Multi-Scenario Simulation and Prediction of Land Use Conflicts Based on the PLUS Model and Optimal Landscape Scale: Evidence from the Urumqi–Changji–Shihezi Urban Agglomeration, Northwest China

Examining the spatiotemporal evolution of land use conflicts, along with projecting their future spatial patterns across various development pathways, is of critical importance for assessing the regional land use conditions, refining spatial layout designs, and advancing sustainable development. Focusing on the Urumqi–Changji–Shihezi Urban Agglomeration (UCSUA) in the arid region of Northwest China, this study utilized land use data from four time periods between 1990 and 2020. To investigate the spatiotemporal evolution of land use spatial conflict potential at the grid level, this work employed an assessment framework devised from a landscape pattern standpoint. To pinpoint the spatial distribution of land use conflicts across divergent development trajectories, the Patch-generating Land Use Simulation (PLUS) model simulated land use patterns projected for 2030 under multiple scenarios. Throughout the entire study timeframe, grassland and unused land emerged as the predominant land use types, together comprising more than 70% of the total study region. Notable land use changes occurred in grassland, cultivated land, and construction land. Grassland area declined from 27,093.48 km2 to 24,457 km2, while cultivated land and construction land expanded from 10,336.92 km2 and 1110.11 km2 to 12,922.39 km2 and 1900.35 km2, respectively. In addition, multi-scenario simulations based on the PLUS model revealed that grassland and unused land generally declined, cultivated land expanded in all scenarios except the ecological protection scenario, and water bodies and construction land increased across all scenarios. Furthermore, from 1990 to 2020, the average potential land use spatial conflict index exhibited phased fluctuations. It declined from 0.296 to 0.264 during 1990–2000, remained stable from 2000 to 2010, and rebounded to 0.271 during 2010–2020 driven by rapid urbanization, agricultural expansion, and intensified land use competition. Finally, multi-scenario simulations revealed controllable units exceeding 70% across all scenarios. The ecological protection scenario minimized potential conflict intensity with the highest stably controllable proportion (65.27%), whereas natural development and economic priority scenarios showed elevated conflict levels, suggesting that ecological protection effectively reduces regional potential land use spatial conflict intensity and promotes coordinated development.