Biodiversity forecasting in natural plankton communities reveals temperature and biotic interactions as key predictors

Abstract

As natural ecosystems experience unprecedented human-made degradation, it is urgent to deliver quantitative anticipatory forecasts of biodiversity change and identify relevant biotic and abiotic predictors. Forecasting natural ecosystems has been challenging due to their complexity, chaotic nonlinear nature and the availability of adequate data. Here, we use four years of daily abundance of a complex lake planktonic ecosystem and its abiotic environment to model and forecast biodiversity metrics. Using a state-of-the-art equation-free modelling technique, we forecast community richness and turnover with a proficiency greater than the constant predictor several generations ahead (30 days). Short-term forecasts improve substantially using biotic predictors (i.e., autoregressive term or community richness). Long-term forecasts require a more complex set of variables (i.e., biotic interactions), and the forecast proficiency depends strongly on including abiotic predictors such as water temperature. Depending on the forecast horizon, biotic and abiotic predictors can interact nonlinearly and synergistically, enhancing each others effects on biodiversity metrics. Our findings showcase the challenges of forecasting biodiversity in natural ecosystems and stress the importance of monitoring focal biotic and abiotic predictors to anticipate undesired changes.