Combining remote sensing and local perspectives uncovers divergent ecosystem service change in the Lower Omo, Ethiopia

Le résumé fourni par la source

Understanding ecosystem service (ES) change in African drylands is critical as land use transformation increasingly affects ecosystems and the communities that depend on them. However, most ES assessments overlook or homogenise local perspectives, limiting their accuracy and relevance. This study develops an approach that combines Landsat-based land cover mapping with participatory assessment to quantify changes in provisioning ES in the Lower Omo, a culturally diverse region of southwest Ethiopia, over the past two decades. The method incorporates the perspectives of multiple beneficiary groups, enabling social and spatial differentiation of ES change across the landscape. Results highlight the importance of natural land covers (shrubland, grassland, and riverine forest) for local provisioning services and show substantial ES declines associated with the expansion of monoculture agriculture linked to large-scale land acquisitions. Disaggregated ES maps reveal that changes that appear uniform in satellite-derived land cover data are experienced unevenly across communities due to differences in livelihoods, values, and exposure to land use change. For example, bush encroachment increases ES value for some groups by supporting cultivation, wild plant harvesting, and wood products, while reducing grazing resources for pastoral communities. Similarly, commercial monoculture expansion produces uneven outcomes, with some communities losing access to high-value ES while others are less affected. These findings demonstrate that bridging participatory perspectives and remote sensing improves ES assessments by revealing distributional impacts often obscured by top-down analyses. The approach offers a replicable framework for producing socially differentiated, landscape-scale ES assessments to inform sustainable land management in dryland regions.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.