Validation of O(S-1) wind measurements by WINDII: The WIND imaging interferometer on UARS

This paper describes the current state of the validation of wind measurements by the wind imaging interferometer (WINDII) in the O(S-1) emission. Most data refer to the 90-to-110-km region, Measurements from orbit are compared with winds derived from ground-based observations using optical interferometers, MF radars and the European Incoherent-Scatter radar (EISCAT) during overpasses of the WINDII fields of view. Although the data from individual passes do not always agree well, the averages indicate good agreement for the zero reference between the winds measured;on the ground and those obtained from orbit. A comparison with winds measured by the high resolution Doppler imager (HRDI) instrument on UARS has also been made, with excellent results. With one exception the WINDII zero wind reference agrees with all external measurement methods to within 10 m s(-1) at the present time. The exception is the MF radar winds, which show large station-to-station differences. The subject of WINDII comparisons with MF radar winds requires further study. The thermospheric O(S-1) emission region is less amenable to validation, but comparisons with EISCAT radar data give excellent agreement at 170 km. A zero wind calibration has been obtained for the O(D-1) emission by comparing its averaged phase with that for O(S-1) on several days when alternating D-1/S-1 measurements were made. Several other aspects of the WINDII performance have been studied using data from on-orbit measurements. These concern the instrument's phase stability, its pointing, its responsivity, the phase distribution in the fields of view, and the behavior of two of the interference filters. In some cases, small adjustments have been made to the characterization database used to analyze the atmospheric data. In general, the WINDII characteristics have remained very stable during the mission to date. A discussion of measurement errors is included in the paper. Further study of the instrument performance may bring improvement, but the utimate limitation for wind validation appears to be atmospheric variability and this needs to be better understood.