Radar interferometry technique and anisotropy of the echo power distribution: First results

Abstract

The theory of radar interferometry developed in the companion paper [ Van Baelen and Richmond, this issue] is applied to a set of measurements obtained with the middle and upper atmosphere (MU) radar in order to demonstrate the technique and to test the theory. The variations with respect to Doppler frequency of the coherencies and phases of the signals cross‐correlated between receiver pairs depend on the amplitude and direction of the wind in the manner predicted by the theory. The horizontal wind derived by the interferometric technique agrees quite well with the so‐called apparent velocity obtained by cross‐correlating signals in the time domain from different receiver pairs, and under our experimental conditions the latter estimate has been determined to be a good representation of the true wind, in spite of slight tendencies to overestimate the wind amplitude. The vertical wind derived by the interferometric technique shows significant differences with the Doppler wind obtained from a vertically pointing antenna, but these differences disappear when the Doppler wind is corrected for the off‐zenith distribution of the received power. An examination of the mean angle‐of‐arrival of the received signal reveals that there is a tendency for the angle to approach, though not attain, perpendicularity with the three‐dimensional wind vector, and perhaps also to tend to be perpendicular to isentropes in the crosswind direction. A postset beam steering analysis of several examples of data illustrates the behavior of the angular distribution of received power in time and space. When the signal‐to‐noise ratio is large, there is a tendency for the power to be fairly isotropic with respect to zenith angle, but highly distorted patterns can be encountered when the signal is weaker. There appears to be a tendency for the patterns to vary relatively slowly over time, but strong variations in altitude are observed in passing from one 300‐m range gate to the next. The direction of most intense received power does not necessarily correspond to the direction of mean angle‐of‐arrival of the signal as determined by interferometry.