Bathymetric Modeling from Time Series of Multispectral Satellite Images by Using Google Earth Engine: Understanding Error Distribution by Depth

Bathymetric data could be extracted by means of remote sensing techniques from satellite sensors known as satellite derived bathymetry (SDB). This study applies the use of remote sensing data for extraction of bathymetry within a specified time range. Logarithmic and linear analytical methods are applied to Sentinel 2A and Landsat 8 imagery to retrieve bathymetry data. We intend to analyse the pattern of error in regards to depth and epoch. The collection of satellite image, the corresponding storage, and processing makes use of Google Earth Engine (GEE). The result shows that the root mean square error (RMSE) of depth is ranging from 1.6 m to 5.4 m from both sources of imageries. Better accuracy is obtained by applying logarithmic method to Landsat imagery.