The effects of the strobilurin fungicide azoxystrobin were studied in brackish water microcosms, with natural plankton communities and sediment. Two experiments were conducted: Experiment 1 (nominal conc. 0, 15 and 60 μg/L, 24-L outdoor microcosms for 21 days) and a second, follow-up, Experiment 2 (nominal conc. 0, 3, 7.5, 15 μg/L, 4-L indoor microcosms for 12 days). The microcosms represent a simplified brackish water community found in shallow semi-enclosed coastal areas in agricultural districts in the Baltic Sea region. Measured water concentrations of the fungicide (Experiment 1) were, on average, 83 and 62% of nominal concentrations directly after application, and 25 and 30% after 21 days, for the low and high dose treatments, respectively, corresponding to mean DT50-values of 15.1 and 25.8 days, for low and high dose treatments, respectively. In Experiment 1, direct toxic effects on calanoid copepods at both test concentrations were observed. Similarly, in Experiment 2, the copepod abundance was significantly reduced at all tested concentrations. There were also significant secondary effects on zooplankton and phytoplankton community structure, standing stocks and primary production. Very few ecotoxicological studies have investigated effects of plant protection products on Baltic organisms in general and effects on community structure and function specifically. Our results show that azoxystrobin is toxic to brackish water copepods at considerably lower concentrations than previously reported from single species tests on freshwater crustaceans, and that direct toxic effects on this ecologically important group may lead to cascade effects altering lower food webs and ecosystem functioning.

Selective Serotonin re-uptake inhibitors (SSRIs) are a class of psychotropic drugs used to treat depression in both adolescents and pregnant or breast-feeding mothers as well as in the general population. Recent research on rodents points to long-lasting behavioural effects of pre- and perinatal exposure to SSRIs which last into adulthood. In fish however, studies on effects of developmental exposure to SSRIs appears to be non-existent. In order to study effects of developmental SSRI exposure in fish, three-spine sticklebacks were exposed to 1.5 A mu g/l of the SSRI citalopram in the ambient water for 30 days, starting two days post-fertilisation. After approximately 100 days of remediation in clean water the fish were put through an extensive battery of behavioural tests. Feeding behaviour was tested as the number of bites against a piece of food and found to be increased in the exposed fish. Aggression levels were measured as the number of bites against a mirror image during 10 min and was also found to be significantly increased in the exposed fish. Novel tank behaviour and locomotor activity was tested in an aquarium that had a horizontal line drawn half-way between the bottom and the surface. Neither the latency to the first transition to the upper half, nor the number of transitions or the total time spent in the upper half was affected by treatment. Locomotor activity was significantly reduced in the exposed fish. The light/dark preference was tested in an aquarium where the bottom and walls were black on one side and white on the other. The number of transitions to the white side was significantly reduced in the exposed fish but there was no effect on the latency to the first transition or the total time spent in the white half. The results in the current study indicate that developmental SSRI exposure causes long-lasting behavioural effects in fish and contribute to the existing knowledge about SSRIs as environmental pollutants.

Contamination can cause a rapid environmental change which may require populations to respond with evolutionary changes. To evaluate the effects of pulp mill effluents on population genetics, we sampled three-spined sticklebacks (Gasterosteus aculeatus) near four pulp mills and four adjacent reference sites and analyzed Amplified Fragment Length Polymorphism (AFLP) to compare genetic variability. A fine scale genetic structure was detected and samples from polluted sites separated from reference sites in multidimensional scaling plots (P < 0.005, 1000 permutations) and locus-by-locus Analysis of Molecular Variance (AMOVA) further confirmed that habitats are significantly separated (F-ST = 0.021, P < 0.01, 1023 permutations). The amount of genetic variation between populations did not differ between habitats, and populations from both habitats had similar levels of heterozygosity (polluted sites Nei's Hs = 0.11, reference sites Nei's Hs = 0.11). Still, pairwise F-ST: s between three, out of four, pairs of polluted-reference sites were significant. A F-ST-outlier analysis showed that 21 (8.4%) loci were statistically different from a neutral distribution at the P < 0.05 level and therefore indicated to be under divergent selection. When removing 13 F-ST-outlier loci, significant at the P < 0.01 level, differentiation between habitats disappeared in a multidimensional scaling plot. In conclusion, pulp mill effluence has acted as a selective agent on natural populations of G. aculeatus, causing a convergence in genotype composition change at multiple sites in an open environment.