Interaction networks of avian mixed‐species flocks along elevation in the tropical Andes

Ecological communities are comprised of species that interact with each other and those interactions ultimately generate community structure. Network theory provides a useful framework to study communities, by simultaneously considering species composition and the interactions among species. In this study, I use mixed‐species flocks as model systems to gain insights on community and network structure. Specifically, I use co‐occurrence network analyses to explore if avian mixed‐species flocks change in richness and composition and/or in network structure and pair‐wise associations, across elevations in the tropical Andes of Bolivia. Networks of flocking species changed both in composition and in the frequency and realization of pair‐wise interactions across elevations, but changes in pair‐wise associations explained most of network turnover along elevation. Pair‐wise interactions changed rapidly, with shared species changing in position and importance within the network. Network dissimilarity was mostly explained by changes in the nature of associations rather than by differences in composition. Altogether, results show that montane mixed‐species flocks are composed of loosely connected species and that most species have the potential of switching associations, often increasing in association strength at high elevations (up to 3150 m). Networks increased in connectivity and cohesion with elevation; flocks in lower elevations had more connections and these were less even. Above 3150 m a.s.l., there was rapid decay suggesting that flocks above this critical point are less connected and cohesive. This study exemplifies how combining community, network and pair‐wise analyses can provide a more holistic view on the responses of species and assemblages to environmental gradients.