Improving accuracy of Eutrophication State Index estimation in Chaohu Lake by moderate resolution remote sensing data driven method

Trophic Level Index (TLI) calculated from several water quality monitoring indicators is often used to assess the general eutrophication state of inland-lake. In this paper, we proposed a data driven inland-lake eutrophication mapping method by using artificial neural network (ANN) to build relationship from remote sensing data and in-situ TLI sampling. Low spatial resolution remote sensing data (MODIS, 250-m and 500-m) and moderate spatial resolution remote sensing data (OLI, 30-m) together with in-situ observations are acquired to train the net. Result demonstrates that TLI obtained from medium-resolution remote sensing images is more accurate than which from low resolution remote sensing data, and more accurate than TLI calculated from the water quality factors retrieved from remote sensing images. This method provides an efficient way of mapping the TLI spatial distribution in-inland lake.