Objective This study aimed to explore the spatial distribution characteristics of soil organic matter (SOM) in soil profiles and its driving factors, which will provide a scientific basis for the precise assessment and management of carbon sink functions in arid regions, and has significant demonstration significance for the sustainable development of ecosystems and the implementation of carbon neutrality strategies under the background of global climate change.

Method This study took the Hui Autonomous Prefecture of Changji in Xinjiang as the research area. Based on its complete vertical zonation of mountainous areas, the samples were systematically collected from 210 soil profiles. These samples were divided into five soil layers (0 - 20 cm, 20 - 40 cm, 40 - 60 cm, 60 - 80 cm, and 80 - 100 cm) by weighted average. The variation characteristics of SOM at different soil depths and the vertical differentiation under different land use types were explored. And the spatial distribution characteristics of SOM in the surface layer (0 - 40 cm) and deep layer (40 - 100 cm) were investigated. Random forest and structural equation methods were used to analyze the influencing factors by combining with six types of driving factors including topography, climate, vegetation, drought index, soil properties, and human activities.

Result From the perspective of soil depth, SOM showed a significant decreasing trend. The average value of the 0 - 20 cm soil layer was 17.93 g kg⁻¹, while that of the 80 - 100 cm soil layer dropped to 8.01 g kg⁻¹, with a change rate of 44.7%. All soil layers exhibited strong spatial variability. Horizontally, SOM in the study area presented a spatial pattern of high values in the southern mountainous area, and low values in the central plain and the northwest desert. High-altitude areas were distributed in a band-like pattern, while the central transition zone showed a mosaic pattern of patches and bands. The study of driving factors indicated that soil bulk density and elevation in the surface soil had more significant regulatory effects. In the deep soil, elevation and precipitation had more significant regulatory effects. Topographic factors were the dominant driving factors for the spatial distribution of SOM in both the surface and deep layers.

Conclusion The results combined with different land use types indicated that human activities such as farming and irrigation had improved SOM to a certain extent, but natural factors remained the main regulatory factors of SOM. These results provided important data to support the precise managements of soil carbon pools, the restoration of degraded ecosystems, and the formulation of climate change adaptation strategies in arid regions.