Pollution signature of water quality using remote sensing data

The aim of this paper is to develop a new methodology to characterise surface water quality using remote sensing data. The general idea is to construct Pollution Signature Draw "PSD" from several water pollution parameters assessed from remote sensing data in order to typify the water quality at any site observed by the satellite. The water quality characterisation is evaluated by the comparison of several signatures to the reference one given by the in-situ measurements. Thus, the pollution degree is highlighted by the spacing deduced from these signatures. In this work, the PSD is constructed from four water parameters which represent Turbidity "Turb", water transparency measured by Secchi Disk Depth "SDD", Suspended Sediments Concentration "SSC" and the Chlorophyll concentration "Ch1". The combination of Landsat, ERS satellite images and fifty (50) samples of in situ measurements using the neural networks "NN" modelling allowed us to map the four parameters cited above. The application of this methodology to the Algiers's bay and the draw of some examples of PSD show that this bay is very affected by the urban and industrial rejects.