Estimasi Kedalaman Berbasis Transformasi Multiband Citra Sentinel-2 Menggunakan Metode Satellite Derived Bathymetry di Waduk Sermo, Kulon Progo Daerah Istimewa Yogyakarta

Abstract

Bathymetric mapping is essential for obtaining accurate information on water depth however, measurements using singlebeam echosounders (SBES) are relatively expensive and time consuming and have a limited detection range. Satellite derived bathymetry (SDB), which is based on multispectral imagery, offers a more efficient alternative for depth estimation. The objective of this study is to compare the results of multiband transformation using the blue, green and red bands, and to determine the optimal multiband transformation for depth estimation in the Sermo Reservoir through accuracy testing. This study employs a quantitative approach by applying the SDB method and the Stumpf algorithm using Sentinel-2 imagery and field data. Data processing included atmospheric correction, masking of non-water areas, image cropping, extraction of the blue (B2), green (B3) dan red (B4) and the application of multiband transformation, which were then calibrated against field depth data through regression analysis. The estimation results were evaluated using the Root Mean Square Error (RMSE) and the coefficient of determination (R²). The results of the study show that the transformation of the green band with the red band produced an RMSE of 7.040 m and R² of 0.4829 the blue band with green band produced an RMSE of 7.567 m and R² of 0.2427 while the blue band with the red band produced an RMSE of 7.899 m and R² of 0.0241. Based on the accuracy test results, the transformation of the green band with the red band had the lowest RMSE and the highest R² values, making it the most optimal transformation in this study.