Taking into account the problem of soil salinization existing in Huanghuaihai plain at present, spatial variability of soil salinity was surveyed using remote sensing and the EM38 (electromagnetic induction instruments) in Fengqiu county of Henan Province. Spatial variability of soil salinity was analyzed by three methods of Kriging, exponential regression and regression-Kriging. The results indicated that soil electrical conductivity had spatial structure characteristics in a certain region, which was in line with the exponential model distribution and had strong spatial autocorrelation. The trends of spatial variability of soil salinity were similar to each other by three methods, but the spatial distributions of soil salinity by index regression and regression-Kriging were more specific and thorough than that by Kriging interpolation. The integrated methods by adopting comprehensively multi data sources, namely integration of remote sensing image data and EM38 measuring data, and integration of exponential regression and regression-Kriging, could improve the spatial prediction precision of soil salinity and were more accurate than that only by Kriging interpolation, which showed that in the study area the distribution of soil salinization appeared a tendency of increasing from north to south and from west to east, and exhibited obvious trend effect.