Bathymetry estimation of shallow water using multi-spectral satellite images

Bathymetry estimation is an important issue but also achallenge problem. There are two main approaches for bathymetry measurement, echo sounder and optical satellite images. Echo sounder can measure the bathymetry thousands of meters with a narrow band along the track. Therefore, it is difficult to estimate the bathymetry in a wide area. On the other hand, optical satellite image can survey a large area but the penetration is less than 20 meters. In this study, we adopt a simple method for estimating water depths from multi-temporal Formosat-2 data for shallow bathymetry. After the correction with the tidal level difference of the acquired satellite images and estimated attenuation coefficient and bottom reflectance, the water depth can be derived. Formosat-2 multi-spectral image scenesin the southeast area of Penghu Island are used for experiment. Theperformance evaluation is conducted bycomparing the result with depth data from echo sounder sonar surveying. Copyright