Soil moisture is a determinant factor in most of complex environmental processes and has an important role in agricultural activities, hydrologic studies, and environmental monitoring. In this study the soil moisture data from the Advanced Microwave Scanning Radiometer 2 (AMSR2)were used and these data were compared with ground-based measurements over Rafsanjan plain. The comparison criteria’ s are consisted of the evaluation of the mean error, root mean squared error, and the correlation coefficient. The results of validation showed that the measurements of the soil moisture at the stations 4, 5, and 6 behaved similarly to AMSR2 soil moisture, and the correlations were above of 70%, and at the stations 1, 2, and 3, the correlations were low. In order to achieve better results, a downscaling technique was used. The technique suggests that remotely sensed soil moisture retrieved from the AMSR2 system at a 25km resolution can be downscaled to 1km resolution by using a simple linear equation based on parameters calculated with a regression model, which is based on three physical properties of 1km resolution retrieved from MODIS. The results revealed that the downscaling technique reduced the errors and increased the correlation coefficient, the correlation coefficients at stations 1, 2, and 3 increased from 0. 295, 0. 552, 0. 237 to 0. 864, 0. 7, 0. 75, respectively.