Surface soil moisture is a key variable in describing the water and energy exchanges at the land surface/atmosphere interface. Despite its importance in the fields of hydrology and meteorology, the lack of knowledge of global soil moisture has long been felt. Various land surface models show great potential for producing estimates of soil moisture with high temporal sampling and on a regional-to-global scale. In order to achieve a "correct" physical description of land surface processes, these models need to be rigorously validated. However, at the present time the availability of the soil moisture observations, especially on a large scale, which can be used for model validation and calibration is often limited. In this study, we are trying to expand the validation capabilities of current land surface models beyond those few areas where in situ data are readily available. Our approach involves coupling the land surface model with the state-of-the-art L-band microwave emission model to assimilate prognostic brightness temperature observable from satellite microwave radiometers. 10¨Cyear (1986-1995) time series of the main variables simulated from the SSiB land surface model in Global Soil Wetness Project (GSWP-2) have been utilized to drive the L-band microwave emission model, including surface soil moisture, soil temperature at the surface and at depth, snow cover characteristics, soil and vegetation characteristics, etc. The assimilated brightness temperature has been assessed with airborne ESTAR measurements from several large-scale campaigns held during the GSWP2 10 year period, including Washita 92 and 94, and Monsoon 90. Our results showed that the coupled models could serve as a potentially useful tool in validating various land surface models and provide a good reference for quality assurance of remotely sensed brightness temperature as well. Through such cross-validation exercise, we expect the consistency between model estimates and remote sensing observations may increase confidence in both products.