The ASR-9 Weather Systems Processor (WSP) is in the process of being deployed by the Federal Aviation Administration (FAA) at 35 medium and high-density airports across the United States. The Machine Intelligent Gust Front Algorithm (MIGFA) developed at Lincoln Laboratory provides important gust front detection and tracking capability for this system by utilizing multi-dimensional image processing, data fusion, and fuzzy logic techniques to recognize gust fronts observed in WSP data.

Lincoln Laboratory continues to actively monitor algorithm performance at initial limited production WSP sites in Austin, TX (AUS) and Albuquerque, NM (ABQ). At AUS, false detections have occurred most often in association with bands of low-reflectivity rain echoes moving with the general steering winds, while missed or late detections have occasionally occurred when gust fronts are near or embedded in the leading edge of approaching line storms. In ABQ, "canyon wind" events emanating from mountains located just east of the airport occur with very little lead time, and often with little or no radar signatures, making detection on the basis of the radar data alone difficult.

Several significant algorithm improvements in the areas of contextual decision-making and additional data fusion have recently been made to address these specific problems as well as to improve general algorithm performance. For example, previous versions of the WSP MIGFA only processed data within the nominal 15 nmi gust front detection range. However, additional processing of the available 60 nmi range data has been found to provide very useful contextual cueing information such as recognition of approaching line storms, improved classification of precipitation regimes, and determination of mid-level steering winds that control general precipitation echo movement. The additional information gleaned from this extended range processing is used to dynamically adjust local and global detection sensitivity in response to the current weather environment. At ABQ, wind measurements from the Low Level Wind Shear Alert System (LLWAS) anemometer network are now being processed by MIGFA to identify wind shifts as they traverse the network, and to generate gust front detections from this information even in the absence of radar evidence.

This paper describes each of the enhancements and presents results of comparison studies of data obtained from the Austin and Albuquerque WSP sites.

* This work was sponsored by the Federal Aviation Administration under Air Force Contract No. F19628-00-C-0002. The views expressed are those of the authors and do not reflect the official policy or position of the U.S. Government. Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the US Government.