Mudanças no uso e na cobertura da terra e seu efeito sobre a estrutura trófica e morfológica de comunidades de formigas (hymenoptera: formicidae) no Brasil central

Abstract

The structural features and abiotic conditions of the environments play a critical role in shaping the communities of the species; however, little is known about how changes in land use and land cover affect the trophic structure and the morphological structure of ant communities. First, I evaluated in what extent the differences in the trophic structure between ground-dwelling and arboreal ant communities vary between forests and savannas. For this, I used stable isotope analyzes of carbon (δ13C) and nitrogen (δ15N) on 70 species of ants collected in areas of semideciduous forest and woodland savanna. Differences in the trophic structure of the arboreal and ground-dwelling ant communities were the same in savanna as they were for forest for both δ15N and δ13C. In both habitats, most arboreal species presented low δ15N values (similar to herbivorous) whereas ground-dwelling ants presented high δ15N values (similar to predator). Although δ13C values were higher in savanna than in forest, reflecting the prominence of C4 grasses in savanna, few savanna ant species obtained most of their carbon from C4 grasses. This indicates that savanna has the vertical segregation in δ15N values as found in the forest, despite the structural differences between these habitats. This can be explained by the much greater differences in taxonomic composition, both at the species and at the genus level, between the different strata than between habitats. After that, I assessed whether the ant species' morphology can be used as a predictor of their trophic position. First, I evaluated the relationships between morphological traits with the relative trophic position of the same species collected in the forest and savanna of the previous chapter. Subsequently, I used multiple regression models in order to create a model that better explains the variation in the trophic position among species. For this I took measurements of 11 morphological traits considered relevant for this study and used the data of the isotopic nitrogen signature (δ15N) to calculate the relative trophic position (RTP) of the ants. Individually, the morphological characters showed weak associations with the species' RTP, while combined in multiple regression models, they explained most of the variation in RTP. Moreover, the predictive power increased significantly when I took into account the foraging/nesting habitat and/or some taxonomic affinities of the species separately. These results suggest that the morphological traits, when combined, have a great potential to be used in the prediction of the trophic position of ant species. Finally, I assessed whether the morphological and/or trophic structure of ant communities vary in different land uses and whether these differences are associated with the tree cover of these habitats. I sampled ants in five types of habitats representing a gradient of increasing tree cover: soy plantation, pasture, woodland savanna, eucalyptus plantation and semideciduous forest. The trophic and morphological structure of ant communities were affected by the type of land use and certain morphological traits were strongly correlated with the structural complexity of the habitat. Habitats with a greater tree cover showed a greater abundance of ant species with larger body size, small and dorsally positioned eyes. In addition, ant communities of the soy plantations and pastures had higher trophic position than those of the forest, savanna or eucalyptus plantation. In general, this thesis provides evidence that the environment and its structural characteristics influence the prevalence of certain functional traits in the ant community.