What are the relationships between species diversity, interaction diversity and resource specialization, and how do these change across spatial scales and abiotic gradients? To answer these questions, spatial scale and the biological context of interactions need to be determined. At the smallest scales, local interaction diversity is an important determinant of network parameters at larger scales, and contributes to ecosystem stability. However, most networks do not consider local interactions and their contributions to regional networks, rather they are created from literature or from coarse grain sampling, thus network dynamics may not be relevant to local interactions between species or even to regional community dynamics. We have used simulation models in combination with extensive natural history data to test hypotheses about tritrophic interaction diversity at the smallest community scales within small forest plots up to global patterns of specialization and interaction diversity.
Results/Conclusions
Inferences from simulated communities indicate a strong impact of scale on multiple network parameters, driven by changes in species richness and specialization. Network parameters for natural communities are also strongly scale dependent. Thus, trophic interaction diversity measured at scales that are utilized for most food webs are often misleading and not relevant to local community processes, and conclusions about networks across latitudinal gradients are misleading. Network ecology should continue to rely on careful natural history for more relevant hypothesis tests of how networks change across gradients.