Author(s): Hossein Amini; Man Yue Lam; Reza Ahmadian
Linked Author(s): Reza Ahmadian, Man Yue Lam
Keywords: Fecal Indicator Organisms Tidal Phase machine learning bating water
Abstract: Fecal Indicator Organisms (FIOs) have caused major health issues in the past years in bathing waters. Exploring the water quality stressors and how the dynamics of the environment can lead to a change in FIO concentrations have always been an important topic among the research around the globe. In coastal environment, the FIO transport and decay processes under ebb and flood tides can be significantly different. Nevertheless, previous Artificial Intelligence (AI) models were developed with the assumption that FIOs in ebb and flood tides are governed by the same process, and an AI model was used for both ebb and flood tides. In this study, Machine Learning (ML) FIO prediction models were developed for ebb and flood tides respectively. The test site was Swansea Bay, UK, because of its availability of data. Initial results show that there is a difference between FIO concentration in Ebb compared to Flood, and Deep Learning (DL) model can predict the E. coli and Enterococci concentrations with high accuracy. As future work, interpretability of the DL models will be tested with hydrodynamic models.
Year: 2025