Coastal Monitoring Using Fluorescence Line Height -Based Remote Sensing and Machine Learning: A Seawater-Cooled Facility Case Study

This study presents a unique approach that combines Remote Sensing (RS) technologies and machine learning methods to assess the impact of water-cooled nuclear power plants on marine ecosystems. Normalised fluorescence line height (nFLH), Chlorophyll-a, sea surface temperature (SST) and particulate organic carbon (POC) parameters obtained using NASA MODIS-Aqua/L3SMI satellite data were analysed. The nFLH, which forms the main focus of the study, stands out due to its sensitivity to phytoplankton activity, ability to respond quickly to sudden changes in pollution, and its high-resolution optical accuracy. nFLH may provide more reliable results than Chlorophyll-a in coastal areas with high levels of coloured dissolved organic matter and suspended solids. The correlation coefficients obtained in the study were nFLH: 0.56, Chlorophyll-a: 0.68 and POC: 0.46. These findings demonstrate that nFLH is a useful and moderately correlated indicator in coastal environmental monitoring studies. The power of nFLH in monitoring coastal pollution is demonstrated through the use of Earth observation technology and machine learning. The study highlights the effectiveness of Earth observation technologies in monitoring coastal ecosystem dynamics.