Effect of interspecific competition on species sensitivity distribution models: Analysis of plant responses to chemical stress.

Species Sensitivity Distributions (SSD) are widely used in environmental risk assessment to predict the concentration of a contaminant that is hazardous for 5% of species (HC). They are based on monospecific bioassays conducted in the laboratory and thus do not directly take into account ecological interactions. This point, among others, is accounted for in environmental risk assessment through an assessment factor (AF) that is applied to compensate for the lack of environmental representativity. In this study, we aimed to assess the effects of interspecific competition on the responses towards isoproturon of plant species representative of a vegetated filter strip community, and to assess its impact on the derived SSD and HC values. To do so, we realized bioassays confronting six herbaceous species to a gradient of isoproturon exposure in presence and absence of a competitor. Several modelling approaches were applied to see how they affected the results, using different critical effect concentrations and investigating different ways to handle multiple endpoints in SSD. At the species level, there was a strong trend toward organisms being more sensitive to isoproturon in presence of a competitor than in its absence. At the community level, this trend was also observed in the SSDs and HC values were always lower in presence of a competitor (1.12-11.13 times lower, depending on the modelling approach). Our discussion questions the relevance of SSD and AF as currently applied in environmental risk assessment.