Fatores associados à diversidade beta de interações entre plantas e beija-flores na Mata Atlântica

Abstract

The composition of species in communities can vary in space and along geographic and environmental gradients. These changes in species composition can affect the interactions between organisms, which may vary according to changes in the identity of species or as a result of changes in species attributes. We investigate how plants and hummingbirds from the Atlantic Forest interact along environmental gradients, as well as the influence of traits in interactions. We expect that (i) the change in species composition (component of taxonomic turnover) is the main cause of the beta diversity of interactions (i.e, change in interactions) between networks; (ii) hummingbirds species has a preference for plant traits rather than a preference for the plant species and (iii) (iii) larger environmental differences and geographic distance results in networks with distinct interactions between plants and hummingbirds. We have compiled from the literature 22 interaction networks of 278 plant and 31 hummingbird species throughout the Atlantic Forest, as well as the organism’s specific traits. Firstly, we analyzed the dissimilarity between networks over geographic distance (km), decomposing this dissimilarity into two components, taxonomic turnover, and rewiring. Later, we investigated species functional traits in the interaction networks. Finally, using Pearson correlations and Multiple Regression on distance Matrices - MRMs, we investigate how environmental variables influence the beta diversity of interactions. The taxonomic turnover was the component that better explained the beta diversity of interactions, increasing its explanatory power according to the distance between networks. Due to the high taxonomic turnover, rewiring was low and its explanatory power decreased with distance. At the functional level, we did not identify changes in the interactions according to the distance gradient, revealing the overlapping of links between the analyzed networks in the functional space of interactions. Hummingbirds, for example, interacted with a wide variety of plant traits, with no evidence of correspondence between commonly associated attributes such as corolla length and hummingbird’s bill. Geographical distance, minimum temperature, and precipitation explained 31% of the taxonomic changes of the interactions, while only geographical distance influenced the functional changes of the links and explained 28% of these changes. Our results reveal patterns in the beta diversity of interactions for plant and hummingbird networks throughout the Atlantic Forest and that species composition turnover is the main component that drives changes in interactions. Hummingbirds interact with plants that have a wide variety of traits, revealing the group's versatility in terms of interactions.