Electronic Theses and Dissertation

Indonesia sebagai negara maritim memiliki wilayah perairan yang luas sehingga membutuhkan data batimetri yang akurat untuk mendukung pengelolaan wilayah pesisir, navigasi laut, dan mitigasi bencana. Namun, ketersediaan data batimetri di perairan dangkal seperti di Alue Naga masih terbatas. Penelitian ini bertujuan untuk mengkaji hubungan antara data reflektansi citra satelit Sentinel-2 dengan data batimetri hasil pengukuran hidroakustik menggunakan Single Beam Echo Sounder (SBES) di perairan Alue Naga, Banda Aceh. Pengolahan data citra dilakukan melalui platform Google Earth Engine (GEE), dengan pemanfaatan band-band optik Sentinel-2 dan perhitungan indeks Normalized Difference Water Index (NDWI). Nilai reflektansi dari citra satelit kemudian dianalisis hubungannya terhadap data kedalaman hasil pengukuran lapangan melalui regresi linier dan regresi berganda, baik dalam bentuk data mentah maupun transformasi logaritma. Hasil penelitian menunjukkan bahwa band 2 memiliki hubungan paling kuat terhadap batimetri dengan nilai R Squared (R²) sebesar 0,371. Model regresi berganda dengan transformasi logaritma menghasilkan akurasi terbaik dengan nilai R² tertinggi dan Root Mean Square Error (RMSE) terendah. Penelitian ini menunjukkan bahwa citra Sentinel-2 memiliki potensi untuk digunakan sebagai metode alternatif dalam pemetaan batimetri secara efisien dan ekonomis, khususnya di wilayah yang sulit dijangkau oleh survei langsung.
Kata kunci: batimetri, sentinel-2, google earth engine, regresi, reflektansi

As a maritime country, Indonesia has a vast territorial sea that requires accurate bathymetric data to support coastal zone management, maritime navigation, and disaster mitigation. However, the availability of bathymetric data in shallow waters such as Alue Naga is still limited. This study aims to examine the relationship between Sentinel-2 satellite image reflectance data and bathymetric data obtained from hydroacoustic measurements using a Single Beam Echo Sounder (SBES) in the waters of Alue Naga, Banda Aceh. Image data processing was carried out using the Google Earth Engine (GEE) platform, utilizing Sentinel-2 optical bands and calculating the Normalized Difference Water Index (NDWI) . The reflectance values from the satellite images were then analyzed for their relationship with field-measured depth data through linear and multiple regression, both in raw data form and logarithmic transformation. The results showed that band 2 had the strongest correlation with bathymetry, with an R Squared (R²) value of 0.371. The multiple regression model with logarithmic transformation yielded the best accuracy, with the highest R² value and lowest Root Mean Square Error (RMSE). This study demonstrates that Sentinel-2 imagery has the potential to serve as an alternative method for efficient and cost-effective bathymetric mapping, particularly in areas difficult to access via direct surveys. Keywords: bathymetry, sentinel-2, google earth engine, regression, reflectance