Modeling system for predicting enterococci levels at Holly Beach

Area of Science:

• Environmental microbiology
• Water quality monitoring
• Computational modeling

Background:

• Enterococci bacteria in coastal waters pose public health risks.
• Accurate and timely monitoring of recreational water quality is crucial for beach management.
• Existing methods for predicting bacterial levels often lack real-time forecasting capabilities.

Purpose of the Study:

• To develop and validate a novel modeling system for nowcasting and forecasting enterococci levels in coastal recreation waters.
• To enable predictions at any time of day by integrating multiple data sources and correction methods.
• To improve the accuracy and timeliness of public health advisories for beachgoers.

Main Methods:

• Development of an artificial neural network (ANN) model for sunrise-time enterococci prediction.
• Incorporation of clear-sky solar radiation and turbidity corrections to the ANN model for all-day forecasting.
• Utilization of remote sensing algorithms for turbidity estimation and integration of nowcasting/forecasting data.
• Validation using seven years of independent swimming season data.

Main Results:

• The sunrise-time model accurately reproduced 82.63% of advisories with low false positive (2.65%) and false negative (14.72%) rates.
• The enhanced modeling system effectively predicted temporal variations in enterococci levels, from hourly to daily cycles.
• The system demonstrated high efficacy in predicting bacterial levels in coastal beach waters.

Conclusions:

• The developed modeling system provides an effective tool for predicting enterococci levels in recreational waters.
• This system can significantly improve beach management strategies and enhance the protection of public health.
• The integration of ANN, solar radiation, turbidity, and forecasting data offers a robust approach for real-time water quality assessment.