This study focuses on the development of a consistent methodology for soil-moisture inversion from synthetic aperture radar (SAR) data with the use of the integral equation model (IEM), developed by A. K. Fung and colleagues, without the need to prescribe time-varying land-surface attributes as constraining parameters. Specifically, the dependence of backscatter coefficients obtained from synthetic aperture radar (SAR) on the soil dielectric constant, surface-roughness height, and correlation length was investigated. The IEM was used in conjunction with an inversion model to retrieve soil moisture by using multifrequency and multipolarization data (L-, C-, and X-bands) simultaneously. The results were cross validated with gravimetric observations obtained during the Washita '94 field experiment in the Little Washita Watershed, Oklahoma. The average error in the estimated soil moisture was of the order of 3.4%, which is comparable to that expected due to noise in the SAR data. The retrieval algorithm performed very well for low incidence angles and over bare soil fields, and it deteriorated slightly for vegetated areas and overall for very dry soil conditions. Although the original IEM model was developed for bare soil conditions only, one important result of this study was the fact that the retrieval algorithm performed well for vegetated conditions, as demonstrated by the fact that the convergence ratio varied between 92% (dry conditions) and 98% (wet conditions) of all pixels for all days of the experiment. The sensitivity of soil-moisture estimates to spatial aggregation of remote-sensing data before and after the retrieval also was investigated. The results suggest that there is potential to improve the operational utility of high-resolution SAR data for soil-moisture monitoring by compressing the SAR data (preaggregation) to a spatial resolution at least one order of magnitude above that of measurement.