Agrupamento ecológico e funcional de espécies florestais na Amazônia Sul Ocidental

Abstract

The Amazon forest is a mixture of hundreds of species, with different ages and growth rates, knowing them, both for researchers as to Amazonian populations, is a major challenge. The aim of this study was to perform species ecological and functional grouping of Acre state forests. This work was developed from data of 95 permanent plots, with one hectare each (100 x 100m), sub-divided into plots of 10 x 10 and installed on Projeto de Colonização (PC) Pedro Peixoto, State Forest of Antimary (FEA) and ST Management. In these sub-plots all individuals with diameter at breast height (DBH) greater than or equal to 10 cm were measured. The collect period was from 1996 to 2007, with measurements at non-regular intervals. The clustering process occurred in three stages: (1) cluster analysis to group the most populous species. For this process, the variables employed were Periodic Annual Increment in diameter - IPADAP Medium in High, Medium and Low competition and 95% percentile of cumulative frequency distribution of diameters using Ward hierarchical method; (2) discriminant analysis, to allocate the least populated species to the groups formed, using variables of the 95% percentile of the cumulative frequency distribution of diameters and the average annual periodic increments in diameter (IPADAP) and the Fisher method, (3) subjective stage, used to allocate species with few individuals to formed groups. Ward method used in cluster analysis and the Fisher method used for discriminant proved to be effective in creating and training groups. The use of these methods resulted in formation of ten groups of species with high growth rates of similarity between individuals of the same group and low similarity between groups. Variables: IPADAP for stand and IPADAP in low, medium and high competition and P95 of cumulative frequency distribution of diameters shown are variables useful for the purposes of grouping of species. Variations in the growth rates for the overall mean data were reduced when calculated for each group of species.