Quantifying the contributions of factors influencing the spatial heterogeneity of soil aggregate stability and erodibility in a Mollisol watershed

Soil aggregate stability and erodibility reflect soil resistance to erosion. Although the factors influencing soil aggregate stability and erodibility have been extensively studied, the driving effects of these factors and their interactions remain limited. 184 sampling sites were selected in topsoil (0-20 cm) of a small watershed (1.42 km²) in a Mollisol watershed to measure soil bulk density (BD), soil porosity (SP), soil moisture content (MC), soil organic carbon (SOC), total nitrogen (TN), >0.25 mm water stable aggregates content (WSA_{> 0.25}), mean weight diameter (MWD) and K factor. Pearson correlation analysis, semi-variance function, redundancy analysis (RDA), and structural equation model (SEM) were used to quantify the impact of environmental variables (individual and interaction) on the spatial variations of WSA_{> 0.25}, MWD, and K factor. The findings indicated that higher values of WSA_{> 0.25} and MWD are observed in the central and western watershed, while the K values tend to be lower in areas with high WSA_{> 0.25} and MWD values within the watershed. The Exponential model optimally described WSA_{> 0.25}, MWD, and K factor with C₀/(C + C₀) indicating moderate spatial dependence for MWD (39.79%) and K factor (42.86%), while strong spatial autocorrelation for WSA_{> 0.25} (7.23%). Soil properties (moisture content, silt content, and bulk density), topography (elevation, SPI, and slope), and land use contributed 46.6%, 41.4%, and 9.1% of the spatial variation in WSA_{> 0.25}, MWD, and K factor, respectively. SEM revealed that silt content, SOC, and water condition played a fundamental role in controlling the spatial variability of WSA_{> 0.25}, MWD, and K factor. Topography exerted both direct or indirect effects by coupling land use or soil properties spatially. Land use had direct or indirect effects on WSA_{> 0.25} and K factor through regulating MC, but it primarily influences MWD indirectly through impacting MC. These results could clarify the roles and influencing paths of factors controlling the spatial heterogeneity of WSA_{> 0.25}, MWD, and K factor, contributing to optimizing land management strategies.