Linking new national active biomonitoring data with stream macroinvertebrate communities suggests large-scale effects of toxic contamination on freshwater ecosystems

Since recent years, an increasingly large number of toxic chemicals enters watercourses threatening freshwater biodiversity. But ecological studies still poorly document the quantitative patterns linking exposure to complex mixture of toxic chemicals and species communities' integrity in the field. In this context, French monitoring authorities have recently deployed at a national scale in situ biotests using the feeding inhibition of the crustacean Gammarus as toxicity indicator. In this paper, we conjointly exploit this new type of biomonitoring dataset and ecological data for macroinvertebrates to gain information about the structuring influence of toxicity on aquatic communities. Especially, we used multivariate analyses with variation partitioning for testing the hypothesis that toxicity (feeding inhibition index) can explain variations in the taxonomical composition between 76 stations on French streams while, for different spatial scales, estimating the confounding influences of other environmental and spatial factors. Our results showed that changes in the toxicity indicator were significantly associated with specific changes in the taxonomic composition of stream macroinvertebrate communities. That association was weakly confounded with the effects of environmental and spatial factors, especially at the largest spatial scale considered. That taxon turnover linked to toxicity was associated with reduced richness at the community scale, and the replacement of native taxa by alien taxa. Overall, our study thus supports the hypothesis that toxic contamination modifies the structure of stream communities and ergo threatens aquatic biodiversity. [Display omitted] •We used new in situ bioassay data to estimate chemical toxicity in streams•Toxicity changed the structure and reduced the richness of invertebrate communities•The effect was independent of confounding physico-chemical and spatial parameters