COS 101-8 - Assessing macroecological patterns in ecological networks: Interaction diversity, turnover, and specialization in plant-frugivore interactions

Friday, August 16, 2019: 10:30 AM
M109/110, Kentucky International Convention Center
Evan C. Fricke, National Socio-Environmental Synthesis Center, Annpolis, MD and Jens-Christian Svenning, Department of Biology, Section for Ecoinformatics and Biodiversity, Aarhus University, Aarhus, Denmark
Background/Question/Methods

Our understanding of broad-scale ecological patterns is based largely on the study of horizontal communities. The opportunity for comparable work on ecological networks is increasing with improving availability and synthesis of interaction data, yet species interactions still remain an understudied dimension of diversity in ecological function. Quantifying broad-scale patterns in species interactions may elucidate proximate drivers of diversity gradients, causes of turnover, and biotic mediators of the links between climate and species composition.

Plant-frugivore networks influence geographic range and population persistence for the roughly half of plant species that are dispersed by animals, and reflect important trophic interactions for vertebrates that feed on fruit. We explore broad spatial patterns in plant-frugivore interaction by assessing latitudinal gradients in interaction specialization and diversity (including functional and phylogenetic) and by quantifying the spatial scale of species and interaction turnover. To advance data availability and synthesis, we compiled the largest database of plant-frugivore networks from the literature and resolving species taxonomy. We address the issue of inconsistent sampling design by incorporating detailed metadata on sampling design from individual studies and including these as covariates in statistical models and by employing null models from the network literature.

Results/Conclusions

Our results show interaction specialization and intensity are related to latitude, with complementary insights provided by metrics used in the network and community ecology literatures. Further, spatial persistence of interactions depends on frugivore taxon (e.g., birds, bats). When considering functional and phylogenetic diversity of individual species’ partners, differences across latitude are small. This mirrors work indicating, for example, similar herbivorous insect diversity on temperate and tropical trees. Contrasting to this flat relationship, animals more frugivorous at low latitudes, suggesting higher interaction intensity in the tropics. Considering specialization measured at the network level, which indicates how different taxa partition their mutualistic partners across the network, greater specialization at low latitudes suggests that tropical mutualists more finely partition resources. When considering how individual interactions differ over space, only for bat-plant interactions are interactions more spatially persistent than expected under a null expectation based on plant and frugivore cooccurrence. Still, roughly 90% of interactions are reported in only a single study, while a small subset appears in many networks. While many interactions appear facultative, some are highly spatially persistent and may play a key role in driving coevolution within plant-frugivore networks.