Article
Analisis Spasial Temporal Menggunakan Citra Landsat-8 dengan Metode Mahalanobis Distance Classification untuk Identifikasi Perubahan Penggunaan Lahan dan Tutupan Lahan
Jurnal Penelitian Pendidikan Geografi, Vol. 11 No. 3 (2026), hlm. 395–408
Abstrak
Changes in urban land cover and land use have widespread impacts on ecosystems, physical environmental conditions, and regional sustainability. This study aims to: 1) identify the spatio-temporal distribution and dynamics of land use change in Kendari City from 2015 to 2025; and 2) evaluate the mapping accuracy of land cover based on the Mahalanobis Distance Classification method. The analysis results for the 2015–2020 period indicate an increase in Forest area (+1,838.19 ha), Agricultural land (+2,181.13 ha), Water bodies (+150.87 ha), Aquaculture ponds (+180.00 ha), and Open land (+218.83 ha), accompanied by a decrease in Shrubland (-1,456.15 ha) and Built-up land (-3,112.87 ha) due to spectral overlap limitations. Entering the 2020–2025 period, a massive urban sprawl phenomenon occurred, marked by a sharp surge in Built-up land of +4,356.96 ha and Open land (+130.57 ha). This rapid growth significantly converted Agricultural land (-2,893.17 ha), Shrubland (-597.92 ha), Forest (-530.57 ha), Aquaculture ponds (-262,05 ha), and Water bodies (-203.82 ha). The accuracy assessment of the Mahalanobis Distance classification yielded an Overall Accuracy of 82.86% and a Kappa Coefficient of 80.26% (good category). Land use trends in Kendari City from 2015 to 2025 were dominated by the conversion of non-built-up and productive land into built-up areas, demanding stricter enforcement of spatial planning policies.