Soil phosphorus (P) plays an important role in soil fertility and availability of micronutrients in soil, especially in arid and semiarid regions. Therefore, monitoring soil P condition is of great importance. The aim of the present study was to investigate the spatial variation of soil phosphorus by taking into account top soil EC data as secondary information. The research was performed on a grid of 0.75-1 km in an area of 367 km2. Soil phosphorus (P), Potassium (K), Zinc (Zn), Iron (Fe), Copper (Cu), Manganese (Mn), Organic Matter (O.M) and electrical conductivity (EC) were measured. Then variogram was built for P dataset and spatial prediction was done on a grid of 500 m using kriging estimator with taking into account the mean variation. Afterwards soil EC was used as covariate to develop cross-semivarograms in prediction of soil P using co-kriging method. Cross-validating the results from P predictions using only kriging estimator to that of co-kriging with EC data revealed that co-kriging offered better estimations with ME and MSE of 0.11 and 0.149, respectively. Kriging estimator had more smoother and diffused boundaries than that of co-kriging and resulted in more bias estimations (ME and MSE of -0.18 and -0.326, respectively). According to the results, co-kriging method and soil EC could be used successfully in improving spatial prediction of soil phosphor.