Continuous soil nutrient data is the basic data of soil information system, soil nutrient spatial interpolation study becomes very important because it can affect the data reliability of soil nutrient data straightly. In this study, soil available kalium data, Zhouzhi county, Shaanxi province, is sampled and got the interpolation map of available kalium with OK, IDW,Spline and TSA according to interpolating sampling points. The cross validation results showed that OK is the best which used spherical model in these methods; although Spline and IDW could also produce fair and reasonable results, both of them were sensitive to sampling density, interpolation result appeared large prediction errors in those regions which had exiguous sampling points,