Simulation of shallow water depth data merging for HJ-1A/HSI and EO-1/Hyperion

Our recent study shows that Hyperion data and Hyperspectral Optimization Process Exemplar (HOPE) can effectively retrieve the shallow water depth in the South China Sea. In this study, the HJ-1A/HSI and EO-1/Hyperion remote sensing reflectance (Rrs) is simulated by the ocean radiative transfer software Hydrolight and used to derive shallow water depth with HOPE. Data merging is simulated in four different types of sea water inherent optical properties and four different bottom types. The results show Lee's weight mean theory performs obviously better than arithmetic mean, geometric mean in different conditions, which relates to bottom reflectance and seawater inherent optical properties. We further find that the difference between arithmetic mean, geometric mean and weighted mean decrease with the increase of water depth in the range of 0 to 20m and which depends on the bottom reflectance and seawater inherent optical properties. It will contribute to data merging of shallow water depth derived from hyperspectral data.