Search

International audience; RivFunction is a pan-European initiative that started in 2002 and was aimed at esta- blishing a novel functional-based approach to assessing the ecological status of rivers. Litter decomposition was chosen as the focal process because it plays a central role in stream ecosystems and is easy to study in the field. Impacts of two stressors that occur across the continent, nutrient pollution and modified riparian vegetation, were exam- ined at >200 paired sites in nine European ecoregions. In response to the former, decomposition was dramatically slowed at both extremes of a 1000-fold nutrient gra- dient, indicating nutrient limitation in unpolluted sites, highly variable responses across Europe in moderately impacted streams, and inhibition via associated toxic and addi- tional stressors in highly polluted streams. Riparian forest modification by clear cutting or replacement of natural vegetation by plantations (e.g. conifers, eucalyptus) or pasture produced similarly complex responses. Clear effects caused by specific riparian distur- bances were observed in regionally focused studies, but general trends across different types of riparian modifications were not apparent, in part possibly because of important indirect effects. Complementary field and laboratory experiments were undertaken to tease apart the mechanistic drivers of the continental scale field bioassays by addressing the influence of litter, fungal and detritivore diversity. These revealed generally weak and context-dependent effects on decomposition, suggesting high levels of redundancy (and hence potential insurance mechanisms that can mitigate a degree of species loss) within the food web. Reduced species richness consistently increased decomposition variability, if not the absolute rate. Further field studies were aimed at identifying impor- tant sources of this variability (e.g. litter quality, temporal variability) to help constrain ranges of predicted decomposition rates in different field situations. Thus, although many details still need to be resolved, litter decomposition holds considerable potential in some circumstances to capture impairment of stream ecosystem functioning. For instance, species traits associated with the body size and metabolic capacity of the con- sumers were often the main driver at local scales, and these were often translated into important determinants of otherwise apparently contingent effects at larger scales. Key insights gained from conducting continental scale studies included resolving the appar- ent paradox of inconsistent relationships between nutrients and decomposition rates, as the full complex multidimensional picture emerged from the large-scale dataset, of which only seemingly contradictory fragments had been seen previously.