A chlorophyll-a retrieval algorithm for the Coastal Zone Imager (CZI) onboard the HY-1C satellite in the Pearl River Estuary, China

Abstract

ABSTRACT Accurate estimation of the near-surface chlorophyll-a (Chl-a) concentration in turbid estuaries via remote sensing is an ongoing challenge. Recently, with the successful launch of the third satellite of the China Ocean Colour Satellite Series, HY-1 C, equipped with the Coastal Zone Imager (CZI) with a 3-day revisit period and 50-m resolution in four bands, this study constructed a Chl-a inversion model via symbolic regression (SR) in the Pearl River Estuary (PRE) considering 60 in situ samples to explore the current ocean colour monitoring potential of the CZI. Compared to five state-of-the-art algorithms, the SR model attained a lower mean absolute percentage error (MAPE, 35.51%), mean absolute error (MAE, 1.56) and root-mean-square error (RMSE, 0.24). Additionally, the Chl-a spatial distributions determined with the SR model also better coincided with those determined with synchronous satellite products. The above results indicate that the SR model performed better than the five state-of-the-art models and could be effectively applied to turbid PRE waters. This research may provide a reference regarding the practicality and feasibility of the CZI in water environment monitoring in coastal and estuarial waters at a high temporal-spatial resolution.