Ecotoxicological assessment by microcosm tests on plankton communities

Abstract: The aim of this thesis is to present a short-term plankton community toxicity test in laboratory microcosms, with which the effect of the test substance on species abundance and composition of the community is evaluated. This laboratory microcosm is an intermediate between single-species tests and outdoor mesocosm tests. The brominated flame retardant HBCDD was chosen as the reoccurring test substance in three experiments. It was applied at similar nominal exposure concentrations in toxicity experiments performed in test systems with different grades of complexity. One experiment was performed in an outdoor artificial pond in which cylinders were inserted to isolate replicate mesocosms. The other experiments were performed in a greenhouse environment in 10 L microcosms with sediment and water from a eutrophic lake and brackish water estuary. The short-time plankton community toxicity test proposed in this work was successful in detecting structural changes in plankton community composition due to exposure to anthropogenic chemicals. Several prerequisites for a sensitive community toxicity test were fulfilled. Ecological interactions, like competition and predation, were present and made detection of community level responses like secondary effects possible. The community which developed in the 10 L microcosms included many plankton species. This is important for a sensitive and representative evaluation of the toxic effect. The rich plankton community and a maintained balance between the autotroph and heterotroph communities during the course of the experiment were possible due to the sediment compartment. The sediment acts as a seed-bank from which plankton continuously hatch and enter the water volume. Lake water inoculation did not increase the species richness and made no change in plankton community composition. Using artificial water decreases both work-load and costs, considerably. Another important issue in the evaluation of toxicity tests is the variance of test endpoints between replicates. The minimal detectable difference from control (MDD) which mirrors the variability of test endpoints was compared between the mesocosm experiment and the two microcosm experiments. The three lowest MDD values were more or less in the same range between the experiments and ranged between 5-39%. A possible increase in MDD was observed with increased experimental duration. An exposure period of 7 days is considered optimal in detecting clear community level responses at a minimum time. The effects of the test compounds in the three experiments were evaluated with population level and community level metrics. Judging from the results, no single toxic endpoint was found to be the most sensitive one. This stresses the importance of analysing both community and population metrics.