Measurement of raindrop fall speeds using Doppler lidar

Although coherent Doppler lidars have been used extensively to study atmospheric boundary layer dynamics under clear-air conditions, much less attention has been devoted to the potential of using these systems to study clouds and precipitation. One limitation of the lidar is that the return signals are strongly attenuated by clouds, making them less useful than radars for probing the internal dynamics of clouds. By contrast, the attenuation of the lidar signal is typically much less severe in precipitation. A vertically staring Doppler lidar is often able to “see” many 100s of meters through precipitation. The principle advantage of the lidar over the radar in this situation is its ability to unambiguously discriminate between the contribution of the clear-air (aerosol) return from the precipitation return in the Doppler spectra.

In this study, we examine the characteristics of lidar Doppler spectra collected during several mid-latitude summer rain events. These Doppler spectra were acquired from a vertically staring Doppler lidar operating at the US Department of Energy’s Atmospheric Radiation Measurement (ARM) site in north central Oklahoma, i.e. the Southern Great Plains (SGP) site, during the period from 1 August to 30 September 2017. A number of stratiform and convective rain events were observed during this period, including at least one severe thunderstorm on 11 August 2017 in which rain rates exceeded 100mm/hr.

We show that the lidar Doppler spectra exhibit a distinct bimodal (sometimes trimodal) structure during rain events. A multimodal Doppler velocity estimation algorithm is used to estimate vertical velocities corresponding to the precipitation and aerosol maxima in the Doppler spectra, enabling calculation of the corrected rain drop fall speeds. The corrected fall speeds from the lidar’s lowest valid range gate are then compared to weighted averages of the rain drop fall speeds using data from a nearly collocated ground-based video disdrometer. The weighting depends on the drop size distribution and the backscatter cross section of raindrops. The results of the lidar-disdrometer comparison are discussed.

We note that the Doppler lidar at SGP is also collocated with a 915MHz radar wind profiler, and two Ka-band radars (one scanning, and one vertically staring), all of which routinely log Doppler spectra. Part of the motivation behind this study is to examine the feasibility of developing a new operational data product that potentially uses lidar and radar Doppler spectra to retrieve height-resolved rain drop size distributions.