Spatial structure of tree species functional traits in response to edge distance in Upper-Montane Forest in southern Brazil

• Main Author: Dreyer, Jaqueline Beatriz Brixner
• Publication Date: 2020
• Other Authors: Schlickmann, Monique Bohora, Cuchi, Tarik, Vieira, Francielle Santos, Moraes, Giselli Castilho, Higuchi, Pedro, Silva, Ana Carolina da
• Format: Article
• Language: por
• Source: Ciência Florestal (Online)
• DOI: 10.5902/1980509833586
• Download full: https://periodicos.ufsm.br/cienciaflorestal/article/view/33586
• topic: Edge effect; Functional gradient; Functional diversity; Efeito de borda; Gradiente funcional; Diversidade funcional

We aimed to evaluate the functional organization of tree species in an Upper-Montane Araucaria Forest remnant, in the municipality of Urubici, Santa Catarina state. We tested the following hypothesis: the species present a high functional heterogeneity that is spatially structured as a function of the remnant edge distance. To determine the most abundant tree species, we plotted a 20 x 100 m transect, perpendicular to the fragment edge, where all trees with diameter at breast height (dbh) ≥ 5 cm were sampled. For the most abundant species, the functional traits were characterized: wood density (DM), leaf area (AF), specific leaf area (AFE) and maximum potential height (Hmax). The community weight mean values of functional traits (CWM) and functional diversity (RaoQ) were determined. The data were analyzed by Principal Component Analysis (PCA) and by Mantel Correlogram. While PCA axis 1 summarized a functional gradient related to Hmax and AFE, axis 2 indicated a variation related to DM and AF. Moreover, the edge distance significantly influenced the leaf traits and the functional diversity. At the edge, which is the environment where the vegetation is most exposed to frost, lower functional diversity was observed, with the predominance of species with small leaves. We concluded that the studied species are not functionally homogeneous and that the functional organization is spatially structured according to the edge distance.