Since 2002, the Advanced Microwave Scanning Radiometer on the Earth Observing System (AMSR-E) has been observing radiances from the atmosphere and surface of the Earth using 6 frequencies in microwave region from 6.9GHz to 89GHz.The instrument measures vertically and horizontally polarized radiation at each frequency. Current microwave imagers, such as SSM/I, SSMIS and AMSR-E, use almost the same channels in microwave region from 19GHz to 89GHz to acquire the information on moisture in the atmosphere. In addition to those channels, AMSR-E has 4 low frequency channels in microwave region (6.9GHz and 10.7GHz, dual polarization), which are sensitive to the sea surface wind and the sea surface temperature and less affected by the atmosphere. Therefore, these measurements contain the information on the sea surface wind and the sea surface temperature and can be assimilated under all weather conditions. The purpose of this study is to examine the impact of AMSR-E radiances including the low frequency channels on the National Centers for Environmental Prediction (NCEP) global data assimilation system (GDAS).

In order to include low frequency channels in the assimilation of AMSR-E radiances, a new microwave ocean emissivity model has been developed. The emissivity model is a two-scale ocean roughness model and the coefficients of the model were derived from the satellite measurements. Using the new emissivity model at low frequencies substantially reduced the biases in the simulated AMSR-E brightness temperature. The new emissivity model and its adjoint model have been implemented in the Community Radiative Transfer Model (CRTM) for the assimilation experiment. In the experiments, AMSR-E radiances from 10.7GHz to 36.5GHz have been assimilated with NCEP Gridpoint Statistical Interpolation (GSI) analysis system and the NCEP operational global forecast model.

In this presentation, we will show a comparison of the emissivity models and the impact of including the AMSR-E radiances in an assimilation system on forecasts.