Connectivity mediates the spatial ecological impacts of a glyphosate-based herbicide in experimental metaecosystems

Metacommunity ecology has shown that connectivity is important for the persistence of a species locally and across connected ecosystems, however we do not know if ecological effects in freshwater ecosystems exposed to biocides leaking from agriculture depend on metaecosystem connectivity. We experimentally replicated metaecosystems in the laboratory using gradostats as a model system. We tested the effects of connectivity, in terms of node distance from the pollutant-source, flow rate, and a glyphosate-based herbicide, on phytoplankton productivity, diversity and stability. Gradostats were composed of interconnected equally spaced nodes where resources and phytoplankton move directionally along a gradient of increasing distance from the source of the polluting herbicide. We hypothesised that ecological effects would be stronger in the node situated closer to the point of herbicide input, but that flow would suppress phytoplankton populations in distant nodes. Overall, RoundUp impacted phytoplankton productivity and stability by reducing algal biomass and abundances. This occurred especially in the node closest to the diluted herbicide point-source and under high flow, where species abundances were heavily suppressed by the effects of the rapidly flowing herbicide. At low flow on the other hand, distant nodes where buffered from the effects of the slow-moving herbicide. No differences in beta and gamma diversity among replicate metaecosystems was found; however, a significant loss of alpha diversity in all metaecosystems occurred through time until the end of the experiment. Together, these results point to the importance of considering aquatic connectivity in management plans for monitoring and mitigating unintended ecological consequences of agrochemical runoff.