Desempenho de clones de teca em sistema agrossilvipastoril

Abstract

The inclusion of the tree component to the crop-livestock integration activities represents an alternative to incorporate the production of wood to the agricultural enterprise. The choice of species, the genetic material and the type of management are important points to consider for the success of the system. The objective of this study was to evaluate the performance of three teak clones (Tectona grandis Linn f.). In an agrosilvipastoral system, in the municipality of Araputanga, southwestern region of Mato Grosso state of Brazil. The study carried out in an agrosilvipastoril system installed in the year 2009, in equidistant 25-meter single lines, conducted in consortium with maize and soybean cultivation between the lines in the first years and cultivation of pasture grasses in subsequent years. In the seventh year, 47 permanent plots of 25 trees each allocated by stratified random sampling. Strict tree densities obtained to obtain volume, and the removal of the trees' disks along the trunk for later trunk analysis. The percentage of survival, the mean values of DAP, transversal area and total height, besides the basal area and the total volume, were calculated. The dominant height determined by the Assmann method was also determined. Eight growth models adjusted to estimate dendrometric variables as a function of age. For the teak clones in the agrosilvipastoril system it was verified: the performances were superior in the sites with higher levels of Ca and Mg; there is intraspecific competition, represented by the growth in asymmetric diameter, with larger dimensions in the interline compared to the planting line; the growth in diameter is asymmetrical, with larger dimensions in the interline compared to the planting line; occurs the eccentricity of the marrow in the trees and may form traction wood; the technique of trunk analysis is efficient to obtain data of growth not registered in previous years; the Takata model was the most adequate to estimate the volume; the ChapmanRichards model was the most adequate to describe the variation of the cross-sectional area and the Mitscherlich model for the mean volume; there is no reduction in the current and average increments in cross-sectional area and average volume, which indicates that there is no need for thinning.