Performance analysis of the C2RCC processor in estimate the water quality parameters in inland waters using OLCI/Sentinel-3A images

Abstract

Remote sensing can be a powerful tool for long-term spatial and temporal water quality monitoring if proper sets of algorithms are available. To estimate optically significant substances (OSS) by satellite images the water-leaving reflectance (pw) must be accurately estimated because it is directly related to the inherent optical properties (IOPs). For an accurate pw an effective atmospheric correction method must be used to remote the contribution of the atmospheric path radiance. The C2RCC processor has a set of algorithms capable of reduce the atmospheric path radiance, estimate the IOPs and then the OSS concentrations. But, the C2RCC was only tested using OLCI/Sentinel-3 images for coastal areas, therefore, is of huge importance to know about their accuracy for inland waters. The results showed that the pw (with errors from 26.57 to 97.48%), IOPs (with errors from 39.77 to 99.90%) and OSS concentrations (with errors from 49.29 to 148.40%) estimated by C2RCC have no correlation with in situ data. For a long-term use of OLCI/Sentinel-3 images researchers must try to use another atmospheric correction and IOPs estimation methods when studying inland waters.