Null model analysis on bat species co-occurrence and nestedness patterns in a region of the Atlantic Rainforest, Brazil

Abstract This study aimed to test through null models analysis the co-occurrence and nestedness patterns of bat species in a region of the Atlantic Rainforest. Data from 10 communities were tabulated in presence-absence matrices. Two metrics have been used to test the co-occurrence patterns: C-Score, which calculates the “checkerboard” pattern testing the interspecific competition between pairs of species, and V-Ratio, which calculates the covariance between species. All analyses were performed with EcoSim software. For nestedness analyses, we utilized the NODF metric through the Aninhado software. None of all the C-Score results showed a significant pattern of species association, indicating a higher coexistence among bat species. For V-Ratio, three results indicated nonrandom associations of species. The communities are highly nested, with poor species sites being subsets of richer sites. The diversification in this group allows the coexistence of species, decreasing the niche overlap. Nested pattern is common in mammal communities, and the significance of our results is in accordance with the results of the co-occurrence test. Competitive interactions may lead to differences in abundance of species in communities. However, as the method of bat sampling is considered biased, the results using binary data, as in our study, seems to be the most trustful method.