Comparação de duas metodologias multivariadas no estudo de similaridade entre fragmentos de Floresta Atlântica

Abstract

Studies carried in the Atlantic Forest has pointed out to the floristic heterogeneity of this vegetation in different regions of Brazil. However, a refined comparison of the floristic relations, becomes necessary to identify the level of affinity among them. The objective of this work was to compare two multivaried methodologies in the study of similarity among fragments of the Atlantic Forest. A bibliographical survey was carried out from 11 fragments of Atlantic Forest, located in the States of Pernambuco, Rio Grande do Norte, Minas Gerais, São Paulo and Rio de Janeiro. The data set for the accomplishment of the study of the floristic similarity was construed using two methodologies of the multivaried analysis techniques. In the usual methodology it was used a binary matrix (presence/absence) of 236 arboreal species present in the 11 fragments, a cluster analysis, using the method of the simple linkage and the coefficient of Jaccard was applied. In the proposal methodology it was used the analysis of principal components for reduction of the dimension of the matrix density and absolute dominancy of the 236 arboreal species,considering the scores of these principal components to apply the cluster analysis, through a method of simple linkage and the Euclidian distance. In the usual methodology it was identified two fragments, one in the Northeastern Region(Pernambuco) and another in the Southeastern Region (Minas Gerais). In the proposal methodology it was identified only one group with fragments in the Northeastern region (Pernambuco),standing out that the quantitative variables are of utmost importance for the association of the forests in different regions. The proposal methodology has potential for use in the study of similarity of the forest fragments.