Inversion of Chlorophyll-a Concentrations in Chaohu Lake Based on GF-1 WFV Images

Abstract

Chlorophyll-a is an important parameter that is used to measure the water quality, and its concentrations indicate the degree of eutrophication. Taking the waters of Chaohu Lake as the research area, four remote sensing inversion models of the chlorophyll-a concentrations were constructed based on GF-1 WFV images using empirical and semiempirical/semianalytical methods. The optimal model was selected by evaluating the model validation accuracies, and this model was used to invert the spatial distributions of the chlorophyll-a concentrations in the waters of Chaohu Lake from May to July 2017. The results show that the NDCI model is the most suitable for inverting chlorophyll-a concentrations in the waters of Chaohu Lake based on GF-1 WFV images. This model has the highest validation accuracy, with a determination coefficient of 0.933, a Root mean square error of 2.85 µg/L and a mean absolute percentage error of 19.69%. The spatial distributions of the chlorophyll-a concentrations in the waters of Chaohu Lake from May to July 2017 all showed decreasing trends from Western Half Lake to Eastern Half Lake, the areas of higher chlorophyll-a concentrations in May and June were mainly concentrated in the northern waters of Western Half Lake, and the area with the highest chlorophyll-a concentrations in July was located in the northwestern waters of Middle Half Lake. The mean chlorophyll-a concentration in the waters of Chaohu Lake from May to July 2017 was 12.5 µg/L, which indicated slight eutrophication.