Novel Technique for Assimilating SSM/I Observations of Marine Atmospheric Storms

Marine storms strongly affect operations at sea, both DoD and commercial. The DMSP SSM/I clearly depicts certain features of marine storms. Improved depiction of these storms will aid nowcasting in support of execution.of operations at sea and fore- casting of the evolution of these storms in support of planning. However, SSM/I data as well as many other satellite data have only small impacts on the forecasts and analyses of the operational meteorological centers. At many centers the SSM/I data are not currently used. The SSM/I data has great potential for monitoring and depicting marine storms. To attain this potential we propose a novel technique for assimilating SSM/I observa- tions of marine atmospheric storms. In plain terms we propose to match the observ- able features, shifting the short term forecast (or background) of the data assimila- tion system to best match the available satellite data. Our approach to this problem is technically a novel characterization of errors for numerical weather predictions. The basic idea is to characterize forecast errors by a distortion, composed of continuous displacement and amplification fields. We will apply these concepts to SSM/I observations of precipitable water and surface wind speed. These data are ideal for pinpointing the location and structure of marine storms. We will develop and test algorithms for a variational analysis method to make use of these data for initializing numerical forecasts of these storms.