Species traits and abundances predict metrics of plant-pollinator network structure, but not pairwise interactions

摘要: Plant–pollinator mutualistic networks represent the ecological context of foraging (for pollinators) and reproduction plants some pollinators). visitation exhibit highly conserved structural properties across diverse habitats species assemblages. The most successful hypotheses to explain these network are neutrality biological constraints hypotheses, which posit that interaction frequencies can be explained by relative abundances, trait mismatches between potential mutualists respectively. However, previous analyses emphasize prediction metrics qualitative structure, may not stringent tests hypotheses. Using a newly documented temporally explicit alpine plant–pollinator network, we show both quantitative structure easy predict, even models predict identity or frequency interactions poorly. A variety phenological morphological as well neutral successfully predicted all tested, without accurately predicting observed interactions. Species phenology alone was best predictor frequencies. were poor predictors pairwise frequencies, suggesting other aspects biology generally considered in studies, such for dipterans, play an important role shaping at this site. Future progress explaining dynamics will require new approaches accurate rather than metrics, better reflect underlying