SWM : Smart Water Management

Quick detection of algal blooms (i.e., high chlorophyll-a concentration) in freshwaters is imperative for effective monitoring of spatial water quality distribution in the context of ecological health and assessment. With the recent advancement of technology, satellite images have been deemed as practically useful tools to identify the spatial distribution of algal blooms. Our study aimed to predict chlorophyll-a concentrations using 13-band satellite images derived from Sentinel-2 (spatial resolution: 10m x 10 m, time interval: 5 days). In order to validate the values from the satellite images, we compared them with simultaneously observed chlorophyll-a concentrations based on fluorescence measurement. The goal of this study is to improve the accuracy of predictions induced from satellite images. The analytical techniques (multiple linear regression, decision-tree classifier, and artificial neural network) were comparatively evaluated. The results showed that artificial neural network exhibited the best performance among them, improving more than 37% accuracy compared to that of multiple linear regression. In the end, it was successful to create algal bloom maps using a new algorithm to analyze spatial algae management.