Brightness temperatures (TB's) measured from a special sensor microwave imager (SSM/I) are used to improve the storm initialization of Hurricane Bonnie (August 1998). SSM/I data is used to locate gross features of the storm such as areas of precipitation and its center. The results of this qualitative analysis on the prediction of Hurricane Bonnie will be presented.

Furthermore, preliminary results from a more quantitative analysis will also be discussed. The TB's obtained from a four-stream radiative transfer model (RTM), which takes as input, numerical weather prediction (NWP) model variables such as pressure, temperature, and moisture, will be compared to SSM/I TB's. The feasibility of using SSM/I TB's directly for storm initialization will be discussed. An adjoint of the RTM is developed, and a sensitivity analysis is performed. The goal of this study is to test the potential application of SSM/I TB's for hurricane initialization, therefore improving the quality of the forecast of Hurricane Bonnie.