Use of artificial neural network in the prediction of algal blooms

A model to quantify the interactions between abiotic factors and algal genera in Lake Kasumigaura, Japan was developed using artificial neural network technology. Results showed that the timing and magnitude of algal blooms of Microcystis, Phormidium and Synedra in Lake Kasumigaura could be successfully predicted. As for the newly occurring dominant Oscillatoria, results were not satisfactory. The evaluation of the importance of factors showed that Microcystis, Phormidium, Oscillatoria and Synedra were alkalophilic. The algal proliferation for Microcystis, Oscillatoria and Synedra decrease due to the increase in total nitrogen, while the growth of Phormidium is enhanced with more nitrogen. In addition, the algal density is affected by zooplankton grazing but with the exception of Phormidium due to it being poor food source. Algal responses to the orthogonal combinations of the external environmental factors, chemical oxygen demand, pH, total nitrogen and total phosphorus at three levels were modeled. Various combinations of environmental factors enhance the proliferation of some algae while other combinations inhibit bloom formation.