High Resolution Wind Forecasts Using Multi-Sensor Satellite Surface Characterization Data

Mission Research Corporation (MRC) will develop a near real-time high resolution multi-sensor satellite surface characterization retrieval aand mesoscale model assimilation technique to predict land surface efffects on atmospheric dynamics. We will include time-varying land cover characteristics and will investigate incorporating satellite-based surface moisture and cloud liquid water data. Atmospheric prediction models currently describe, with reasonable accuracy,the weather where there is relatively strong synoptic forcing. Weak circulations driven by the diurnal heating and cooling of the surface are currently more difficult to predict. These flows depend on timely, accurate descriptions of surface vegetation information and moisture content. The proposed effort will take advantage of related developments to produce nearreal-time input data to rectify this deficiency. United States Geological Survey static land cover data at one kilometer resolution will be augmented with current Advanced Very High Resolution Radiometer satellite imagery to produce near real-time land cover data. The investigation of surface soil moisture retrieval will be based on calculating microwave surface emittance from the DMSP SSM/I passive microwave sensor and its use for forecast model initialization will also be explored. The near real-time data system developed under this effort will significantly improve the accuracy of weather forecasts worldwide. Weather forecasts are currently based on static surface characterizations. Near real-time satellite data is available worldwide and can be applied to forecasts in regions for which surface data is not available. These improvements will be very important to worldwide opal weather forecasting and to predicting the flow of hazardous materials under all conditions.