A physical descriptive model for predicting bacteria level variation at a dynamic beach.

Area of Science:

• Environmental Science
• Water Quality Monitoring
• Microbiology

Background:

• Escherichia coli (E. coli) contamination in recreational waters poses a public health risk.
• Predicting E. coli levels is challenging due to complex interactions between environmental factors and dynamic beach conditions.
• Traditional models like multiple linear regression may not adequately capture non-linear relationships influencing bacteria levels.

Purpose of the Study:

• To develop and validate a rational-based physical descriptive model (PDM) for predicting E. coli levels in dynamic beach environments.
• To improve the accuracy of E. coli predictions compared to existing statistical and geometric mean models.
• To account for the cumulative effects, intensity, duration, and timing of storm events on bacteria loadings.

Main Methods:

• Developed a comprehensive Physical Descriptive Model (PDM) incorporating precipitation, creek discharge, and time-related factors.
• Included analysis of lag times between environmental events and sample collection, and parameter thresholds.
• Compared PDM performance against previously developed statistical models and the geometric mean method.

Main Results:

• The PDM demonstrated improved accuracy in predicting E. coli levels in beach water.
• The model successfully accounted for complex, non-linear relationships between environmental factors and bacteria concentrations.
• An overall accuracy of 75% was achieved for E. coli level predictions across five case years.

Conclusions:

• The developed PDM offers a more accurate and comprehensive approach to predicting E. coli in dynamic coastal environments.
• The PDM's ability to integrate multiple physical and temporal factors enhances its predictive power.
• This model provides a valuable tool for water quality management and public health protection at recreational beaches.