Electronic Theses and Dissertation

Wilayah pesisir merupakan kawasan yang memiliki perubahan tutupan lahan yang tinggi akibat aktivitas manusia dan proses alami. penelitian ini bertujuan untuk menganalisis dan memetakan perubahan tutupan lahan di wilayah pesisir aceh barat khususnya kecamatan arongan lambalek dan kecamatan samatiga menggunakan citra sentinel-2 level-2a tahun 2017, 2021, dan 2025. metode yang digunakan adalah klasifikasi terbimbing (supervised classification) dengan algoritma maximum likelihood classification. variabel input yang digunakan terdiri atas band 2 (blue), band 3 (green), band 4 (red), band 8 (near infrared), serta indeks spektral normalized difference vegetation index (ndvi) dan normalized difference water index (ndwi). hasil klasifikasi divalidasi menggunakan confusion matrix untuk memperoleh nilai overall accuracy dan koefisien kappa. hasil penelitian menunjukkan bahwa wilayah penelitian terdiri atas tujuh kelas tutupan lahan, yaitu hutan, lahan basah, lahan pertanian, lahan terbuka, padang rumput, permukiman, dan badan air. selama periode 2017, 2021, dan 2025 terjadi perubahan tutupan lahan yang ditandai dengan penurunan luas hutan dan lahan terbuka, peningkatan lahan pertanian, padang rumput dan permukiman, serta perubahan yang berfluktuasi pada lahan basah dan badan air. hasil uji akurasi menunjukkan nilai overall accuracy sebesar 88% pada tahun 2017, 89% pada tahun 2021, dan 88% pada tahun 2025, dengan koefisien kappa masing-masing sebesar 0,85. kata kunci: tutupan lahan, wilayah pesisir, sentinel-2 level-2a, maximum likelihood classification, ndvi, ndwi

Coastal areas are regions that experience significant land cover changes due to both human activities and natural processes. This study aims to analyze and map land cover changes in the coastal area of Aceh Barat, specifically in Arongan Lambalek and Samatiga Districts, using Sentinel-2 Level-2A imagery acquired in 2017, 2021, and 2025. The study employed a supervised classification approach using the Maximum Likelihood Classification algorithm. The input variables consisted of band 2 (Blue), band 3 (Green), band 4 (Red), band 8 (Near Infrared), and the spectral indices Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI). The classification results were validated using a confusion matrix to determine the Overall Accuracy and Kappa coefficient. The results indicate that the study area comprises seven land cover classes: forest, wetlands, agricultural land, open land, grassland, settlements, and water bodies. During the 2017, 2021, and 2025 period, land cover changes were characterized by a decrease in forest and open land areas, an increase in agricultural land, grassland, and settlements, and fluctuating changes in wetlands and water bodies. The results of the accuracy test show an Overall Accuracy value of 88% in 2017, 89% in 2021, and 88% in 2025, with a Kappa coefficient of 0,85 each. Keywords: Land cover, coastal area, Sentinel-2 Level-2A, Maximum Likelihood Classification, NDVI, NDWI