Assessing entomological collection data to build pollen interaction networks in the tropical Amazon forest

Abstract

Interactions between species are the driving force behind ecological dynamics within communities. For this reason, community structure has been described as networks of interacting species for over a century. However, the high canopies in forests such as the tropical Amazon pose an additional challenge to understand interactions. The main objective of the present study is to infer the interactions between bees and plants collected in the Carajás National Forest (Eastern Amazon) and its surroundings from the analysis of pollen loads from bee specimens deposited in biological collections and to determine the trophic niches of the evaluated bee species. This study also aims to contribute to the restoration planning of deforested areas in the eastern Amazon by identifying priority plant species that could help in restoring interaction networks. Pollen loads samples were collected from bees deposited in two entomological collections. The samples obtained were prepared according to Erdtman's acetolysis method, and the pollen grains were counted, measured and microphotographed. We found 154 bee-plant interactions. The pollen material identified in the 72 bee samples (51 species; 16 genera) showed 82 pollen types (plant taxa), distributed in 28 families and 54 genera, with 43 fully identified species and three indeterminate pollen types. The amplitudes of the trophic niche indicated that the genera Trigona, Centris and Xylocopa had more diverse diets. This work also emphasizes the importance of understanding the plant species that make up the diets of bees for management, conservation and restoration strategies.