Coastal water classification using remote sensing data

In this paper, we propose a coastal water quality classification using remote sensing data combined with an optimized fuzzy system. The water classification is based on pollution map derived from surface water map's of four water quality parameters: Turbidity “Turb”, Secchi Disk Depth “SDD”, Suspended Sediments Concentration “SSC” and Chlorophyll-A “Chl-A” estimated from satellite data. Each water map is processed by an hybridization of fuzzy model and genetic algorithm which is modeled by a set of fuzzy rules extracted from the data through two steps procedure. First, fuzzy rules are generated by unsupervised fuzzy clustering of the input data. In the second step, genetic algorithm is applied to optimize the rules. Our contribution is focused on the use of Pollution Signature Draw (PSD) tool in characterizing water quality of some typical sites. This PSW gives the concentration of each parameter and presents useful information to highlight the pollution degrees of the studied sites. The proposed approach was tested on Algiers bay and has highlighted four pollution levels corresponding to High Pollution (HP), Medium Pollution (MP), Low Pollution (LP) and Clear Water (CW).