed the best performance and was recommended to estimate the number of trees per hectare by diameter class in the study region.