Can we predict community-wide effects of herbicides from toxicity tests on macrophyte species?

International audienceMacrophyte communities play an essential role in the way freshwater ecosystems function. It is thus of great concern to understand how environmental factors, especially anthropogenic ones, influence their composition and diversity. The aim of this study was to examine whether the effects of a herbicide mixture (50% atrazine, 35% isoproturon, 15% alachlor) on single macrophyte species can be used to predict its impact at a community level. In a first experiment we tested the sensitivity of six species (Azolla filiculoides, Ceratophyllum demersum, Elodea canadensis, Lemna minor, Myriophyllum spicatum and Vallisneria spiralis) grown separately and exposed to 0.6–600 gL−1 of the herbicide mixture. In a second experiment, conducted in microcosms, we tested the effects of herbicides on macrophyte assemblages composed of the same six species exposed to 0, 6 or 60 gL−1 of the herbicide mixture. Species grown separately exhibited growth inhibition at 60 and 600 gL−1. At 600 gL−1 the sensitivity differed significantly between species. V. spiralis was the most resistant species, C. demersum, M. spicatum and E. canadensis exhibited intermediate sensitivities, and A. filiculoides and L. minor were the most sensitive species. In microcosms, community biomass and Shannon evenness index were reduced after 8 weeks at 60 gL−1. Communities also exhibited changes in their composition: the relative and absolute abundance of C. demersum increased at 6 gL−1, while the relative abundance of V. spiralis increased at 60 gL−1. These results are in agreement with the individual responses of these species to the herbicides. It is therefore concluded that short-term effects of herbicides on simple macrophyte communities can be predicted from the sensitivity of individual species. However, further investigations are required to examine whether longer term effects can be predicted as well, especially in more complex communities