Food web properties vary with climate and land use in South African streams
MetadataShow full item record
Land use intensification and climate change are two prominent drivers of variation in biological communities. However, we know very little about how these two potential environmental stressors interact. Here we use a stable isotope approach to quantify how animal communities respond to urban and agriculture land use, and to latitudinal variation in climate (rainfall and temperature), in 29 streams across South Africa. Community structure was shaped by both land use and climatic factors. The taxonomic diversity of invertebrates was best explained by an independent negative effect of urbanization, while abundance declined in summer. However, we could not use our variables to predict fish diversity (suggesting that other factors may be more important). Both trophic functional diversity (quantifed usingisotopic richness) and food chain length declined with increasing temperature. Functional redundancy (quantifed usingisotopic uniqueness) in the invertebrate community was high in wet areas, and a synergistic interaction with urbanization caused the lowest values in dry urban regions. There was an additive effect of agriculture and rainfall on abundance-weighted vertebrate functional diversity (quantified usingisotopic dispersion), with the former causing a decline in dispersion, with this partially compensated for by high rainfall. In most cases, we found that a single dominant driver (either climate or land use) explained variation between streams. We only found two incidences of combined effects improving the model, one of which was amplified (i.e. the drivers combined to cause an effect larger than the sum of their independent effects), indicating that management should first focus on mitigating the dominant stressor in stream ecosystems for successful restoration efforts. Overall, our study indicates subtle food web responses to multiple drivers of change, only identified by using functional isotope metrics-these are a useful tool for a whole-systems biology understanding of global change. A freePlain Language Summarycan be found within the Supporting Information of this article.