[Spatiotemporal Simulation and Driving Force Analysis of Carbon Storage in Heihe River Basin Based on LUCC Changes]

Exploring the land use characteristics, spatial distribution patterns of carbon storage, and driving factors of spatial differentiation in the Heihe River Basin research area, as the second largest inland river basin in the northwest region of China, under future scenarios is crucial for guiding land use planning and regional coordinated development. Using land use data from Heihe River Basin in 2000, 2010, and 2020, the PLUS-InVEST-Geodetector model combined with 15 driving factors was used to simulate and predict the spatial distribution pattern of land use and carbon storage in the River Basin in 2030 under four scenarios: natural development, urban development, ecological protection, and cultivated land protection. The results showed that: ① From 2000 to 2020, cropland area increased by 574.00 km², forest land increased by 2 860.00 km², and water bodies increased by 488.00 km². The proportion of grassland decreased from 22.42% to 20.89%, with a decrease in the area of 2 214.00 km², and the area of unused land decreased by 2.25%. Grassland mainly converted to forest land and unused land, and unused land mainly converted to grassland and cropland. ② Carbon storage in Heihe River Basin increased by 0.213×10⁹ t over 20 years, mainly due to 70.17% of unused land being transferred to grassland. ③ Compared to that in 2020, the carbon storage under all four scenarios for 2030 showed an increasing trend, with carbon storage values of 6.653×10⁹, 6.654×10⁹, 6.730×10⁹, and 6.667×10⁹ t, respectively. The growth rates from high to low were in the order of ecological protection scenario (4.42%) > cropland protection scenario (3.44%) > urban development scenario (3.24%) > natural development scenario (3.23%). ④ Land use type, NDVI, annual precipitation, and DEM were the main driving factors affecting the spatial differentiation of regional carbon stocks. The interaction type was dominated by two-factor enhancement, in which the interaction detection between land use type and annual precipitation had the greatest influence, with a q value of 0.99. The research results provide data support for optimizing the land structure and promoting sustainable development in the Heihe River Basin.