Assessment of bathymetry using Landsat-8 Imagery in shallow water of Seberang Takir to Teluk Ketapang Coastal, Terengganu / Mohammad Shahrul Nizam Zaid

Hydrography is a geomatics field that contributes greatly to the charting of water depths such as rivers, oceans, and lakes. Furthermore, bathymetry is one of the hydrography products that determine the depth of water below the required datum reference in a lake, river, or ocean. The data collected during the survey will be evaluated to create a navigation map, seafloor profile, sea level rise predictions, ocean management, and so on. Thus, sonar, a traditional approach, and Airborne Laser Bathymetric (ALB), a new technology, are utilised to gather bathymetry information. However, those techniques are costly, time-consuming, and require maintenance services. As a result, utilising Satellite Derived Bathymetry (SDB) to measure depth using satellite imagery is a good approach. The study aims to create the SDB map using Landsat-8 imagery for the Seberang Takir to Teluk Ketapang coastal using an empirical and semi-analytical method. The objectives are to produce an SDB map using Landsat-8 through the Stumpf and Dierssen model with a bathymetry map for the study area for the year 2020 and to assess the accuracy assessment of the SDB and Bathymetry map. Following processing, the SDB map will be evaluated using the Root Mean Square Error (RMSE) as a statistical study to determine its accuracy. The RMSE from the final satellite determined bathymetry for empirical and semi-analytical methods is 0.797 and 0.811 respectively. Furthermore, with 10% of the ground data evaluated, the depth correlation between ground data and the SDB depth imagery is 0.930 for the empirical method and 0.927 for the semi-analytical method. Overall, the findings suggest that it is possible to compute depth for a large area using remote sensing without replacing the traditional method. For future research, it is recommended that high-resolution satellites such as Worldview or Pleiades be used along with a more detailed algorithm.