Spurious Velocities in Doppler Radar Data Caused by a Moving Antenna Feedhorn

Abstract

Modern Doppler weather radars use clutter filtering to reduce the strength of ground targets and enhance the detection of meteorological echoes. WSR-88D [NEXRAD (Next Generation Radar) and TDWR (Terminal Doppler Weather Radar)] radars, for example, will automatically reject up to 50 dB of signal from stationary targets. Clutter rejection operates on the assumption that the targets to be rejected are really stationary. This assumption is correct for all true ground targets. Unfortunately, these targets are detected by radars using moving antennas. The design of these antennas usually places the feedhorn of the antenna some distance from the center of azimuthal rotation of the antenna. The WSR-88D, for example, has the feedhorn 4.78 m from the center of azimuthal rotation. One consequence of this design feature is that there is a relative motion between the feedhorn and stationary ground targets, which introduces a radial velocity that depends upon the distance from the feedhorn to axis of rotation, the azimuthal rotation rate, and the relative angle between the mainlobe direction and the target. Consequently, stationary targets can produce sidelobe echoes that have velocities. For example, data from the Lincoln Laboratory S-band FL2 radar show antenna rotation-produced moving sidelobe echo velocities as fast as 3 m s?1; the UND C-band radar produced erroneous velocities as fast as 0.8 m s?1. WSR-88D radars will generate erroneous velocities as large as 2.5 m s?1 for a 30° s?1 scan rate.