Abstract: In this paper, based on the principle for function approaching of BP neural network, the method of BP artificial neural network for field information interpolation was studied, and soil available N spatial distribution maps of BP artificial neural network interpolation and Kriging interpolation were drawn using ArcView3.2, the errors for interpolation results of the two methods were analyzed. The result showed that interpolation method of BP artificial neural network excels Kriging interpolation method, and the neural network interpolation method can be used to express the field information spatial distribution visually and exactly, which can be used as a reference for precision fertilization, precision irrigation and other precision management in the field.