Raindrop fall velocity in turbulent flow: an observational study

Merhala Thurai,Mathew Wingo,Viswanathan Bringi,Patrick Gatlin

doi:10.5194/asr-18-33-2021

Abstract

Abstract. Laboratory measurements of drop fall speeds by Gunn–Kinzer under still air conditions with pressure corrections of Beard are accepted as the “gold standard”. We present measured fall speeds of 2 and 3 mm raindrops falling in turbulent flow with 2D-video disdrometer (2DVD) and simultaneous measurements of wind velocity fluctuations using a 3D-sonic anemometer. The findings based on six rain events are, (i) the mean fall speed decreases (from the Gunn–Kinzer terminal velocity) with increasing turbulent intensity, and (ii) the standard deviation increases with increase in the rms of the air velocity fluctuations. These findings are compared with other observations reported in the literature.

Highlights

Measurements of the terminal fall speed of drops by Gunn and Kinzer (1949) under laboratory conditions with the pressure correction of Beard (1976) has been the “gold” standard since 1949
The mechanisms are not clear, but sub-terminal velocity after a collision-coalescence event and the super-terminal velocity after breakup have been proposed. The latter mechanisms depend on the collision frequency, and after a transient period of several 100 ms the drops recover to their terminal fall speeds (Szakáll et al, 2010)
Thériault et al (2015) using Computational Fluid Dynamics software documented in detail the airflow in the vicinity of the Double Fence Intercomparison Reference (DFIR) as well as inside the DFIR for two situations: (a) the flow is along the vertex of the octagon and (b) the flow is along the “flat” side (22.5◦) of the octagon

Summary

Introduction

Measurements of the terminal fall speed of drops by Gunn and Kinzer (1949) under laboratory conditions with the pressure correction of Beard (1976) has been the “gold” standard since 1949. The recent article by Ren et al (2020) uses direct numerical simulation (DNS) to study the drop dynamics of 2 and 3 mm sizes in turbulent flow Their conclusion was that both sized drops showed a decrease of the mean fall speeds (settling speeds) relative to terminal by 5 %–7 %. The Doppler spectrum measured by vertical pointing radar for which the terminal fall speed versus drop diameter relation is needed and fits to the Gunn–Kinzer are universally used (e.g., Williams and Gage, 2009) Another application is the numerical solution of the stochastic coalescence-breakup equation where the gravitational kernel involves the terminal fall speeds of the different sized drops that are assumed to follow fits to Gunn–Kinzer (e.g., Morrison and Grabowski, 2006). The results reported here provide an extension to our previous study by Thurai et al (2019)

Experimental set-up

Brief background

Findings