Sea surface reflectance is a key underlying-surface parameter for remote sensing retrievals of sea fog, sea wind, and marine aerosols, and it exhibits significant variability across different times and ocean-current-controlled regions, with maximum variations of up to approximately 15%. Long-term use of a fixed seawater reflectance as an a priori input may introduce systematic biases in radiative transfer simulations and atmospheric correction. Based on hyperspectral observations acquired by the EO-1/Hyperion satellite from 2003 to 2017 over major global ocean-current regions, this study statistically retrieved multi-regional and multi-seasonal sea surface reflectance characteristics and constructed a lookup table constrained by ocean current distribution and seasonal variation. Radiative transfer forward simulations were conducted in representative regions influenced by the Canary Current, California Current, and East Australian Current using both Landsat-8 OLI and Sentinel-2 MSI sensors. The results show that when the lookup-table-based reflectance is used as the ocean surface input, the mean relative error between simulated and observed at-sensor radiance in the visible to near-infrared bands is stably controlled within approximately 6%–9%, and consistent error convergence behavior is observed across different sensors, which is consistently lower than that of the constant-reflectance scheme. These results indicate that incorporating ocean-current and seasonal information into reflectance characterization can effectively improve the consistency and engineering feasibility of marine background radiative modeling, thereby providing a practical improvement for underlying-surface parameter specification in multi-source ocean remote sensing applications. However, the proposed dataset should be interpreted as a statistically stratified regional parameterization rather than a globally complete operational product.
Jiang et al. (Wed,) studied this question.