Forest fragmentation and land use land cover change analysis in Naqamte City Western Ethiopia using machine learning and landscape metrics
摘要
Rapid land use/land cover (LULC) change and forest fragmentation have become major environmental challenges in rapidly urbanizing regions of Ethiopia, particularly in Naqamte City, where urban expansion and agricultural pressure are intensifying. This study analyzed multi-temporal LULC dynamics, quantified forest fragmentation, and identified the major drivers of forest fragmentation in Naqamte City from 1991 to 2025 using machine learning techniques and landscape metrics. Landsat images from 1991, 2011, and 2025 were classified into five LULC classes (built-up, cropland, forest, grassland, and waterbody) using SVM, with accuracy assessed via confusion matrices, overall accuracy, and Kappa. LULC changes were analyzed through post-classification comparison. Forest fragmentation was quantified using FRAGSTATS 4.2 by analyzing patch, edge, perforated, and core forest categories. Socio-economic drivers were identified through key informant interviews, focus group discussions, and spatial overlay analysis. SVM, offer improved performance due to their ability to handle complex spectral signatures and high-dimensional feature spaces [