Processing Millimeter Wave Profiler Radar Spectra

Abstract

Spectral processing algorithms employed in millimeter-wave profiling radars typically obtain good signal-to-noise ratios from weakly scattering clouds by incoherently averaging many spectra. Radar operating characteristics dictate sampling times on the order of a few seconds. Presented here are analyses showing that changes in the vertical wind during the sampling period can be a major contributor to the measured spectrum width. Such broadened spectra violate the assumptions made in spectral inversion techniques, and may lead to incorrect interpretations of the turbulent and microphysical characteristics of the radar volume. Moreover, it is shown that there are several factors involved in determining the measured spectral shape: the averaging time window and horizontal advection velocity of the cloud, as well as horizontal inhomogeneities in cloud vertical velocity and microphysical fields. Current processing algorithms do not allow for distinction between these effects, leading to potential for large errors in retrievals. In this paper a simple technique is presented to remove this effect for monomodal spectra. A side product of this algorithm is high temporal resolution estimates of the volume-mean vertical wind.