Terminal Doppler Weather Radar (TDWR) Low Level Wind Shear Alter System 3 (LLWAS 3) Integration Studies at Orlando International Airport Airport in 1991 and 1992

In 1993 the Federal Aviation Administration (FAA) began deploying two new wind shear detection systems: .the Terminal Doppler Weather Radar (TDWR) and the third-generation Low Level Windshear Alert System (I&WAS 3). Currently, nine airports are scheduled to receive both aTDWR and an LLWAS 3. This number may eventually increase to as high as 45. When co-located, the systems will be integrated to provide a single set of wind shear alerts and improve’system performance. The TDWR production schedule required one of three integration algorithms to be chosen for specification by fall 1991. The three algorithms are the prototype integration algorithm developed at the National Center for Atmospheric Research (NCAR) and two algorithms developed at MIT Lincoln Laboratory (MIT/LL). To assess the performance of the three algorithms, MITiLL performed a study of the integration, TDWR, andLLWAS 3 algorithms at Orlando International Airport (MCO) in the Summer of 1991. Based on the results of this study, Lincoln Laboratory and NCAR issued a joint recommendation that the FAA procure one of the integration algorithms developed at MIT/LL. This algorithm was demonstrated at the Orlando International Airport in the summer of 1992. We discuss results of the 1991 comparative study and a follow-up study of the TDWR, LLWAS 3, and. Message Level integration algorithms at Orlando in 1992. All of the algorithms met the requirement of detecting 90 percent of microburst level wind shear with loss events. LLWAS 3, Build 5 TDWR, and the MIT/LL integration algorithms run with Build 5 TDWR, all met the requirement that less than 10 percent of wind shear alerts be false. TheNCAR prototype did not utilize Build 5 TDWR. Build 4 TDWR and all the integration algorithms run with Build 4 TDWR did not meet the false-alert requirement. Detailed descriptions of the algorithms are given. The methodology for estimating various algorithm performance statistics based on a comparison with a dual-Doppler algorithm is detailed.