Bathymetry, an extended version of topographic mapping is considered as important area of studies as it is useful for shipping traffic safety, knowing sea bed geology, aquatic plants and other hazardous features which relies on accurate depth measurements regularly. Since the introduction of remote sensing using satellite imageries, bathymetry came up as an application with great possibilities and challenges which constantly kept on improving over the years. Also, satellite derived bathymetry is giving under water information very economically compared to echo soundings or airborne LIDAR data. This thesis is about assessing different possible ways to estimate bathymetry, starting from correction methods for atmospheric and sun glint effects, and hence calculating depth using some most acceptable methods based on literature. Four sensor datasets i.e. Landsat 8, Sentinel-2, LISS-4 and Worldview-2 having increasing spatial resolution respectively, were used over Chidiya Tapu (South Andaman) and Mumbai region. These areas were chosen to compare bathymetry over two contrasting places, Mumbai being highly turbid and human affected area whereas Chidiya Tapu being more natural and less crowded. All possible combinations were tried and statistically analysed to get the best possible methodology for particular study area. Also, how different datasets were performing based on their properties and study area on which they were applied is studied. Validation was done using insitu data and nautical charts.
Keywords: Satellite derived bathymetry, echo sounding, Automatic depth estimation, spatial resolution