Raindrop Size Distributions Retrieved from Vertically Staring Coherent Doppler Lidar

Abstract. The US Department of Energy’s Atmospheric Radiation Measurement (ARM) program operates pulsed coherent Doppler lidars at all its fixed and mobile sites. These instruments provide accurate range- and time-resolved measurements of line-of-sight (radial) velocity and aerosol backscatter in the atmospheric boundary layer under clear-air conditions. Consequently, they have been used extensively to study winds and turbulence in the atmospheric boundary layer, while their ability to measure precipitation properties has received much less attention. In this study, Doppler spectra from a vertically staring pulsed coherent Doppler lidar (XR+) are used to retrieve clear-air vertical velocities in the presence of precipitation, and raindrop size distributions (RSDs) across multiple convective precipitation events at ARMs Southern Great Plains (SGP) site. RSDs derived from the lidar’s lowest valid height bin are compared to a nearby surface-based laser disdrometer. The disdrometer and lidar-derived RSDs show a high degree correlation, particularly under light to moderate wind speed conditions. For wind speed less than 2 ms-1, the median RSD correlation was 0.94. As the wind speeds and rain rates increased the correlations exhibit more scatter such that the median correlation for all observed wind speeds was 0.88. The median percent difference in the modal diameters was found to be 3.4% with the lidar producing slightly higher estimates. Significantly better agreement is obtained when the disdrometer is used to calibrate the lidar’s backscatter efficiency. In that case, median correlation coefficient for light winds increases to 0.96 and the median percent difference in the modal diameters decreases to -0.7%, with the lidar producing slightly lower modal diameter estimates than the disdrometer. The results of this study indicate that coherent Doppler lidars could be used operationally to obtain accurate height-resolved RSD retrievals. This would require careful data management given the large data volumes associated with the Doppler lidar spectra.