The NEXRAD in SPACE: Potential Improvements to Hurricane Analysis and Prediction

The NEXRAD in Space (NIS) is a geostationary satellite featuring a 35 GhZ (~1 cm, Ka band) Doppler scanning radar, which has been under development in NASA's incubator program for nearly 5 years. NIS technology is being developed principally as a tool to provide measurements to improve tropical cyclone analysis and prediction. This will be accomplished by providing temporally continuous measurement of reflectivity and line-of-sight (LOS) Doppler winds on a three dimensional spiral span grid spanning a circle of 5300 km diameter covering the Western tropical Atlantic, Eastern Tropical Pacific, the Caribbean Sea and Gulf of Mexico. The proposed .02 degree beam width yield 12 km horizontal resolution at the sub point with a range resolution of 300m all the way to the surface.

Over the past 5 years, the University of Wisconsin has been experimenting with assessing the impact of this instrument on the potential for cloud resolving assimilation of tropical cyclone internal structure. For this, OSSE experiments are being conducted, based on the application of NIS to a cloud resolving assimilation system using an EnKf technique. Numerical cloud resolving prediction of the resulting initial ensemble forward in time, are used to create a probabilistic forecast and evaluate the impact of NIS on the forecast.

Results of OSSEs are very promising, indicating a strong likelihood that the storm internal structure can be initialized with good fidelity, leading to impressive gains in accuracy of prediction of hurricane structure and overall intensity. Ongoing studies are experimenting further with a variety of storm conditions and to evaluate the value gained in the context of other low earth orbit satellite based microwave, radar and IR measurements, such as TRMM and SSMI, AMSR as well as observations of the operational GOES platforms. Additional OSSE experiments are being conducted to study closely the relationship between possible improvements or compromises on NIS pixel size, scan frequency and additional capabilities on even greater improvements.