The spatial and temporal structures of soil moisture and soil properties of SGP '97, Wahita '92 and Monsoon '90 experiments were characterized to obtain a better understanding of the spatial and temporal variabilities of soil moisture under present climatic conditions. The sample autocorrelation of temporal soil moisture has a long memory. The periodogram of detrended data has a sharp peak at the diurnal cycle, which highly contributes to the total variance in soil moisture evolution. The phase difference between rainfall and soil moisture is less than data interval (30min). The cross-correlation shows little correlation between the soil moisture and the rainfall. The two-dimensional correlograms of soil moisture and porosity field are isotropic and highly correlated. The two-dimensional spectra of soil moisture and porosity field are significantly maximum centered in zero frequency, which shows a two-dimensional trend caused by the natural variation. The two-dimensional correlation function and spectrum of soil moisture field are similar to those of porosity field. This similarity in structural shape can reduce the efforts to characterize the soil moisture field. It is important to know the maximum correlation length of soil moisture and properties in large scale sampling of the soil moisture field. The correlation length of soil moisture and properties in the fixed area remains unchanged with the variation of observing resolution. The relation between the size of mapping area and correlation length shows that correlation length of soil properties increases with size increasement up to a certain size and then converges a value. Washita '92 and SGP '97 soil moisture mapping area is insufficient to estimate the maximum correlation length of soil moisture field. Maximum correlation length of soil properties is estimated by using STATGO data.