Modelagem do crescimento e da produção de Pinus taeda L.

Abstract

The present dissertation aimed to adjust mathematical models from complete stem analysis data to obtain the equations that best represents the growth and future production of a Pinus taeda L. plantation of a company located in the Southern region of Parana State, and with this assist on the strategic planning of their forest management. The database is composed by a Pinus taeda plantation with initial spacing of 3.0 x 3.0 m, 2.5 x 3.0 m and 2.5 x 2.5 m, established between the years 1992 to 2005. To achieve the aim, it was used as initial parameters data from a recent forest inventory conducted in 4594.30 hectares, grouped in site classes by Schumacher model and the guide curve method. At each site we conducted stratification in diameter classes by Sturges method for the selection of the 144 trees to be sampled for the complete stem analysis technique. The disks were removed from the base and at 1.30 m from the ground, and in relative sections with 15%, 25%, 35%, 45%, 55%, 65%, 75%, 85% and 95% of its total height, totaling 11 disks per tree, measured in the LINTAB equipment, which helped achieve these measurements quickly and with high degree of accuracy, and after that processed using Florexel software. Two models were adjusted to obtain the diameters with bark at ages prior to sampling. The information on the actual amount of trees per hectare and harvested by thinning along the years were obtained from previously forests inventories made by the company. Finally, these data were used to carry out the modeling of growth and yield by adjusting some of the mathematical models available in the literature for stand-level modeling. The adjustment of these models was performed by evaluating the Adjusted Coefficient of Determination and the Estimate Standard Error, beyond the individual distribution of residual graphs. The Clutter model (1963) had the best resulting projections of basal area and volume for the studied plantation. The results showed that the complete stem analysis may be considered as a suitable tool to modeling the growth and yield.