Abstract: As a nationally significant energy and chemical industry base, Shanxi Province plays a critical role in achieving China′s ‘Dual Carbon’ goals (carbon peaking and carbon neutrality). However, the spatiotemporal evolution and driving mechanisms of land use and ecosystem carbon storage at the provincial scale remain insufficiently understood. This study addresses this gap by integrating the InVEST model, the PLUS (Patch-generating Land Use Simulation) model, and Geodetector to analyze carbon storage dynamics in Shanxi Province from 2000 to 2023, predict changes under a natural development scenario (NDS) for 2030, and identify the key driving factors. The results reveal that: (1) From 2000 to 2023, forest area increased by 0.46%, contributing to an average annual carbon storage growth of 0.2%, while built-up land expanded by 105.70% and cropland decreased by 2.46%. (2) High carbon storage areas (>1.9 hundred million tons) were concentrated in Lüliang and Xinzhou, where NDVI and precipitation exert synergistic effects, while low-value areas (<0.7 hundred million tons) were found in highly urbanized basins such as the Taiyuan Basin. (3) Projections for 2030 indicate a 3.28% increase in forest area will further enhance carbon storage, but a 27.65% expansion of built-up land may intensify localized carbon loss. (4) Geodetector analysis identified NDVI (q-value: 0.383-0.567) and precipitation (q-value: 0.061-0.455) as the primary natural drivers, with their interaction showing a nonlinear enhancement effect (q-value>0.5). Among socioeconomic factors, population density and GDP were the dominant drivers of land use change. This study demonstrates that forest expansion is the main driver of carbon storage growth in Shanxi Province, effectively offsetting carbon losses from urbanization. The synergistic effects of natural and socioeconomic factors amplify spatial differentiation, highlighting the need for targeted mountain-basin management strategies to optimize carbon sink patterns. These findings provide a scientific basis for strengthening territorial carbon management and supporting regional carbon neutrality goals. Future research should incorporate dynamic carbon density data and policy-driven factors to improve simulation accuracy of carbon storage and support more targeted ecological restoration policies.