Mitigating carbon loss in China's arid and semi-arid regions through optimization and management under CMIP6 scenarios

Carbon storage (CS) in arid and semi-arid regions (ASR) constitutes a key part of the global terrestrial carbon pool but remains highly sensitive to climate change. Optimizing the spatial distribution of carbon sink services is essential for maximizing their sequestration potential. Using arid and semi-arid regions of China (CASR) as a case study, we applied the InVEST model to quantify CS patterns in 2020 and 2030 under CMIP6 scenarios, examined the contributions of environmental drivers, and optimized spatial distribution with a Bayesian belief (BBN). Results show that CS decreases from east to west, with grasslands contributing over 50% of the total CS. Unused land stores more carbon than forestland and should not be overlooked. Under SSP126 scenario, carbon sink potential is projected to increase, while under SSP245 and SSP585 scenarios, CS is expected to decline by 1.41×10⁷ t and 2.84×10⁷ t, respectively. CS is primarily influenced by precipitation and elevation, but the influence weight of socioeconomic factors gradually increases under the SSP245 and SSP585 scenarios. The region was divided into four optimized subsets based on carbon sink potential—ecological conservation, ecological restoration, cropland optimization, and desert optimization—with zoned management recommended for their carbon pools. After optimization, the ecological conservation subset was projected to cover 4531×10⁴ ha, 4223×10⁴ ha, and 4132×10⁴ ha under SSP126, SSP245, and SSP585, respectively, thereby enhancing carbon sink potential. This study offers scientific insights into carbon sink management in ASR.