Abstract:Soil saturated hydraulic conductivity (Ks) is an important parameter influencing hydrological processes and the accuracy of hydrological model simulation. Understanding the profile distribution of Ks and their controlling factors under different vegetations at the slope scale are conductive to better understand the hydrological process and its regulation mechanism. In this study, the Ks and soil basic properties were measured of 12 profiles (0—200 cm) under different vegetations on slopes. These profiles were chosen according to different vegetations (grassland and forestland) and different slope aspects (northeast orientation and west orientation). The objective of this study is to examine the changes in the vertical distribution of Ks and to identify the main controlling factors for the variations of Ks at the slope scale in a relative small region. The results showed that Ks under different vegetations decreased initially with depth and then tended to increase downward, and the average values of Ks (Ks-average) in 0—20 cm soil layer were significantly higher than that in 20—200 cm soil layer (P < 0.05). There were no significant differences of the Ks-average in 0—20 cm soil layer among different vegetation types in the same slope as well as between different slopes for the same vegetation type (P > 0.05). However, due to the differences of soil texture and organic matter, the Ks-average in 20—200 cm soil layer under grassland was higher than that under forestland in the northeast organization slope, and the Ks-average in 20—200 cm soil layer in northeast orientation slope was higher than that in west orientation slope whether under grassland or forestland (P< 0.05). Ks under different vegetations were positively correlated with capillary porosity, saturated water content, sand and organic matter (except grassland in the west orientation slope) (P < 0.05), but were negatively correlated with bulk density, clay and silt content (P< 0.05). The factors influencing Ks can be attributed to three principal components: soil water retention capacity (49.36%), soil texture (24.98%) and soil fertility characteristic (13.92%). Based on the effects of soil properties, the pedotransfer functions (PTFs) of Ks under different vegetations were proposed by using multiple stepwise regression analysis. The PTFs with inputs of bulk density, soil texture and organic matter had a better fitting ability of Ks (R2 = 0.60 ~ 0.86,P < 0.001), which could be used as a reference for the simulation and prediction of Ks at the slope scale in Liudaogou watershed.