Multi-Decadal Land-Cover Transitions and Vegetation Condition in a Protected Montane Forest, Ethiopia (2000–2024)

Understanding spatiotemporal land-use/land-cover (LULC) dynamics and their effects on vegetation condition is critical for managing protected forests exposed to increasing anthropogenic pressure. This study assessed multi-decadal LULC change and associated vegetation dynamics in the Harenna Forest, Bale Mountains National Park, Ethiopia, from 2000 to 2024 using Landsat imagery, Random Forest classification, and NDVI analysis. Classification accuracy ranged from 80.9% to 87.2%, with Kappa values between 0.739 and 0.826, indicating substantial agreement across all years. Transition matrix analysis revealed progressive forest conversion primarily to smallholder agriculture and localized settlement expansion, reflecting cumulative edge encroachment within the protected landscape. NDVI patterns consistently differentiated LULC classes, with forests exhibiting significantly higher vegetation greenness than agriculture, settlement, and bare land (p < 0.001). These findings indicate that land-cover conversion is associated with measurable reductions in vegetation vigor, particularly along forest margins where transition intensity is highest. Although Harenna Forest remains largely intact, the observed edge-driven transitions demonstrate sustained anthropogenic pressure within the protected area. By integrating multi-decadal land-cover transition analysis with inferential NDVI assessment, this study provides a robust and scalable quantitative framework for evaluating vegetation condition responses to land-use change and supports adaptive forest management and long-term satellite-based monitoring in tropical montane forest systems. These results further emphasize the importance of integrating remote sensing tools into conservation planning frameworks.