Modeling Fecal Coliform Bacteria Levels at Gulf Coast Beaches

Abstract

This study presents and compares two models for predicting fecal coliform levels at Gulf Coast beaches in Louisiana, USA. One was developed using the artificial neural network (ANN) in MATLAB toolbox and the other one was developed based on the multiple linear regression method (MLR). A total of six independent environmental variables, including rainfall, tide, wind, salinity, temperature, and weather type along with eight different combinations of the independent variables are capable of explaining about 76 % of variation in fecal coliform levels for model training data and 44 % for independent data. The findings are obtained from the ANN model and the MLR model using six years of bacteriological and environmental monitoring data. The results show that the ANN model performs consistently better than the MLR model. Applications of the ANN model can significantly reduce potential health risks of fecal pollution to beachgoers. The paper provides new insights into environmental processes responsible for the variation in levels of fecal coliform bacteria in coastal beach waters as well.