Effect of logging and thinning intensities in the species ecological grouping

Abstract

This study was conducted with data from Projeto Bom Manejo (Embrapa-CPATU/CIFOR/ITTO, Jari area, Vitória do Jari - AP. The statistical design was in random blocks. The applied treatments are combinations of three logging intensities (15%, 25% and 35%), followed by four levels of original basal area reduction (0%, 30%, 50% and 70%). There were 40 permanent sample plots (36-logged and 4-unlogged). The stand was measured in 1984, logged in 1985 and remeasured in 1986, 1988, 1990, 1994, 1996 and 2004 (trees with dbh ≥ 20 cm). The process for grouping species involved three stages: (1) clustering analysis to make the groups using the most populous species; (2) discriminant analysis to add the less populous species to the existing groups; and (3) subjective stage where species with little or no data were assigned into the existent groups. The characteristic variables of each species in the cluster analysis were: Periodic annual increment in diameter and the 95-percentile point in the DBH frequency distribution. It was concluded that both logging of commercial trees species and the application of thinnings stimulated the growth of trees of the plants already established, and different from expected, did not favor the growth of fast growing pioneer species.