The objective and quantitative analysis of spatial heterogeneity of soil nutrients and their influential factors can provide scientific basis for precision fertilization. Apple production area at Liquan County, Shaanxi Province was selected as the research area, soil samples at 0-40cm depth were collected by the “S” shape sampling method, the mapping accuracies of soil nutrients including organic matter, alkali hydrolysable N, available K and available P were studied by geostatistics and GIS. The optimal prediction models were determined by cross-validation and relative tolerance, and spatial distribution maps of soil nutrients were drawn, the spatial distribution characteristics of soil nutrients were comprehensively analyzed. Moreover, influential factors of soil nutrients and their weights were decided by Pearson correlation and multiple regression analysis. The results showed that: 1) Under the optimal variogram model, the interpolation accuracy of Cokriging were better to predicate organic matter, available K and available P compared by Ordinary Kriging, while, Ordinary Kriging was better to predicate alkali hydrolysable N. 2) Soil orgamic matter showed a decrease trend from Junma Town-Qiandong Town, Jianling and Zhaoling Town to the middle part area; the high value area of hydrolysable N was from Zhao Town to the east of Shide Town, decreasing toward the east, north and southwest respectively; the high value area of available K was located in the south of the studying area; The high value area of available P was located among Fenghuo and Junma Town, Shitan-Zhaoling Town. 3) Soil organic matter and terrain factors were significantly correlated. Alkali hydrolysable N was affected by NDVI, soil types and terrain factors. Available K content was significantly correlated with slope and aspect, however, no significant correlation was found for available P.