Objectives Sunglint is a significant phenomenon in marine optical remote sensing, and presents a dual character: its variability aids in marine target monitoring, yet it degrades the retrieval accuracy of water color information. Therefore, the precise calculation or assessment of sunglint signals has become a key research focus in ocean remote sensing. Affected by the remote sensing scale effect, statistical sunglint models suitable for coarse spatial resolution data are difficult to apply directly to high spatial resolution remote sensing data. The inter-band relationship approach, leveraging the absorption characteristics of water in the near-infrared (NIR) to shortwave infrared (SWIR) spectrum, provides an effective pathway for sunglint correction in high spatial resolution marine optical remote sensing data.

Methods This paper utilizes data from the 50 m spatial resolution coastal zone imager (CZI) aboard the Haiyang-1 C/D (HY-1 C/D) satellites. Based on the optical processes of sunglint and the image characteristics of marine targets, we conducted a detailed analysis of the image features of sunglint and the remote sensing scale effects within high spatial resolution marine optical data. Furthermore, a dedicated sunglint correction method for CZI data was developed. The results demonstrate that post-correction CZI data effectively highlight water color information such as eddies and water masses.

Results The uncertainty of Rayleigh correction affected by sunglint reflection at 460 nm, 560 nm, and 650 nm is reduced by 65%, 80%, and 89%, respectively.

Conclusions The proposed method does not rely on atmospheric parameters, facilitating the rapid dynamic monitoring and retrieval of coastal water color using CZI data.