Predator phylogenetic diversity decreases predation rate via antagonistic interactions

Background/Question/Methods

Predator species can have strong top down effects on community composition and ecosystem function, yet it is difficult to extend these results to predict the effects of diverse predator assemblages. Ecologists have recently begun using the phylogenetic diversity (PD) of an assemblage to predict function, most notably in studies of primary producers. Here we use PD to understand the distribution, feeding behaviour and ecosystem-level effects of a diverse predator assemblage.  Specifically, we ask if predator assemblages with higher PD have higher non-additive effects on prey mortality, and therefore higher rates of ecosystem function.

Predators may increase ecosystem function through either additive (complementary) or nonadditive (synergistic) effects on ecosystem function. Higher predator PD will lead to complementary if co-occurring predators show differences in feeding preferences while nonadditive effects are caused by trait-mediated effects of predators on prey species or on each other.

We examined predator phylogenetic diversity in the diverse community of macroinvertebrates found in bromeliads, a natural aquatic mesocosm. We use measures of predator PD to combine three distinct datasets: an observational dataset of predator distribution among bromeliads, a series of feeding trials, and a manipulative experiment of predator PD at the mesocosm level.

Results/Conclusions

We found that phylogenetic distance does not predict differences in predator distribution, indicating that a range of predator PD is found in nature. We did, however, find evidence of a slight negative correlation between predator PD and diet preference. This suggests that distantly-related predators might eat different prey. However, our manipulative experiment showed that increasing predator PD had a negative non-additive effect, caused by an increase in antagonistic interactions among more diverse predators. This did not, however, create any measurable differences in ecosystem function (as measured by rates of decomposition and nitrogen cycling). Our study demonstrates a way to quantify the effects of predator phylogenetic diversity on the structure and behaviour of food webs.