Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/162163| Title: | Monitoring Perubahan Tutupan dan Penggunaan Lahan Berbasis Machine Learning di Kesatuan Hidrologis Gambut Rokan Hilir, Riau |
| Other Titles: | Monitoring of Land Use and Land Cover Changes Based on Machine Learning in Rokan Hilir Peat Hydrological Unit, Riau. |
| Authors: | Putra, Erianto Indra Saputri, Hanum Resti |
| Issue Date: | 2025 |
| Publisher: | IPB University |
| Abstract: | Kabupaten Rokan Hilir adalah salah satu kabupaten di Riau yang mengalami
konversi lahan gambut yang tinggi. Penelitian ini bertujuan untuk mengetahui
tutupan dan penggunaan lahan apa yang banyak mengalami perubahan dan
mengestimasi luas tutupan dan penggunaan lahan yang paling banyak mengalami
perubahan di Kesatuan Hidrologis Gambut (KHG) Rokan Hilir pada tahun 2005
hingga 2023. Algoritma indeks yang digunakan sebanyak 12 indeks. Kelas tutupan
dan penggunaan lahan dibagi menjadi 6 kelas, yaitu badan air, lahan terbangun,
lahan terbuka, perkebunan sawit, hutan alam, dan hutan tanaman. Machine learning
yang digunakan adalah RF (Random Forest). Penelitian ini menunjukkan bahwa
pada periode 2005-2023 tutupan lahan yang paling banyak mengalami penurunan
luasan adalah hutan alam (119.186,3 ha) dan hutan tanaman (72.135,6 ha).
Penggunaan lahan yang paling banyak mengalami kenaikan luasan yaitu
perkebunan sawit (146.384,7 ha) dan lahan terbangun (22.101,2 ha). Rokan Hilir Regency is one of the regencies in Riau that has experienced a high rate of peatland conversion. This study aims to identify the land cover and land use types that have undergone the most significant changes and to estimate the area of land cover and land use with the greatest changes in the Rokan Hilir Peat Hydrological Unit (PHU) from 2005 to 2023. A total of 12 algorithms index were used in the analysis. Land cover and land use were classified into 6 categories e.g water bodies, built-up land, bare land, oil palm plantations, natural forests, and forest plantations. The machine learning method used was Random Forest (RF). The results of the study show that between 2005 and 2023, the land cover types that experienced the most area decrease were natural forests (119,186.3 ha) followed with forest plantations (72,135.6 ha). The land use types that showed the most area increase were oil palm plantations (146,384.7 ha) and built-up land (22,101.2 ha). |
| URI: | http://repository.ipb.ac.id/handle/123456789/162163 |
| Appears in Collections: | UT - Silviculture |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_E4401211018_2a3c9a0c12b1419fba04edb6a3d8ea96.pdf | Cover | 281.17 kB | Adobe PDF | View/Open |
| fulltext_E4401211018_e5a05bb54a964a8092ac6c039207581b.pdf Restricted Access | Fulltext | 1.42 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.