Interactions between specialist and generalist natural enemies: Parasitoids, predators, and pea aphid biocontrol

Abstract

Most biological control systems involve a diverse community of natural enemies. We investigated how specialist and generalist natural enemies differ as biological control agents of pea aphids (Acyrthosiphon pisum), and how interactions among natural enemies affect successful control. In alfalfa, pea aphids are attacked by a specialist parasitoid wasp, Aphidius ervi, and a guild of generalist predators primarily made up of Nabis and Orius bugs, coccinellid and carabid beetles, and web-building spiders. In three field experiments, we manipulated the parasitoid, then the generalist predator guild, and finally both classes of natural enemy, and recorded resulting impacts on pea aphid population control. The parasitoid caused little immediate reduction in aphid population growth but caused a large decline after a delay corresponding to the generation time of the parasitoid. In contrast, the generalist guild caused an immediate decline in the aphid population growth rate. However, the generalists did not exert density-dependent control, so aphid densities continued to increase throughout the experiment. The third field experiment in which we simultaneously manipulated parasitoids and predators investigated the possibility of "nonadditive effects" on aphid control. Densities of parasitoid pupae were 50% lower in the presence of generalist predators, indicating intraguild predation. Nonetheless, the ratio of parasitoids to aphids was not changed, and the impact of the two types of natural enemies was additive. We constructed a stage-structured model of aphid, parasitoid, and predator dynamics and fit the model to data from our field experiments. The model supports the additivity of parasitoid and predator effects on aphid suppression but suggests that longer-term experiments (32 d rather than 20 d) would likely reveal nonadditive effects as predation removes parasitoids whose response to aphid densities occurs with a delay. The model allowed us to explore additional factors that could influence the additivity of parasitoid and predator effects. Aphid density-dependent population growth and predator immigration in response to aphid density would likely have little influence on the additivity between parasitism and predation. However, if a parasitoid were to show a strong Type II functional response, in contrast to A. ervi whose functional response is nearly Type I, interactions with predators would likely be synergistic. These analyses reveal factors that should be investigated in other systems to address whether parasitism and predation act additively on host densities.