Shallow aquatic systems are frequent in agricultural regions. They often suffer from enhanced anthropogenic eutrophication and from organic or inorganic pesticides. A good ecological status exists with clear water and submerged macrophyte dominance, while phytoplankton blooms represent a bad ecological state. However, regime shifts between both states occur depending on the level of disturbance. Our French-German project CLIMSHIFT investigates how agricultural run-off (ARO) composed of nitrates, copper and organic pesticides in combination with climate warming affects such systems. We use indoor microcosms to investigate the impact of different exposure scenario to ARO on primary producers alone or in combination with primary consumers. Primary producers consisted of selected strains of planktonic and benthic algae and three different species of macrophytes. Primary consumers were Daphnia magna, Dreissena polymorpha and Lymnaea stagnalis, reflecting different feeding modes. Microcosms ran for 3 or 4 weeks after ARO addition, temperature was 22 and 26°C. We performed 4 microcosm experiments. Exp-1 was based on only the three groups of primary producers, and Exp-2 included all primary consumers. Exp-1 and -2 were exposed to individual or multiple pulses of ARO. Exp-3 investigated if nitrate would alleviate negative effects of pesticides on primary producers and consumers. In a separate experiment, negative effects of ARO on L. stagnalis were evaluated. ARO with or without nitrate negatively affected L. stagnalis and D. magna. In Exp-2, positive effects on periphyton were likely due to indirect negative effects of ARO on snails. In Exp-3, Daphnia developed well in all treatments without pesticides, suppressing phytoplankton and enhancing macrophyte growth. Several pulses instead of one, or nitrate additionally to pesticides increased ARO effects of on periphyton or phytoplankton, respectively. ARO had an indirect positive effect on the development of periphyton or phytoplankton through trophic interactions. Our results underline the necessity to understand the mechanisms behind the observed effect, determined here using structure equation modelling (SEM). Such studies are needed for a realistic risk assessment of these complex yet important aquatic ecosystems.