Artificial neural network for algal blooms


  • Randy Showstack


Incorporating factors such as turbulence, water temperature, turbidity, color, and nitrogen and phosphorus concentrations, a doctoral candidate at the University of Adelaide in South Australia has developed a computer model to predict outbreaks of blue-green algae (cyano-bacteria) in rivers weeks before they occur.

Artificial neural networks (ANNs) used in the computer-based modeling enable the model to learn key factors that can lead to an algal outbreak, according to Gavin Bowden, a student in the Department of Civil and Environmental Engineering who developed the model. ANNs, according to Bowden, attempt to imitate the complex, non-linear, and parallel mechanisms involved in the interpretation of information by biological neural networks.