Optimizing size thresholds in a plant-pollinator interaction web: towards a mechanistic understanding of ecological networks

摘要: Using functional traits together with abundance effects strengthens the prediction of interactions between pairs species in ecological networks. Insights into way interact as well accuracy can be gained when thresholds for trait value combinations that make possible are optimized through model selection. I present novel data two subalpine plant–pollinator communities and build several stochastic models integrating flower morphological threshold rules allow or restrict species. The number correctly predicted was highest were set so insect’s proboscis not shorter than nectar-holder depth minus 1–1.6 mm, wider width 0.5 mm. In comparison based solely on plant effects, incorporating size better distribution differences plants insects. This indicates a mechanistic approach interaction webs provides valuable information community structure. implications functioning discussed.