Cold-arid regions cover more than half of China’s land area. However, their climatic and environmental conditions have long constrained agricultural development. To fully exploit the advantages of agroclimatic resources, expand the endowment of climatic resources, and explore localized agricultural development pathways, it is essential to scientifically understand the distribution characteristics of agroclimatic resources and study the intrinsic traits of climatic resources in China's cold arid regions. Based on meteorological data with a 4 km resolution from 2000 to 2020, this study constructed a Cold-Arid Synergy Index (CASI), integrating the dual dimensions of thermal limitation (Thermal Restriction Index, TRI) and moisture stress (Hydrological Stress Index, HSI). This index quantifies the spatial differentiation pattern of cold-arid interactive stress and analyzes the contribution mechanisms of its dominant factors. The results showed that cold-arid agroclimatic zones in China account for 16.42% of the national land area and exhibit a northeast-southwest belt-shaped distribution. CASI-based zoning revealed a five-category gradient: Grade I zones (extreme stress) are primarily located on the Tibetan Plateau, Grades II and III zones (Hexi Corridor, Inner Mongolia Plateau) form the main body of the agro-pastoral ecotone, and Grades IV and V zones (Northeast Plain to Central Gansu) have optimal hydrothermal matching and are suitable for large-scale agricultural production. At the provincial scale, provinces such as Gansu and Inner Mongolia have over 40% of their area classified as cold-arid, with CASI standard deviations reaching 0.10~0.18, highlighting the climatic sensitivity of transition zones. The LISA spatial autocorrelation analysis indicated that 28.52% of the cold-arid area comprises Low-Low clusters (LL-type), representing core advantageous production areas; while 28.24% belong to High-High cluster areas (HH-type), forming agroclimatic high-risk islands that require targeted enhancement of climate resilience. The dominant factor analysis revealed that aridity-dominant zones account for 73%, whereas cold-dominant zones account for 27%, with the contribution of cold stress increasing significantly with elevation. The CASI system constructed in this study provides a novel methodology for cold-arid agroclimatic zoning, and the zoning results offer a scientific basis for optimizing the agricultural spatial layout, enabling precise resource allocation, and developing specialized industries within cold-arid regions in the context of climate change.