Running summarizing tools on a new article

Abstract: Accurate Bathymetryc information in shallow coastal waters is critical for marine navigation, coastal zone management, habitat assessment, and environmental monitoring. Conventional Bathymetryc surveys are limited by high operational costs, restricted spatial coverage, and time-consuming fieldwork. To overcome these challenges, the present study assesses the potential of Satellite-Derived Bathymetry (SDB) using Sentinel-2 imagery as an alternative, comparing it with in situ Bathymetryc measurements obtained via Single Beam Echosounder (SBES) in the coastal waters of Dubibir, Situbondo, Indonesia. Bathymetryc data were collected with an SBES GPSMAP 585c, and Sentinel-2 multispectral imagery was processed to estimate water depth using a Random Forest (RF) machine learning model. Results indicate that SBES measurements reached a maximum depth of 53.34 m, while Sentinel-2-derived bathymetry captured depths up to 29.61 m. Model evaluation yielded a coefficient of determination (R²) of 0.83, a mean absolute error (MAE) of 1.90 m, and a root mean square error (RMSE) of 3.56 m, demonstrating strong predictive performance in shallow-water environments. However, the findings also show reduced SDB capability in deeper, optically complex waters, particularly where turbidity limits light penetration and weakens the satellite signal. Overall, Sentinel-2 imagery combined with the RF algorithm offers a practical, cost-effective, and spatially efficient solution for shallow-water Bathymetryc mapping, while SBES remains essential for validation and for representing deeper seabed conditions. This study advances the application of machine learning-based SDB for coastal mapping and provides a relevant approach for generating Bathymetryc data in data-limited coastal regions.