With 4 figures and 1 table in the text Abstract. Since 1995, Dikerogammarus villosus SOWINSKI, a Ponto-Caspian amphipod species, has been invading most of Western Europe's hydrosystems. D. villosus geographic extension and quickly increasing population density has enabled it to become a major component of macrobenthic assemblages in recipient ecosystems. For this study, the ecological characteristics of D. villosus on a mesohabitat scale were investigated at a station in the Moselle River. The results show that this amphipod is able to colonize a wide range of substratum types, thus posing a threat to all freshwater ecosystems. Rivers whose dominant substratum is cobbles and which have tree roots along the banks could harbour particularly high densities of D. villosus. A relationship exists between substratum particle size and the length of the individuals, and spatial segregation according to length was shown. This allows the species to limit intra-specific competition between generations while facilitating reproduction. A strong association exists between D. villosus and other Ponto-Caspian species, such as Dreissena polymorpha and Corophium curvispinum, in keeping with Invasional Meltdown Theory. Four taxa (Coenagrionidae, Calopteryx splendens, Corophium curvispinum and Gammarus pulex) exhibited spatial niches that overlap significantly that of D. villosus. According to the predatory behaviour of the newcomer, their populations may be severely impacted.

The growing need to evaluate the quality of aquatic ecosystems led to the development of numerous monitoring tools. Among them, the development of biomarker-based procedures, that combine precocity and relevance, is recommended. However, multi-biomarker approaches are often hard to interpret, and produce results that are not easy to integrate in the environmental policies framework. Integrative index have been developed, and one of the most used is the integrated biomarker response (IBR). However, an analysis of available literature demonstrated that the IBR suffers from a frequent misuse and a bias in its calculation. Then, we propose here a new calculation method based on both a more simple formula and a permutation procedure. Together, these improvements should rightly avoid the misuse and bias that were recorded. Additionally, a case study illustrates how the new procedure enabled to perform a reliable classification of site along a pollution gradient based on biomarker responses used in the IBR calculations.

No abstract