Widiyanti, Lilis (2026) Analisis Perubahan Tutupan Lahan Sebelum Dan Sesudah Banjir Bandang Berbasis Karakteristik Backscatter Menggunakan Citra Sentinel-1 (Studi Kasus: Kabupaten Agam, Sumatera Barat). Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5016221030-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5016221030-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (10MB) | Request a copy |
Abstract
Banjir bandang yang terjadi di Kabupaten Agam, Sumatera Barat pada tahun 2025 menyebabkan perubahan tutupan lahan yang signifikan di wilayah terdampak. Pemantauan perubahan ini menggunakan citra optik terkendala oleh tutupan awan yang tebal di wilayah tropis, sehingga citra Synthetic Aperture Radar (SAR) Sentinel-1 digunakan sebagai alternatif yang tidak bergantung pada kondisi atmosfer. Penelitian ini menganalisis perubahan tutupan lahan sebelum (22 November 2025) dan sesudah banjir bandang (9 Januari 2026) di Kabupaten Agam menggunakan nilai backscatter polarisasi VV dan VH serta fitur tekstur Gray Level Co-occurrence Matrix (GLCM) untuk klasifikasi algoritma Random Forest. Komposit RGB dibentuk dari kombinasi kanal polarisasi sebagai alat bantu visualisasi pola hamburan antar kelas tutupan lahan. Empat kelas tutupan lahan yang diklasifikasikan adalah vegetasi, badan air, permukiman, dan lahan terbuka. Pra-pengolahan citra meliputi kalibrasi radiometrik, speckle filtering, dan koreksi geometrik, dilanjutkan dengan ekstraksi nilai backscatter dan enam parameter GLCM. Evaluasi akurasi menggunakan confusion matrix menghasilkan Overall Accuracy 82,50% (KA: 0,767) untuk periode sebelum banjir dan 80,00% (KA: 0,733) untuk periode sesudah banjir. Hasil analisis menunjukkan penurunan lahan terbuka sebesar 17.726,12 ha dan peningkatan badan air sebesar 5.547,88 ha, diikuti peningkatan vegetasi sebesar 14.709,49 ha dan penurunan permukiman sebesar 2.271,05 ha. Perubahan ini mengindikasikan bahwa sebagian lahan terbuka berubah menjadi area tergenang maupun kembali tertutup vegetasi akibat perbedaan musim tanam antara kedua waktu akuisisi citra, dan informasi spasial yang dihasilkan dapat digunakan sebagai dasar perencanaan mitigasi dan rehabilitasi lingkungan di Kabupaten Agam.
===============================================================================================================================================
The flash flood that occurred in Agam Regency, West Sumatra in 2025 caused significant land cover changes in the affected area. Monitoring these changes using optical imagery is constrained by thick cloud cover in tropical regions, making Sentinel-1 Synthetic Aperture Radar (SAR) imagery a suitable alternative that is independent of atmospheric conditions. This study analyzes land cover changes before (November 22, 2025) and after the flash flood (January 9, 2026) in Agam Regency using VV and VH polarization backscatter values and Gray Level Co-occurrence Matrix (GLCM) texture features for Random Forest classification. An RGB composite was constructed from polarization channel combinations as a visual aid for identifying backscatter patterns among land cover classes. Four land cover classes were classified: vegetation, water body, settlement, and open land. Image preprocessing included radiometric calibration, speckle filtering, and geometric correction, followed by extraction of backscatter values and six GLCM parameters. Accuracy evaluation using a confusion matrix yielded an Overall Accuracy of 82.50% (Kappa: 0.767) for the pre-flood period and 80.00% (Kappa: 0.733) for the post-flood period. The analysis reveals a decrease in open land of 17,726.12 ha and an increase in water body of 5,547.88 ha, accompanied by an increase in vegetation of 14,709.49 ha and a decrease in settlement area of 2,271.05 ha. These changes indicate that part of the open land converted into inundated zones or regrew with seasonal vegetation due to the difference in acquisition timing between the two images, and the spatial information produced can serve as a basis for mitigation planning and environmental rehabilitation in Agam Regency.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Banjir Bandang, Backscatter, GLCM, Random Forest, Sentinel-1, Kabupaten Agam. Flash Flood, Backscatter, GLCM, Random Forest, Sentinel-1, Agam Regency. |
| Subjects: | G Geography. Anthropology. Recreation > GA Mathematical geography. Cartography > GA102.4.R44 Cartography--Remote sensing |
| Divisions: | Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geomatics Engineering > 29202-(S1) Undergraduate Thesis |
| Depositing User: | Lilis Widiyanti |
| Date Deposited: | 24 Jul 2026 07:07 |
| Last Modified: | 24 Jul 2026 07:07 |
| URI: | http://repository.its.ac.id/id/eprint/137920 |
Actions (login required)
 |
View Item |