Intelligent Assimilation of Satellite Data into a Forecast Model Using Sensor Web Processes and Protocols

The goal of the Sensor Management for Applied Research Technologies (SMART) On-Demand Modeling project is to develop and demonstrate the readiness of the Open Geospatial Consortium Sensor Web Enablement capabilities to integrate both space-based Earth observations and forecast model output into new data acquisition and assimilation strategies. Working with NASA’s Short-term Prediction Research and Transition Center, the SMART team is developing a sensor web-enabled processing workflow to intelligently assimilate Atmospheric Infrared Sounder (AIRS) satellite temperature and moisture retrievals into a regional configuration of the Weather Research and Forecast model over the southeastern United States. The SMART workflow involves mining North American Mesoscale forecasts for interesting weather phenomena, then determining whether AIRS observations are coincident with the detected weather events. The assumption is that assimilating AIRS observations of anomalous weather conditions will improve the forecast and is worth the extra computational cost; whereas, if there is no significant weather there is no advantage to assimilating additional data and not worth the extra computational cost. A variety of SWE protocols are used for data access and alert services, and for process chain definition.