Base on 3 m×3 m grid in sloping cornfield with soil auger in Yingwugou Small Watershed of the Dan River Basin, a total of 39 sampling points were collected, and soil water content and nutrient content were measured in different soil depths. Meanwhile, the soil properties of different depth have been analyzed by traditional statistical and geo-statistics approaches. The results showed: the mean value of total nitrogen and soil organic carbon reduced as soil depth increased in general. But soil water content increased as the soil depth increased. The change of total phosphorus with soil depth was not obvious. The total nitrogen, soil water content, soil organic carbon and total phosphorus presented a moderate intensity variation and strong spatial dependence. In these four sampling depths, semi-variance model can simulate the precisions of total nitrogen, soil water content and total phosphorus in the spatial structures of 0 to 10 cm and 10 to 20 cm well. But the spatial structure of soil organic carbon was not good, which could not be simulated with semi-variance model. The analysis with Kriging interpolation showed that, the total nitrogen, soil water content and total phosphorus presented layered distribution in the spatial structures of 0 to 10 cm and 10 to 20 cm; when the spatial distribution changed to 10 to 20 cm from 0 to 10 cm, the average total nitrogen content reduced to 0.310 g/kg from 0.598 g/kg, while the average water content and total phosphorus increased from 12.988% to 15.439% and from 0.229 g/kg to 0.366 g/kg, respectively.