An exposure-based, ecology-driven framework for selection of indicator species for insecticide risk assessment

In the current “tiered” paradigm for evaluating risks of insecticidal products, one of the first decisions that must be made is the selection of indicator species to be used in toxicity assays. However, as yet, no formal system has been developed to determine whether proposed indicator species are relevant to the ecology of the crop system where the product will be released. Here, we propose a protocol that provides information on the ecology and trophic linkages of organisms within agro-ecosystems, and demonstrate its implementation within maize agro-ecosystems, which have been a major focus of recent insecticidal developments. We use molecular gut-content assays and network analysis to identify species that are likely to be exposed to plant-incorporated products, and that likely have important functional roles in interaction webs in the maize ecosystem. The vast majority of arthropod abundance was found in the soil (97% of specimens per m2 were found in the soil column). Only nine of the 382 morphotaxa met all three of the ecological criteria (high abundance, corn consumption, degree of connectedness within the network) for inclusion as indicator species, only one of which, Orius insidiosus (Say) (Hemiptera: Anthocoridae), has routinely been considered in risk assessment. Ecological data collected in studies such as this one can be used to ensure that insecticide risk assessments are ecologically relevant. We advocate the use of large-scale field bio-inventories, combined with molecular gut-content assays and ecological network analysis as regular components of the preparation and design phases of all future risk-assessment programs.