A hybrid agent-based framework for modeling land use and land cover change in the gidabo watershed, Ethiopia

Abstract Land use and land cover (LULC) change is a major driver of ecosystem service shifts, hydrological processes, and biodiversity loss in Ethiopia’s Rift Valley. This study investigates historical LULC dynamics in the Gidabo watershed (2000–2024) and simulates future trajectories to 2080 using a hybrid modeling framework that integrates agent-based modeling, cellular automata, and multi-criteria evaluation. The model explicitly links human decision-making, spatial suitability, and policy scenarios, enabling dynamic representation of human–environment interactions. LULC classification achieved strong accuracy (80.83%–86.22%, Kappa 0.76–0.82), confirming the reliability of the ABCSMM framework. Results reveal a clear transition from tree-based systems toward agriculture and settlements, with agroforestry declining by nearly 50%, while farmland and built-up areas expanded substantially. Forest cover showed a non-linear pattern, increasing initially before declining under growing land pressure. These changes reflect the combined influence of accessibility, population growth and economic conditions. Future projections indicate non-linear land-use trajectories driven by interactions between agent behavior and spatial constraints. Farmland expands in the medium term before declining, while forest cover shows partial recovery under reduced pressure. Agent-based simulations highlight the role of household economics: financially stable conditions promote cropland expansion, whereas economic stress increases the likelihood of land degradation. Scenario analysis demonstrates that policy choices strongly shape land-use outcomes. Ecological protection scenarios enhance forest and grassland coverage while limiting urban expansion, whereas development-oriented scenarios accelerate the growth of farmland, agroforestry, and built-up areas. Overall, the findings show that LULC dynamics in the Gidabo watershed are governed by the interaction of socioeconomic drivers, environmental suitability, and policy interventions. The integrated modeling framework provides a useful tool for evaluating alternative land-use strategies and supports the design of policies that balance agricultural development, urban growth, and ecosystem conservation.