Basin‐wide sediment transport affects estimates of basin sediment yield, which is a fundamental scientific issue in drainage basin studies. Many studies have been conducted to examine erosion and deposition rates in drainage networks. In this study, we proposed a new approach using grain‐size standard deviation model of sedimentary samples from different geomorphological units for numerical analysis and paleo‐climate interpretation in the Shiyang River drainage basin, arid China. 1043 sedimentary samples were obtained from the upper reaches, the midstream alluvial plain and the terminal lake area; chronological frames were established based on 58 radiocarbon ages. Grain‐size standard deviation model was introduced to examine sediment components according to grain‐size and transport forces. In addition, transient paleo‐climate simulations, including the Community Climate System Model version 3 and the Kiel models, were synthesized, as well as the results from PMIP 3.0 project, to detect the long‐term climate backgrounds. Totally, we found four major common components, including fine particulates (<2 μm), fine silt (2–20 μm), sandy silt (20–200 μm), coarse sand (>200 μm), from basin‐wide sedimentary samples. The fine particulates and fine silt components exist in all the sedimentary facies, showing long‐term airborne aerosol changes and its transport by suspended load. There are some differences in ranges of sandy silt and coarse sand components, due to lake and river hydrodynamics, as well as the distance with the Gobi Desert. Paleo‐climate simulations have shown that the strong Asian summer monsoon during the transition of the Last Deglaciation and Holocene was conducive to erosion and transport of basin‐wide suspended load, also enhancing sediment sorting effects due to strong lake hydrodynamics. Our findings provide a new approach in research of long‐term basin‐wide sediment transport processes.