Entendendo a diversidade de anfíbios anuros em gradiente elevacional na Chapada Diamantina, nordeste do Brasil

Abstract

Understanding the composition and spatial distribution of mountain biotas’ has been a study subject for several working groups. It is highlighted, as one of the questions to be answered, the identification of which processes regulate the distribution of biotic communities throughout elevational gradients, a theme of great interest to various areas of knowledge. Nevertheless, in Brazil, only four studies tested the effect of elevation as a predicting variable for structuring amphibian communities, all restricted to the Atlantic Forest. The primary objective of this study was to identify which environmental variables along an altitudinal gradient in the Caatinga biome can explain variations, if there are, in the structure of anurans’ communities that inhabit lotic environments without canopy coverage. We tested the following hypotheses: (i) the physical and biological environment inhabited by the anurans is gradually altered under the effect of the gradual increase in elevation and, consequently, (ii) richness, (iii) abundance and (iii) composition of anurans’ species directly respond to the gradual variation of the physical and biological characteristics of the local environment. The study area is inserted in the Chapada Diamantina National Park and regions in the surrounding area. We have sampled 20 sample units (SUs) in two periods (of higher and lower rainfall), distributed along 360 m to 1310 m (elevational amplitude of ca. 950 m) and equally grouped in five elevational quotas: 400, 600, 800, 1000 and 1300 m. We seek (i) SUs’ structuring patterns as a function of local environmental variables through the principal component analysis (PCA) and (ii) the ordering patterns of the anurans’ community through the non-metric multidimensional scaling method (NMDS). We tested the hypothesis that the elevational gradient (independent variable) is able to explain variations in the characteristics of the environment, richness, and abundance of anurans’ species (dependent variables) through the linear regression analysis. We have hypothesized that the set of sampled environmental variables (independent variables) are capable of explaining the distribution pattern of anurans’ species (dependent variable) along environmental gradients using the multiple linear regression analysis (MLRA). We registered 613 specimens of anurans, distributed in five families, nine genera and 21 species. The PCA resulted in four more representative PCs. The NMDS (2D) solution showed the structuring of the anurans’ community and the “turnover” of species along the altitudinal gradient. The physical and biological characteristics of the environment and the structure of the anurans’ community were linearly associated with the altitudinal gradient, i.e., increase in elevation followed by an increase in abundance and decrease in species richness. The MLRA revealed that the set of local environmental variables could explain the structure of the anurans’ community along the altitudinal gradient, highlighting climate variables and marginal herbaceous/shrubby vegetation to lotic environments. It is the first study to analyze the simultaneous effect of elevation and local environmental variables on anurans’ community structure in the Caatinga biome and, in Brazil, with the highest elevational range.