ANNA - Artificial Neural Network model for predicting species abundance and succession of blue-green algae

Predictive potential of deductive and inductive phytoplankton models are compared regarding their usefulness for forecasting and control of harmful algal blooms. While applications of deductive models still seem to be restricted by lack of knowledge, ad hoc inductive models sometimes prove to be straightforward and useful. The inductive neural network model ANNA is documented by means of an application to Lake Kasumigaura, Japan. ANNA was validated for five blue-green algae species where predictive accuracy has improved with increased event and time resolution of training data. A scenario analysis on species succession has demonstrated the potential of ANNA for hypothesis testing. Finally, implications for use of ANNA for operational algal bloom control are discussed.