[Multi-scenario Optimization and Effect Assessment of Land Use in Southern Jiangsu Based on "Potential-constraint"...

Land use conflicts are becoming increasingly severe during rapid urbanization processes. Addressing the limitations of existing land use optimization research， including insufficient multi-objective coordination mechanisms and separation between structural and spatial layout optimization， this study constructs a "potential-constraint" collaborative optimization framework and develops a Multi-Objective Spatial Genetic Algorithm （MOSGA）. The framework systematically integrates three potential objectives （net carbon sink， economic development， and food production） with three constraint objectives （land use suitability， compatibility， and transition costs）. Taking Southern Jiangsu as a case study， four development scenarios-low-carbon， economic， food security， and coordinated-were established to comparatively analyze land use optimization results for 2030 and 2060， achieving collaborative optimization allocation of land use structure and spatial layout at the grid scale. The MOSGA algorithm integrates genetic algorithms with NSGA-II through non-dominated sorting， crowding distance calculation， and patch-based crossover and mutation operations. The potential objectives are quantified using machine learning models， while constraint objectives are evaluated through land use suitability assessment， compatibility analysis， and transition cost evaluation. The optimization process involves weight adjustment across different scenarios and incorporates spatial constraints including permanent basic farmland and ecological protection red lines. The findings reveal： ① Land use changes in Southern Jiangsu from 2000-2020 were characterized by rapid urban land expansion and continuous farmland reduction. ② Through the MOSGA algorithm， potential objectives showed significant improvements， and constraint objectives achieved stable enhancement across all scenarios. ③ Four land use optimization scenarios presented differentiated development pathways： By 2030 and 2060， the low-carbon scenario showed net carbon sink improvements of 6.62% and 13.24%， respectively， the economic scenario achieved economic growth of 29.17% and 156.16%， the food security scenario demonstrated food production initially declining by 13.13% then increasing by 6.25%， while the coordinated scenario emphasized multi-objective balance. ④ Multi-scenario land use optimization demonstrated that rural residential land reduction released development space， ecological land agglomeration enhanced environmental efficiency， and construction land intensification degree continued to improve. This study provides new technical paradigms and decision support for multi-objective collaborative optimization of land use in rapid urbanization regions.