Validation of a green-red quasi-analytical algorithm for inherent optical properties in East China Sea

Abstract

A green-red quasi-analytical algorithm, QAA-GRI, was calibrated to derive inherent optical properties (IOPs) using an in situ dataset from Lake Qiandaohu (QDH). First, 510 nm was chosen as the reference band based on the general structure of the quasi-analytical algorithm (QAA). Second, a green-red index (GRI), which was calculated from the remote sensing reflectance at the three wavelengths (510, 560 and 620 nm), was used to retrieve the total absorption coefficients at the reference band, a(510) . A semi-analytical model based on a(510) and the GRI was proposed to replace the empirical model in original QAA. Subsequently, QAA-GRI, is calibrated to analytically retrieve total absorption coefficient for Lake Qiandaohu (QDH). The algorithm was further validated using the in situ data set collected in East China Sea (ECS) on January 1-12, 2016 and May 25-June 2, 2017. The QAA-GRI’s performance in ECS was compared with that of QAA-v5. Our results show that the QAA-GRI performs better in ECS with mean R2 of 0.87, compared with that the QAA-v5 of 0.53, and a mean absolutely percentage error of 19.2%, compared with that the QAA-v5 of 24.8%, respectively. These results indicate the potential of QAA-GRI to accurately estimate the IOPs for coastal and inland waters.