Nitrate as a predictor of cyanobacteria biomass in eutrophic lakes in a climate change context

We aimed to predict cyanobacteria biomass and nitrate (NO₃^-) concentrations in Lake Võrtsjärv, a large, shallow, and eutrophic lake in Estonia. We used a model chain based on the succession of a mechanistic (INCA-N) model and an empirical, generalized linear model. INCA-N model calibration and validation was performed with long term climate and catchment parameters. We constructed twelve scenarios as combinations of climate forcing from the Intergovernmental Panel on Climate Change (IPCC, 3 scenarios), land conversion (forest to agriculture, 2 scenarios), and fertilizer use (2 scenarios). Models predicted 46% of the variance of cyanobacteria biomass and 65% of that of NO₃^- concentrations. The model chain simulated that scenarios comprising both forest conversion to agricultural lands and a greater use of fertilizer per surface area unit would cause increases in lacustrine NO₃^- (up to twice the historical mean) and cyanobacteria biomass (up to a four-fold increase compared to the historical mean). The changes in NO₃^- concentrations and cyanobacteria biomass were more pronounced in low and moderate warming scenarios than in high warming scenarios because of increased denitrification rates in a warmer climate. Our findings show the importance of reducing anthropogenic pressures on lake catchments in order to reduce harmful pollutant and microalgae proliferation, and highlight the counterintuitive effects of multiple stressor interactions on lake functioning.