Development, Pesticide Exposure and Repeatability of an Aquatic, Tri-trophic Laboratory Microcosm

The environmental risk of pesticides is routinely assessed in single-species tests. Multi-species systems are only employed once effect concentrations (divided by safety factors) compared to predicted environmental concentrations give reason for concern. However, direct chemical effects at the individual level often don’t directly translate into impacts observed at higher levels of organization because species interactions play an important role in mediating indirect chemical effects. In this thesis, I aimed to develop a tool that combines the advantages of single-species tests (repeatability, interpretability) and multi-species tests (ecological realism) for the repeatable study of indirect chemical effects mediated by ecological interactions. I show the standardization and testing of a tri-trophic laboratory scale microcosm (Pseudokirchneriella subcapitata, Ceriodaphnia dubia, Hydra viridissima) to better understand and quantify the effects of multiple stressors (e.g. chemicals, food availability and predation) on organisms and their interactions. We found close repeatability of system dynamics in the short term indicating the system’s ability to detect small pesticide effects and bottom-up and top-down effect propagation. Yet, inter-experimental differences between dynamics in controls were found in the long term. An investigation of the influences of a variety of experimental factors showed that deviations from standardized population dynamics were likely caused by medium related factors that acted on algal populations and led to bottom up effects. These likely masked the effects of a herbicide in exposure experiments and I did not gain conclusive results on direct and possibly indirect herbicide effects on grazer and predator populations. My findings demonstrate that considerable consistency and in-depth understanding of the characteristics of all system components are required to achieve repeatability even in apparently simple multi-species systems. My work illustrates possible pitfalls of tools aimed at the generation of repeatable effect data on ecologically relevant endpoints and identifies future research needs to achieve repeatable dynamics in the tri-trophic microcosm and to enhance its applicability.