The importance of snow to efficient water resources management, especially in the mountainous western U.S., has long been recognized. Current model-based approaches to hydrologic forecasting are limited by model biases and input data uncertainties, while ground based measurements have limited coverage and are inappropriate for regional scale hydrology. Passive microwave remote sensing offers an opportunity for all-weather monitoring of snow properties, including areal extent and water equivalent, over larger areas. The Advanced Microwave Scanning Radiomater (AMSR-E),which was launched in 2003 with the EOS-AQUA satellite, provides better spatial and spectral resolution in comparison with other operational passive microwave sensors. However, most snow properties retrieval algorithms are not accurate enough for operational applications. Data assimilation offers a framework for optimally merging information from remotely sensed observations and hydrologic model predictions, and ideally overcoming limitations of both. An additional advantage of data assimilation is that it simultaneously solves the state inversion problem, i.e. calculating snow depth or water equivalent from microwave brightness temperature. This work describes the use of an ensemble Kalman filter (enKF), which is a Monte Carlo variation of the traditional Kalman filter, to assimilate AMSR-E brightness temperatures over the Snake River basin for the winter of 2004. The approach is built around the Variable Infiltration Capacity (VIC) macroscale hydrology model, which balances water and energy over each model grid cell at each timestep. The VIC model represents the effects of subgrid variability in soil moisture, vegetation, topography and precipitation. Model simulations were performed at a spatial resolution of 1/8o with a daily timestep. Brightness temperature values are estimated using the Dense Media Radiative Transfer (DMRT) model which is based on the quasicrystalline approximation. Results showed that the enKF is an operationally feasible solution for the assimilation of remotely sensed observations. The performance of the assimilation system is evaluated using snow water equivalent surface observations from the SNOTEL station network. Also, the effect of the assimilation on streamflow prediction is examined, and specific limitations to the approach are discussed.