Biodiversity, foodweb complexity, and autonomous pest control in coffee agroecosystems

Background/Question/Methods

The simplification of communities is thought to have important consequences for ecosystem functioning. In agriculture, these consequences may have immediate ramifications for species interactions that are valuable to humanity, including pest control and pollination. However, understanding if interactions between foodweb complexity and diversity are important for regulating functions and services is still relatively understudied question. Here, we present a series of empirical studies that evaluate the importance of foodweb complexity and diversity in regulating the pest control services of ant predators of coffee insect pests. We used lab and field studies in shade coffee plantations to ask three questions: (1) Does non-crop plant species identity modify the effect strength of ant pest control services? (2) Does a parasitoid of a dominant predator influence multiple predator biodiversity effects? (3) Does pest species richness facilitate predator biodiversity effects?

Results/Conclusions

(1) Two years of surveying Azteca sericeasur ants nesting in Inga spp. trees and non-Inga spp. trees revealed that nesting on non-Inga spp. trees resulted in 2.6 times better pest control per bush, 2 times more coffee bushes foraged, and 3.6 times better estimated pest control services at the nest level compared to the ants nesting in Inga spp. trees. (2) Laboratory experiments revealed that including a parasitoid phorid fly of A. sericeasur modified the importance of redundant ant predators. In the presence of the phorid fly, A. sericeasur no longer provided control of the coffee pest, but additional ant species compensated for the loss of pest control. Thus, more predator species were needed to reduce pest infestation in the presence of a phorid parasitoids. (3) In a third study, we altered the number of pest species (1 or 3) and ant predator species (0,1, or 3) in concert. Our results suggested that multiple predator effects were dependent on the diversity of pest species. With only a single pest species, sole predators were equivalent to multiple predators. However, with three pest species, multiple predators limited damage to a greater extent than did single predator species treatments. These results suggest diversity and species identity at lower trophic levels, as well as food chain length, modify single and multiple predator effects on coffee pests. Further, they have important implications for understanding the complex network of interactions involved in autonomous pest control in agroecosystems.