Analisis Perubahan Tutupan Lahan Akibat Letusan Gunung Semeru Tahun 2021 dengan Algoritma RF dan SVM (Studi Kasus : Kecamatan Pronojiwo, Kabupaten Lumajang)

Azzahra, Erika (2023) Analisis Perubahan Tutupan Lahan Akibat Letusan Gunung Semeru Tahun 2021 dengan Algoritma RF dan SVM (Studi Kasus : Kecamatan Pronojiwo, Kabupaten Lumajang). Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 03311940000001_Undergraduated_Thesis.pdf] Text
03311940000001_Undergraduated_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2025.

Download (6MB) | Request a copy

Abstract

4 Desember 2021 terjadi erupsi Semeru yang membawa perubahan lingkungan termasuk perubahan tutupan lahan di daerah letusan, salah satunya di Kecamatan Pronojiwo, Kabupaten Lumajang. Informasi mengenai perubahan tutupan lahan diperlukan untuk para pengambil kebijakan atau pemangku kepentingan terkait untuk pengelolaan sumber daya lahan berkelanjutan. Pemantauan perubahan tutupan lahan dapat dilakukan menggunakan teknologi penginderaan jauh dengan data multi-temporal Sentinel-2 Level 2A dan menggunakan berbagai cara, salah satunya dengan metode supervised machine learning dengan algoritma Random Forest (RF) dan Support Vector Machine (SVM). Berdasarkan hasil klasifikasi akan diperoleh metode yang memiliki akurasi tinggi dan cocok digunakan untuk pemantauan perubahan tutupan lahan di area Semeru. Klasifikasi tutupan lahan yang diterapkan pada penelitian ini terdiri enam kelas tutupan lahan, yaitu sungai, lahan terbuka, area terbangun, sawah, hutan lahan kering, dan perkebunan. Berdasarkan hasil uji akurasi dengan menggunakan Confussion matrix menunjukkan bahwa algoritma Support Vector Machine (SVM) merupakan algoritma terbaik dibandingkan dengan Random Forest (RF) dengan nilai overall accuracy berturut-turut sebesar 88,11% dan 85,02%. Algoritma Support Vector Machine (SVM) digunakan untuk mengetahui perubahan tutupan lahan akibat letusan Gunung Semeru tahun 2021. Hasil penelitian menunjukkan terjadi penambahan luas pada kelas sungai, lahan terbuka, dan hutan lahan kering yaitu sebesar 2,293 km2, 6,381 km2, 6,49 km2. Sedangkan untuk kelas sawah, perkebunan, dan area terbangun mengalami penurunan luas sebesar 12,863 km2, 2,495 km2, and 0,445 km2. Terjadinya perubahan luasan tutupan lahan yang signifikan dapat dikarenakan adanya aliran lahar dan lava akibat letusan Gunung Semeru dan kesalahan klasifikasi yang disebabkan oleh beberapa faktor seperti nilai reflectance tiap piksel yang hampir sama dan lainnya. Hasil tersebut dapat dijadikan pertimbangan bagi instansi terkait dalam memperhatikan faktor-faktor yang dapat mempengaruhi hasil klasifikasi tutupan lahan yang diperoleh pada Kecamatan Pronojiwo, Kabupaten Lumajang.
=================================================================================================================================
December 4, 2021, brought about environmental changes including changes in land cover in the explosion area, Pronojiwo Sub-District, Lumajang District. Information abaout land cover changes is needed for policy makers or relevant stakeholders for sustainable land resource management. Monitoring land cover change can be done using remote sensing technology with Sentinel-2 multi-temporal data. There are many approaches to classifying land cover using satellite image data, including using supervised machine learning method with the Random Forest (RF) and Support Vector Machine (SVM) algorithms. Based on the classification results, a method that has high accuracy and is suitable for monitoring changes in land cover in the Semeru area will be obtained. The land cover classification used in this study consists of six land cover classes, namely rivers, open land, built-up areas, farm land, dryland forest, and estate. Based on the results of the accuracy test using the Confussion matrix, it shows that the Support Vector Machine (SVM) algorithm is the best algorithm compared to the Random Forest (RF) with overall accuracy values of 88,11% and 85,02%, respectively. The Support Vector Machine (SVM) algorithm was used to determine changes in land cover due to the eruption of Mount Semeru in 2021. The results showed that there was an increase in the area of the rive class, barren land, and dryland forest, namely 2,293 km2, 6,381 km2, 6,49 km2. Meanwhile, the farm land, estate, and built up area experienced a decrease area of 12,863 km2, 2,495 km2, and 0,445 km2. The occurrence of significant changes in land cover area can be due to lava flows and lava due to the eruption of Mount Semeru and misclassification caused by several factors such as the reflectance value of each pixel which is almost the same and others. These results can be used as a consideration for related agencies in paying attention to factors that can influence the land cover classification results obtained in Pronojiwo District, Lumajang Regency.

Item Type: Thesis (Other)
Uncontrolled Keywords: Erupsi Semeru, Perubahan Tutupan Lahan, Random Forest, Support Vector Machine, Sentinel-2, Semeru Eruption, Land Cover Changes, Random Forest
Subjects: G Geography. Anthropology. Recreation > G Geography (General) > G70.217 Geospatial data
G Geography. Anthropology. Recreation > G Geography (General) > G70.5.I4 Remote sensing
Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions: Faculty of Civil, Environmental, and Geo Engineering > Geomatics Engineering > 29202-(S1) Undergraduate Theses
Depositing User: Erika Azzahra
Date Deposited: 05 Sep 2023 05:17
Last Modified: 05 Sep 2023 05:17
URI: http://repository.its.ac.id/id/eprint/101259

Actions (login required)

View Item View Item