Padrões e mecanismos envolvidos na estruturação de uma comunidade de formigas arbóreas no cerrado

Abstract

In this study, I evaluated some of the most important mechanisms structuring arboreal ant communities. For this I took advantage of the easy accessibility and high diversity of the arboreal ant fauna of the savannas of central Brazil (Cerrado). First, I assessed the co-occurrence patterns of the arboreal ant species and evaluated to which extent these patterns can be explained by interspecific competition or by habitat association. For that, I used a dataset of 240 trees and performed null model analyses, first accounting only for the species incidence in the community and later incorporating habitat attributes. I performed analyses with all the ant species in the community and then with only the most common species using both a matrix level and a pairwise approach. I found that competition is the most important factor behind the distribution of these arboreal ants, with a much smaller influence of habitat attributes. The effect of competition was detected when I accounted for the 14 most common ant species in the community, both in the matrix level and the pairwise analyses. The importance of competition is consistent with the observed differences in some important biological characteristics of the coexisting ant species. After that, I focused on the patterns of resource use and how they can influence the diversity of arboreal ants. First, I assessed how the heterogeneity of nesting resources and the presence of a dominant species can affect the nesting patterns of arboreal ants. For this, I performed an experimental manipulation in 80 trees belonging to the same species. On each tree I placed 27 cavities with different entrance holes divided in two treatments: high entrance diversity (HE), with 9 different entrance sizes (0.15 to 0.55 mm²) and low entrance diversity (LE), with only 3 different entrance sizes (0.15 to 0.25). I considered Cephalotes pusillus as the dominant species over shelter resources, since it was the species which occupied the highest number of cavities and was also the most common in the experimental trees. I found more nesting ant species nesting in the HE treatment trees, and also a change in the used entrance sizes with an increase in the range and diversity of entrance sizes. I also found significant effects of the presence of the dominant species in nesting patterns, with more ant species and colonized nests when it was absent. There were also changes in the mean entrance size used by the ants in the presence of C. pusillus. Finally, I assessed the competitive outcomes of arboreal ants at food baits. I quantified the differences in resource use by the different ant species and asked if there was a significant relationship between the ability to find resources and to dominate them. I further asked which aspects of the ant are responsible for the observed outcomes. I performed observations in 175 medium sized trees from several species during the day and in a subset of 44 trees at night. I placed a teaspoon of sardine bait in the main branch of each tree and observed the turnover of species at the baits. I found a positive relationship between the discovery ability and the dominance of the ant species, supporting a priority effect over food resources. Overall, I found that colony size was the most important characteristic determining the outcomes at the baits. The present works fits in the growing demand for empirical studies assessing the factor defining species coexistence, showing the importance of competition for the structure of an arboreal ant community.