Knowing your enemies: Integrating molecular and ecological methods to assess the impact of arthropod predators on crop pests.

The importance of natural enemies as the foundation of integrated pest management (IPM) is widely accepted, but few studies conduct the manipulative field experiments necessary to directly quantify their impact on pest populations in this context. This is particularly true for predators. Studying arthropod predator-prey interactions is inherently difficult: prey items are often completely consumed, individual predator-prey interactions are ephemeral (rendering their detection difficult) and the typically fluid or soft-bodied meals cannot be easily identified visually within predator guts. Serological techniques have long been used in arthropod predator gut-contents analysis, and current enzyme linked immunosorbent assays (ELISA) are highly specific and sensitive. Recently, polymerase chain reaction (PCR) methods for gut-contents analysis have developed rapidly and they now dominate the diagnostic methods used for gut-contents analysis in field-based research. This work has identified trophic linkages within food webs, determined predator diet breadth and preference, demonstrated the importance of cannibalism and intraguild predation within and between certain taxa, and confirmed the benefits (predator persistence) and potential disadvantages (reduced feeding on pest species) of the availability of alternative nonpest prey. Despite considerable efforts to calibrate gut-contents assays, these methods remain qualitative. Available techniques for predator gut-contents analysis can provide rapid, accurate, cost-effective identification of predation events. As such, they perfectly compliment the ecological methods developed to directly assess predator impacts on prey populations but which are imperfect at identifying the key predators. These diagnostic methods for gut-contents analysis are underexploited in agricultural research and they are almost never applied in unison with the critical field experiments to measure predator impact. This paper stresses the need for a combined approach and suggests a framework that would make this possible, so that appropriate natural enemies can be targeted in conservation biological control.