PRECIPITATION IDENTIFICATION FROM RADAR WIND PROFILER SPECTRAL MOMENTDATA - VERTICAL VELOCITY HISTOGRAMS, VELOCITY VARIANCE, AND SIGNAL POWER VERTICAL VELOCITY CORRELATIONS

Correlations between range-corrected signal power S-rc and radial vert ical velocity V-r from the vertical beam of a UHF wind profiler can be used to distinguish between air- and precipitation-dominated echoes u sing an S-rc-V-r correlation diagram. While there is no clear correlat ion between vertical air motions and S-rc there is a strong correlatio n between the precipitation fall velocity and S-rc, in snow, and to a lesser extent, in rain. This is illustrated through intercomparison of three types of precipitation events. and two types of clear-air event s. Using a histogram of V-r from an event where there is evidence of p recipitation in its S-rc-V-r correlation diagram, and from other infor mation, it is possible to objectively determine a threshold value of V -r, referred to as V-r, that approximately identifies which measuremen ts are dominated by Rayleigh scattering from precipitation in that eve nt. A method is introduced that uses the histogram of observed V-r fro m that event to provide an estimate of how many measurements are incor rectly attributed to Bragg scattering or Rayleigh scattering as a func tion of V-T. The error estimates can be used to select V-T on a case-b y-case basis and according to the needs of the particular application. An objective dual-optimization technique results in an estimated over all error of less than 6%, averaged over three case studies. In additi on, it is shown that inclusion of velocity variance from the vertical beam in the S-rc-V-r correlation diagrams can help distinguish between rain and snow, and between convective and stratiform precipitation.