Spatial-Temporal Variation of the Radiometric Color of the Largest Ten Lakes in China

The optical properties of inland waters are complicated, limiting the use of remote sensing to monitor inland water quality in large regions over long time periods. The radiometric color of water, reflecting overall water quality, is an important parameter in traditional water quality investigations. In this study, we retrieved the Forel-Ule (FU) class radiometric color parameters from MODIS surface reflectance product (MOD09) image, and analyzed its long term time-series variation in the largest ten lakes in China during 2000-2012. Based on validation by in situ measured reflectance data, the MOD09-derived FU products are reliable, with a linear correlation coefficient of 0.97, RMSE of 1.2, and average relative error of 7.7%. Since FU class is an optical parameter, it can be derived from optical remote sensing data without seasonal and area limitations. The FU class products were used to analyze the spatial and temporal variations of turbidity and trophic states in the ten lakes. The largest ten lakes in China include typical types with different optical properties, so the FU class can also be used to assess the quality of other inland waters. The FU class retrieval method can also be applied to other remote sensing data after a simple normalization process. As a result, the FU class radiometric color can be used to assess inland water quality in large regions over long time periods.