The coupling of the Koster and Suarez MOSAIC Land Surface Model (LSM) with the GEOS Data Assimilation System (DAS) General Circulation Model (GCM) has revealed previous biases in the application of prescribed soil moisture. Before the implementation of the MOSAIC LSM, the GEOS DAS GCM relied on prescribed monthly mean soil moisture values combined with a specified simple evapotranspiration coefficient function to compute the surface moisture flux. Predicted soil moisture values from the coupled system are significantly lower than previously prescribed values resulting in a strong reduction in surface moisture flux. As expected, the reduction in surface moisture flux results in a drier planetary boundary layer as well as a drier mid-troposphere. The response of the GEOS DAS GCM convective and stratiform precipitation as well as cloud radiative effects to this drier climate indicate the implementations of the moist physics parameterizations is strongly dependent on the surface moisture flux.