Improving the selection of focal species exposed to pesticides to support ecological risk assessments.

Risk assessment investigates the potential impacts of chemicals on non-target organisms. To assess the risk, ecotoxicologists study the responses of a panel of species to different substance exposure. Among the different parameters used to select indicator species (i.e. focal species), their frequency of occurrence is considered as the key parameter. Although species occurrence within a given habitat is easy to determine, we argue that it does not totally reflect the dependence of a species on a given habitat or its potential exposure to chemicals. In this study, we combined the occurrence of species with their habitat-specificity to identify focal species for risk assessment in cereals. We showed that ranking species by occurrence or by habitat-specificity produced different results, with generalist species ranking high in the occurrence list, and species with specialised habitats ranking high in the abundance list. Integrating frequency and abundance of species into one single indicator (the "Indicator Value") allows us to rank species with specialised habitats as high as generalist species in the top rank species list. Although habitat-specificity is an ecologically meaningful concept, it is largely overlooked in eco-toxicological risk assessment, despite the fact that specialists are good indicators of various environmental pressures. This method could be used extensively at different scales and could contribute to studies on risk assessment issue by (re)introducing ecological and population-level concepts and opening up new trait-based approaches.