Partitioning components of risk reduction in a dragonfly–fish intraguild predation system

Risk to prey imposed by intraguild predation (IGP) can be influenced by a number of factors, yet to date, few studies have measured the contributions of these factors to overall risk. A three‐species IGP system with larvae of the dragonfly Anax junius as IG (top) predators, larvae of the dragonfly Plathemis lydia as IG prey (intermediate predators), and fathead minnow hatchlings (Pimephales promelas) as shared prey was used to estimate the contribution of the following three factors to shared‐prey mortality rate in combined predator treatments: (1) the trophic effect of the IG predator on IG prey density; (2) the effect of reduced shared prey consumption by the IG prey in the presence of the IG predator; and (3) the effect of alternative prey for the IG predator. These factors were integrated into a model of multiple predator effects. To quantify minnow mortality, P. promelas were exposed to A. junius only, P. lydia only, A. junius and P. lydia, or neither in a two‐by‐two factorial design. Additional treatments, in which one or both predators were unable to feed, were used to isolate behavioral (activity level) changes in dragonfly larvae. When predators preyed in combination on P. promelas their impact was less than that of the summed effects of the two predators, each in the absence of the other—a result termed risk reduction. A. junius consumed a significant number of P. lydia when they were present (i.e., IGP), and behavioral interactions between A. junius and P. lydia were asymmetric. The presence of A. junius caused P. lydia to become less active, while the presence of P. lydia elicited a diet shift in A. junius to include some P. lydia. Interactions between predator species, specifically IGP, influenced prey survival. Trophic and behavioral effects of IGP were similar in magnitude. These results highlight the importance of trophic and behavioral interactions in predator–prey systems and also suggest that effects of multiple predators may not be predictable based on the sum of individual effects. Determining the effects of multiple predators requires the identification of mechanisms that contribute to nonadditive prey responses. Corresponding Editor: O. J. Schmitz.