Land use assessment under dynamic evolution: Multi-objective optimization and multi-scenario simulation analysis.

The efficient use of land resources is key to achieving the dual goals of "carbon neutrality" and high-quality development while addressing the challenges of imbalance between ecological protection and economic development in river basins. This study combines remote sensing data with land use change modeling to generate maps of land use changes in the past and present, and by integrating the Grey multi-objective optimization-Patch-level land use simulation (GMOP-PLUS) model with the Coupled Model Intercomparison Project 6 (CMIP 6) development pathways, it has defined 12 target scenarios to simulate and predict the trends of changes over the next 30 years, providing a basis for formulating future land management policies. We found that from 1980 to 2020, grassland dominated (41%), with the largest increase in built-up land, and a decrease in unused land and cropland by 5.22% and 4.29%, respectively. After 2000, the complexity of land use structure has been increasing annually (1.41-1.45), especially in the central and western regions. In the future, the SSP126 scenario is more aligned with the achievement of sustainable development goals in the basin: woodland will expand rapidly, reaching 12.11% under the Maximizing Carbon Storage (MCS) target; grassland will shrink year by year, with the highest (46.20%) under the Maximizing Economic Development (MED)-SSP126 target scenario, and both face the risk of transferring to unused land. In the SSP585 scenario, grassland under the MCS and Maximizing Ecological Value (MEV) targets will mostly transform into woodland (39.06%) and unused land (39.28%). It is predicted that this trend will lead to the instability of terrestrial ecosystems, and optimizing land spatial configuration can help reduce trade-offs between different ecosystem services. The information obtained from the simulation indicates that the modeling method is also applicable to other types of regions.