COMPARISON OF 3 AIRBORNE DOPPLER SAMPLING TECHNIQUES WITH AIRBORNE IN-SITU WIND OBSERVATIONS IN HURRICANE GUSTAV (1990)

Three different airborne Doppler radar sampling strategies were tested in Hurricane Gustav (1990) on 29 August 1990. The two new strategies were the fore-aft scanning technique (FAST) and airborne dual-platform Doppler sampling. FAST employs radar scans in cones pointing alternat ely fore and aft of the vertical plane that is perpendicular to the fl ight track. The airborne dual-platform sampling uses two Doppler radar s, each aboard a separate aircraft. The Doppler radars scan strictly i n the vertical plane normal to the flight track. The aircraft fly simu ltaneously along different, preferably perpendicular, tracks. The thir d strategy tested in Hurricane Gustav was single-platform sampling, wh ich uses one Doppler radar on one aircraft that flies two consecutive, usually orthogonal, flight tracks. The antenna scans in the plane nor mal to the flight track. The third technique had been used previously in hurricanes and other disturbed weather. The rms differences between the aircraft in situ winds and the Doppler winds derived near the air craft by single-platform sampling, dual-platform sampling, and FAST ar e found to be 7.8, 5.1, and 2.5 m s-1, respectively. These results sug gest that in hurricanes dual-platform flat-plane sampling and FAST bot h enable substantial improvements in the accuracy and temporal resolut ion of airborne Doppler wind fields over those obtained from single-pl atform, flat-plane scanning. The FAST results should be applicable to dual-beam sampling, which began in 1991. The actual rms errors of Dopp ler winds far from the flight tracks, at levels well above flight leve l, and in highly sheared environments may be significantly higher than the above differences.