Agricultural pesticides are known to significantly impact the composition of communities in stream ecosystems. Moreover, agricultural streams are often characterised by loss of physical habitat diversity which may impose additional stress resulting from suboptimal environmental conditions. We surveyed pesticide contamination and rates of leaf litter decomposition in 14 1st and 2nd order Danish streams using litter bags with coarse and fine mesh sizes. Two sites differing in physical habitat complexity were sampled in each stream, and we used this approach to differentiate the effects of pesticides between sites with uniform (silt and sand) and more heterogeneous physical properties. Microbial litter decomposition was reduced by a factor two to four in agricultural streams compared to forested streams, and we found that the rate of microbial litter decomposition responded most strongly to pesticide toxicity for microorganisms and not to eutrophication. Moreover, the rate of microbial litter decomposition was generally 50% lower at sites with uniform physical habitats dominated by soft substrate compared to the sites with more heterogeneous physical habitats. The rate of macroinvertebrate shredding activity was governed by the density of shredders, and the density of shredders was not correlated to pesticide contamination mainly due to high abundances of the amphipod Gammarus pulex at all sites. Our study provides the first field based results on the importance of multiple stressors and their potential to increase the effect of agricultural pesticides on important ecosystem processes.

Summary The spatial organisation of biotic communities derives from factors operating at a wide range of spatial and temporal scales. Despite strong scientific evidence of prevalent spatial control of community composition in freshwater ecosystems, local environmental factors are often considered as the main drivers of community change. Furthermore, taxonomic approaches are most frequently used, and few studies have compared the relative importance of local and regional control of trait versus the taxonomic composition in stream ecosystems. Using a spatially dense data set covering all stream sizes in a lowland European region of c. 42 000 km2 and three organism groups (macrophytes, macroinvertebrates and fishes), we compared the relative importance of spatial and environmental determinants of species and trait composition in the study streams, classified into headwaters (stream order 1–2) and downstream sites (stream order >2). We hypothesised that (i) there is a higher correspondence between environmental conditions and trait composition than with species composition, (ii) dispersal limitation (pure spatial structuring) is greater in headwaters than in downstream sites and (iii) dispersal limitation (pure spatial structuring) is weakest for macroinvertebrates, intermediate for macrophytes and strongest for fishes. The most consistent pattern across organisms and stream order groups was a higher correspondence between environmental variation and trait composition as well as a higher number of environmental variables significantly related to trait composition than with species composition (hypothesis 1). Spatial structuring peaked in headwater macrophyte communities and downstream fish communities (hypotheses 2 & 3) – a pattern that was amplified when separate analyses of traits describing species dispersal potential were undertaken. Our study highlights the potential of traits to capture multiple environmental changes in stream ecosystems and illustrates how organism‐specific and highly context‐dependent patterns in community organisation can emerge as a consequence of interactions between habitat connectivity (i.e. top versus lower parts of the stream network) and organism dispersal potential.

We revealed a history of legacy pesticides in water and sediment samples from 19 small streams across an agricultural landscape. Dominant legacy compounds included organochlorine pesticides, such as DDT and lindane, the organophosphate chlorpyrifos and triazine herbicides such as terbutylazine and simazine which have long been banned in the EU. The highest concentrations of legacy pesticides were found in streams draining catchments with a large proportion of arable farmland suggesting that they originated from past agricultural applications. The sum of toxic units (SumTUD.magna) based on storm water samples from agriculturally impacted streams was significantly higher when legacy pesticides were included compared to when they were omitted. Legacy pesticides did not significantly change the predicted toxicity of water samples to algae or fish. However, pesticide concentrations in bed sediment and suspended sediment samples exceeded safety thresholds in 50% of the samples and the average contribution of legacy pesticides to the SumTUC.riparius was >90%. Our results suggest that legacy pesticides can be highly significant contributors to the current toxic exposure of stream biota, especially macroinvertebrate communities, and that those communities were primarily exposed to legacy pesticides via the sediment. Additionally, our results suggest that neglecting legacy pesticides in the risk assessment of pesticides in streams may severely underestimate the risk of ecological effects.

using Malaise traps placed across the brook, parallel to the brook at distances of 2, 20 and 40 m, and perpendicular to it at a distance of 1-3 m. 2. Of the total of 4175 specimens caught, lotic spedes breeding in Stamperenden accounted for 95.6%. The dominant species were Agapelus fuscipes, Lype reducia, Plectrocnemia conspersa, Potamophylax nigricornis and Silo pallipes. 3. The sex ratios in the catches of most species were uneven. Agapetus fuscipes males were in excess, probably because males fly more frequently than females. 4. The dominant species flew predominantiy above the surface of the brook. Whereas both sexes oi A. fuscipes,' L. reducta and S. pallipes were rarely found even as little as 20-40 m from the brook, the dispersal range of P. conspersa was much wider. 5. The implications of low dispersal for the biological recovery of damaged but subsequently restored streams and stream catchments are discussed. It is concluded that complete biological recovery may be a process requiring decades unless the original aquatic fauna is re-established artificially.

Summary Species richness and assemblages of Trichoptera from 157 ‘pristine’ Danish lowland stream sites were analyzed, for dependence on geographical position of the sites and simple physical variables, using two way indicator species analysis and detrended correspondence analysis. A total of 106 species were recorded, representing ≈ 90% of the species pool known from Danish streams. Only seven species occurred at more than half the sites, whereas an additional 11 species were found at more than a quarter of the sites. Although sites showed significant regional differences in environmental variables (stream order, width, slope and presence/absence of riparian forest), species richness and assemblages were primarily correlated with stream order, width and slope. Maximum richness was found at the largest (5th order) stream sites. Regional differences in species assemblages were found, with several species being absent from the islands Funen and Bornholm. Species assemblages also differed between forested and non‐forested stream sites. We found evidence that stream temperature may be of only minor importance in determining Trichoptera species richness and assemblage composition in Danish streams compared to other size‐related physical factors.