Machine intelligent gust front algorithm for the Terminal Doppler Weather Radar (TDWR) and Integrated Terminal Weather System (ITWS)
Thunderstorms often generate gust fronts that can have significant impact on airport operations. Unanticipated changes in wind speed and direction are of concern from an air traffic safety viewpoint (hazardous wind shear) as well as from an airport planning point of view (runway configuration). Automated gust front detection is viewed by FAA and the air traffic community as an important component of current and future hazardous weather detection systems including the Terminal Doppler Weather Radar (TDWR), ASR-9 with Weather Systems Processor (ASR-9 WSP), and the Integrated Terminal Weather Systems (ITWS) for which TDWR is a principal sensor. In cooperation with the FAA, Lincoln Laboratory has successfully developed and tested a real-time Machine Intelligent Gust Front Algorithm (MIGFA) for use with Doppler weather radars. This algorithm resulted from the successful fusion of two complementing technologies developed at Lincoln Laboratory: computer vision/machine intelligence techniques originally developed for automated target recognition, and automated product-oriented weather radar data processing. Using these techniques, a version of MIGFA designed for use with TDWR has demonstrated substantial improvement over the existing TDWR gust front algorithm, detecting more and greater extents of gust fronts with fewer false alarms. MIGFA is slated to eventually replace the existing TDWR gust front algorithm and will be used as the gust front algorithm for the planned ITWS and ASR-9 WSP systems. A brief overview of techniques used by MIGFA to identify and track gust fronts will bre presented in this paper. More details, along with recent detection performance results, can be obtained from prior publications. However, detection and tracking of a gust front is only part of the task. Once the location of a gust front has been determined, the associated wind shear estimate and wind shift forecast must be computed. Several issues arises. For example, a gust front can be tens of kilometers in length, with outflow strength and contrasting environmental winds varying considerably along its length. Where along the front should the wind shear analysis be performed? Also, for airport planning purposes, air traffic controllers and managers need to plan runway configuration based on winds that may change suddenly when a gust front moves over the airport. Depending on the nature of the gust front, some of these winds are relatively transient while others are more persistent. How should the wind shift advisory produced by the algorithm take this into account? MIGFA uses a consensus derived from a variety of estimation techniques as a robust means of generating wind shear and wind shift estimates for detected gust fronts. These techniques, and some of their limitations, are discussed. Results of comparisons of MIGFA-generated wind shear and wind shift reports against observations are also presented. The paper concludes by outlining planned enhancements to incorporate additional information available under ITWS that should further improve the quality of MIGFA's wind shear and wind shift forecasts.