Analisis Spasial Sebaran Mangrove Tahun 2025 Menggunakan Citra Satelit Sentinel-1 dan Sentinel-2 dengan Algoritma Random Forest (Studi Kasus: Kabupaten Sampang)

Zanianti, Priska Rahma (2026) Analisis Spasial Sebaran Mangrove Tahun 2025 Menggunakan Citra Satelit Sentinel-1 dan Sentinel-2 dengan Algoritma Random Forest (Studi Kasus: Kabupaten Sampang). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5016221073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (9MB) | Request a copy

Abstract

Ekosistem mangrove berperan penting dalam menjaga keseimbangan ekologi pesisir, melindungi pantai dari abrasi, serta menjadi habitat berbagai biota laut. Berdasarkan Peta Mangrove Nasional 2023, Kabupaten Sampang merupakan salah satu wilayah di Pulau Madura yang memiliki hutan mangrove dengan luas sekitar 723,4085 hektare. Namun, kondisinya terus mengalami penurunan akibat alih fungsi lahan menjadi kawasan pembangunan, tambak udang, dan tambak garam. Di sisi lain, pemerintah juga terus mendorong berbagai program rehabilitasi dan pengembangan kawasan mangrove sebagai upaya konservasi untuk memulihkan ekosistem yang terdegradasi. Oleh karena itu, penelitian ini bertujuan mengetahui luasan dan sebaran spasial vegetasi mangrove tahun 2025. Penelitian ini menggunakan data citra satelit Sentinel-1 dan Sentinel-2. Metode yang digunakan adalah algoritma Random Forest. Pemrosesan citra dilakukan melalui software SNAP. Sentinel-1 menyediakan informasi yang tidak terpengaruh tutupan awan, sedangkan Sentinel-2 memberikan informasi spektral beragam. Hasil penelitian ini menunjukkan bahwa luas sebaran mangrove tahun 2025 yang terdeteksi Sentinel-1 sebesar 841,8602 ha, Sentinel-2 sebesar 1.082,4981 ha, dan penggabungan kedua citra sebesar 1.511,2128 ha. Sebaran spasial vegetasi mangrove berada di sepanjang garis pantai, alur sungai, dan di sekitar tambak. Penggabungan Sentinel-1 dan Sentinel-2 memiliki akurasi terbaik dibandingkan penggunaan citra tunggal. Nilai overall accuracy yang diperoleh sebesar 94,2857% dan Kappa accuracy sebesar 0,8750 (peluang akurasi sangat baik). Sementara itu, pada citra Sentinel-1 diperoleh nilai overall accuracy sebesar 80% dan Kappa accuracy sebesar 0,7000 (peluang akurasi baik). Pada citra Sentinel-2 diperoleh nilai overall accuracy sebesar 91,4286% dan Kappa accuracy sebesar 0,8714 (peluang akurasi sangat baik). Hasil tersebut mendukung hipotesis bahwa penggabungan Sentinel-1 dan Sentinel-2 menunjukkan hasil yang lebih baik.
===============================================================================================================================
Mangrove ecosystems play a vital role in maintaining coastal ecological balance, protecting shorelines from erosion, and providing a habitat for various marine species According to the 2023 National Mangrove Map, Sampang Regency is one of the areas on Madura Island that has mangrove forests covering an area of approximately 723.4085 hectares. However, their condition continues to deteriorate due to land conversion for development, shrimp farms, and salt ponds. On the other hand, the government continues to promote various mangrove rehabilitation and development programs as conservation efforts to restore degraded ecosystems. Therefore, this study aims to determine the extent and spatial distribution of mangrove vegetation in 2025. This study utilized Sentinel-1 and Sentinel-2 satellite imagery. The method employed was the Random Forest algorithm. Image processing was performed using SNAP software. Sentinel-1 provides information unaffected by cloud cover, while Sentinel-2 offers diverse spectral information. The results of this study show that the mangrove distribution area in 2025 detected by Sentinel-1 is 841.8602 ha, by Sentinel-2 is 1,082.4981 ha, and the combination of both images is 1,511.2128 ha. The spatial distribution of mangrove vegetation is along the coastline, river channels, and around fish ponds. The combination of Sentinel-1 and Sentinel-2 has the best accuracy compared to the use of a single image. The overall accuracy obtained was 94.2857%, and the Kappa accuracy was 0.8750 (very good accuracy). Meanwhile, the Sentinel-1 imagery yielded an overall accuracy of 80% and a Kappa accuracy of 0.7000 (indicating good accuracy). For Sentinel-2 imagery, the overall accuracy was 91.4286% and the Kappa accuracy was 0.8714 (indicating very good accuracy). These results support the hypothesis that the combination of Sentinel-1 and Sentinel-2 yields better results.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Mangrove, Random Forest, Sampang, Sentinel-1, Sentinel-2 Mangrove, Random Forest, Sampang, Sentinel-1, Sentinel-2
Subjects:	G Geography. Anthropology. Recreation > G Geography (General) > G70.212 ArcGIS. Geographic information systems. G Geography. Anthropology. Recreation > G Geography (General) > G70.217 Geospatial data G Geography. Anthropology. Recreation > G Geography (General) > G70.5.I4 Remote sensing
Divisions:	Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis
Depositing User:	Priska Rahma Zanianti
Date Deposited:	19 Jun 2026 06:10
Last Modified:	19 Jun 2026 06:10
URI:	http://repository.its.ac.id/id/eprint/133933

Actions (login required)

View Item