Topographic and Climatic Factors Driving Spatial Heterogeneity of Soil Quality in Arid Regions: An Assessment Based on Cotton Fields in Typical Watersheds of Xinjiang, China

Soil quality is a critical factor impacting agricultural productivity and ecosystem functions. Accurate assessment of soil quality is crucial for sustainable agricultural development. Xinjiang is the primary cotton-producing region in China. Its unique geographical condition, characterized by two basins surrounded by three mountains, results in distinct climatic conditions, soil-forming factors, and soil physical and chemical properties across different cotton-growing areas. The spatial differentiation patterns of soil quality and their primary factors in cotton-growing regions of different river basins are not yet fully understood. This study focused on four typical cotton-growing areas of Xinjiang, China. A total of 1588 plow-layer soil samples were collected, and 21 indicators covering soil physical, chemical, and environmental properties were measured. By constructing a minimum data set (MDS) and comparing the performance of linear (LS) and non-linear (NLS) scoring functions, the effects of geographical environmental factors on the spatial distribution patterns of soil quality in cotton fields of different basins were analyzed. The results showed the MDS, composed of data on soil bulk density and the contents of organic matter, available iron, available zinc, nickel, sand, and silt, could replace the total data set. The NLS-MDS was identified as the optimal assessment model. Its Nash–Sutcliffe efficiency coefficient (Ef = 0.84) and coefficient of determination (R2 = 0.70) were both higher than those of the linear model (Ef = 0.79, R2 = 0.66). The study also revealed significant spatial heterogeneity in soil quality across different cotton-growing areas. The average soil quality index (SQI) in the Aksu River Basin (SQINLS-MDS = 0.53) and Xiaohaizi Basin (SQINLS-MDS = 0.50) was significantly higher than that in the Kuitun River Basin (SQINLS-MDS = 0.44) and Manas River Basin (SQINLS-MDS = 0.41). Random forest analysis demonstrated that the relative importance of topographic (digital elevation model) and climatic factors (annual mean temperature, annual mean precipitation) on SQI was higher than that of the vegetation factor (normalized difference vegetation index). The strong interaction between topographic and climatic factors was the primary driver of the spatial distribution of soil quality. This study confirms significant spatial heterogeneity of soil quality in cotton fields across different river basins in Southern and Northern Xinjiang, and identifies the synergistic interaction between topographic and climatic factors as the dominant driver of this heterogeneity in arid regions. These findings provide a scientific basis for implementing site-specific agricultural management in arid regions.