Estimating bathymetry using sentinel-2a for shallow-waters mapping: A case study in Kelapa Dua Island

Abstract

Many algorithms have been developed to map shallow waters from satellite images, but there is no standard algorithm for all waters. The accuracy of the algorithms developed for each body of water has a different level of accuracy. This research aims to use the Stumpf and Lyzenga algorithms to estimate shallow water bathymetry mapping in Kelapa Dua Island. Shallow water bathymetry mapping with both algorithms uses a combination of blue-green bands from Sentinel-2A satellite image data. In addition to Sentinel-2A satellite image data, the data used in this research are depth, water clarity, and tide data. Satellite image processing includes pre-processing, algorithm application, and fit testing. The fit test through determination test and root mean square error analysis was used to see the level of fit of the shallow water bathymetry model. The analysis shows that the log-ratio transform algorithm tends to be better at mapping bathymetry in Kelapa Dua Island, Kepulauan Seribu with a coefficient of determination (R2) of 0.56 and an RMSE of 0.648, while the log-linear transform algorithm has a coefficient of determination (R2) of 0.54 and an RMSE of 0.742. The depth of shallow water on Kelapa Dua Island is in the range of 0 7 m depth.