Accuracy of Shallow Water Bathymetry Mapping Based on Landsat 8 Satellite Imagery Data In Prigi Bay, Trenggalek Regency

Abstract

Bathymetry plays a crucial role in supporting navigation routes within Indonesian waters. Given the highly dynamic nature of the marine environment, continuous bathymetric mapping is essential. Particularly in Prigi Bay, which holds considerable potential in terms of natural resources, tourism, and serves as the location of the Nusantara Fisheries Port (PPN Prigi). Conventional methods such as echosounders still face limitations in terms of spatial coverage and operational efficiency. A more advanced alternative is Satellite Derived Bathymetry (SDB), which utilizes satellite imagery combined with digital processing algorithms. This study employs Landsat 8 imagery to map seafloor depth and topography using the Stumpf algorithm. The analysis involves comparing band ratio outputs from the Landsat 8 imagery with in-situ depth measurements to evaluate the accuracy of depth estimation. Results indicate that actual water depths in Prigi Bay range from 4.92 m to 67.61 m, while estimations derived from Landsat 8 imagery range from 7.05 m to 57.10 m. The regression model between Landsat 8-derived depths and in-situ measurements produced a coefficient of determination (R²) of 0.6287, a correlation coefficient (r) of 0.792, a Root Mean Square Error (RMSE) of 12.697, and a Normalized Mean Absolute Error (NMAE) of 28.48, indicating a reasonably reliable level of accuracy for satellite-based bathymetric estimations in coastal waters.