In order to define the spatial distribution of soil air permeability and do correlation analysis of its impact factors in basin scale, researches in Jinghui irrigation district at the middle of Shaanxi province were made. Based on the spatial analysis function of geographic information systems, the results show that the optimal fitting model for air permeability is an exponential model, and the statistical method is suitable to simulate the spatial structure and variation characteristics of soil air permeability. The bulk density, saturated moisture content, saturated degree and soil air permeability’s spatial autocorrelation C/(C+C0) are all greater than 0.9. It shows the spatial heterogeneity induced by the autocorrelation has a strong degree in the spatial heterogeneity of the study scale. Recommend sampling distances of air permeability is 7.5km. Significant correlation is found between bulk density and soil air permeability, and the absolute value of Pearson correlation coefficient is 0.595. High correlation is found between soil air permeability and saturated degree, and the absolute value of Pearson correlation coefficient is 0.959. The bulk density and the saturated degree are proved to be the main impact factors by interactive correlation analysis between air permeability and soil physical parameters. The influence ranges of bulk density and soil saturation degree on air permeability are both from -20km to 20km at 95% confidence level.