Spatio-temporal Dynamics and Driving Factors of Carbon Storage in the Mu Us Sandy Land: A Case Study in Wushen Banner

Advancements in sand control measures have significantly enhanced vegetation cover in the Mu Us sandy land with the growth in grassland and forest vegetation mainly stemming from the transformation of desert vegetation types. Since the 1990s, socioeconomic growth and ecological engineering initiatives have markedly altered land use patterns in the region, impacting carbon storage. Yet, accurately assessing carbon storage and its determinants remains challenging. This study focuses on Wushen Banner, a representative county in the Mu Us sandy land. Based on the land use data from 1990 to 2020, the PLUS-InVEST model was used to explore the spatial and temporal variation of carbon storage under four scenarios, i.e. natural development scenario (NE), sustainable coordinated development scenario (SC), ecological protection scenario (EP) and economic development scenario (ED), from 1990 to 2020 and in 2035, and with the help of geography-detector model and Log-mean Dez exponential method (LMDI) to explore the drivers of the changes of carbon storage. The results of the study show that: (1) Analysis over the past three decades indicates increases in cultivated, forest, construction, and watershed areas, while grassland and sand areas have declined. The expansion of forest areas has been the primary contributor to the 266.59×10⁴ ton increase in the ecosystem's carbon stock; (2) By 2035, among four forecast scenarios, the scenario of economic development exhibits the highest proportional increase in regional carbon storage, while the scenario of natural evolution sees the largest proportional decrease in carbon storage areas. (3) The geodetector model highlights DEM, NDVI, and transferred forest land area as key explanatory variables for changes in carbon stock, while the LMDI model confirms a positive correlation between transferred forest land area and economic development impacts on carbon stock. These findings underscore the importance of multi-scenario analysis in understanding the evolution of carbon stocks and in crafting land use strategies that support urban intensification and high-quality ecological growth, alongside an increase in carbon sinks.