Answers are grounded in this paper and cite their sources.
Land-use and land-cover change (LUCC) strongly influences terrestrial carbon sequestration and climate mitigation, making accurate future simulation essential for sustainable land management. However, many existing studies separately address LUCC prediction and carbon assessment, with limited use of interpretable multi-scenario frameworks. This study developed an integrated framework combining Random Forest (RF) machine learning, CA–Markov spatial simulation, and the InVEST Carbon Storage and Sequestration model to project LUCC dynamics and associated carbon changes from 2024 to 2050 in metropolitan Prague, Czech Republic. RF-derived transition potentials based on topographic, climatic, and anthropogenic drivers were incorporated into the CA–Markov allocation process to simulate Business-as-Usual, Ecological Optimization, and Urban Growth scenarios. SHAP analysis revealed a class-specific gradient in driver dominance: anthropogenic accessibility factors dominated transitions in six of seven land-cover classes (from 51.0% for water to 75.7% for built-up areas), whereas only hydrologically sensitive flooded vegetation was primarily controlled by elevation and climate. The Ecological Optimization scenario produced the greatest gains in vegetation cover and carbon storage, while the Urban Growth scenario resulted in the largest conversion of agricultural land to built-up areas. InVEST analysis identified persistent high-carbon areas and an overall positive carbon sequestration trend associated with vegetation recovery and forest expansion. The proposed framework provides an interpretable and reproducible approach for linking LUCC dynamics with ecosystem carbon storage and supports sustainable land-use planning and climate mitigation strategies in metropolitan regions.
No takes yet. Share an insight, caveat, or question.
Gholamnia et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: