SALINITY DETERMINATION AT THE PARAÍBA DO SUL RIVER DELTA USING EMPIRICAL CORRELATIONS AND THE GOOGLE EARTH ENGINE PLATFORM

Abstract

Salinity is one of the most relevant factors when monitoring water quality because of its impact in both aquatic environment and water supply destined to human consumption. Empirical correlations applied to satellite images can be utilized to determine salinity in estuarine areas. The present study used three empirical correlations, Cilamaya and Cimandiri algorithms and the Normalized Difference Salinity Index at points located in the Paraíba do Sul River estuary and at open sea, with the purpose of testing the efficacy of the applied methods. The results indicated that the Cimandiri algorithm showed better correlations at points with higher salinity, and the Cilamaya algorithm presented better correlations at points located at areas within the estuary. It was concluded that these algorithms can be used for salinity determination. Although these algorithms can be used for salinity determination, recent studies show that application of machine learning techniques can obtain better results for this purpose.