The sparse temporal sampling rates (2-8 times daily) expected in next-generation missions to retrieve rainfall rates from space lead to potentially large errors in estimated rainfall accumulations at short time scales (daily to weekly) and may reduce the utility of such observations for surface water and energy balance applications. Other types of remote sensing observations may be able to mitigate the impact of rainfall sampling errors. For instance, surface wetness observations from microwave radiometry provide a temporally integrated measure of past rainfall intensities. If properly assimilated into a land surface model, these observations can correct land surface model predictions for errors arising from an inaccurate representation of rainfall forcings.

This paper will explore the use of an Ensemble Kalman filter (EnKF) to provide the framework for the integration of - temporally sparse - rainfall observations and - vertically shallow - surface brightness temperature observations into a surface water and energy balance model. The focus will be on the potential of the EnKF and L-band surface brightness temperature observations to correct model predictions for the impact of temporal sampling errors in precipitation forcings. Results will be shown for both a synthetic fraternal twin assimilation experiment as well as real brightness temperature observations taken during the 1997 Southern Great Plains Hydrology Experiment (SGP97).