Multi-thermals and high concentrations of secondary ice: a modelling study of convective clouds during the Ice in Clouds Experiment – Dust (ICE-D) campaign

Zhiqiang Cui,Gary Lloyd,Rachel Hawker,Alan Blyth,Keith Bower,Yahui Huang,Paul Field,Thomas Choularton,Lindsay Bennett

doi:10.5194/acp-22-1649-2022

Abstract

Abstract. This paper examines the mechanisms responsible for the production of ice in convective clouds influenced by mineral dust. Observations were made in the Ice in Clouds Experiment – Dust (ICE-D) field campaign which took place in the vicinity of Cape Verde during August 2015. Measurements made with instruments on the Facility for Airborne Atmospheric Measurements (FAAM) aircraft through the clouds on 21 August showed that ice particles were observed in high concentrations at temperatures greater than about −8 ∘C. Sensitivity studies were performed using existing parameterization schemes in a cloud model to explore the impact of the freezing onset temperature, the efficiency of freezing, mineral dust as efficient ice nuclei, and multi-thermals on secondary ice production by the rime-splintering process. The simulation with the default Morrison microphysics scheme (Morrison et al., 2005) that involved a single thermal produced a concentration of secondary ice that was much lower than the observed value of total ice number concentration. Relaxing the onset temperature to a higher value, enhancing the freezing efficiency, or combinations of these increased the secondary ice particle concentration but not by a sufficient amount. Simulations that involved only dust particles as ice-nucleating particles produced a lower concentration of secondary ice particles, since the freezing onset temperature is low. The simulations implicate that a higher concentration of ice-nucleating particles with a higher freezing onset temperature may explain some of the observed high concentrations of secondary ice. However, a simulation with two thermals that used the original Morrison scheme without enhancement of the freezing efficiency or relaxation of the onset temperature produced the greatest concentration of secondary ice particles. It did so because of the increased time that graupel particles were exposed to significant cloud liquid water in the Hallett–Mossop temperature zone. The forward-facing camera and measurements of the vertical wind in repeated passes of the same cloud suggested that these tropical clouds contained multiple thermals. It is possible of course that several mechanisms, some of them only recently discovered, may be responsible for producing the ice particles in clouds. This study highlights the fact that the dynamics of the clouds likely play an important role in producing high concentrations of secondary ice particles in clouds.

Highlights

Ice particles in clouds contribute to the formation of more than half of the world’s precipitation and greatly enhance the amount of precipitation process over the warm-rain-only process (McFarquhar et al, 2017)
We report on a modelling study designed to explore several aspects of the production of ice particles, that is, the impact of the freezing onset temperature and the impact of the efficiency of freezing in microphysics schemes, to inspect the relationship between the measured dust as efficient ice-nucleating particles (INPs) and the secondary ice production, and to investigate the role of multi-thermals in producing high secondary ice concentrations
Additional simulations were run with adjusted onset freezing temperature, freezing efficiency, the Paukert scheme for dust-only INPs, and a two-thermal simulation

Summary

Introduction

Ice particles in clouds contribute to the formation of more than half of the world’s precipitation and greatly enhance the amount of precipitation process over the warm-rain-only process (McFarquhar et al, 2017). Precipitation in deep convective clouds is closely related to the riming and melting of ice particles (Cui et al, 2011). Dust particles can serve as INPs at higher temperatures if the fraction of potassium-rich feldspar (K-feldspar) is high because feldspar particles have high ice-nucleating active sites (Atkinson et al, 2013). A full functionalization of the ice-nucleating sites of a feldspar particle with hydroxyl groups enables a strong bonding to the prismatic plane of ice and prompts the formation of ice crystals on the surface of the feldspar (Kiselev et al, 2017), which favours a higher freezing temperature. Burrows et al (2013) suggested strong regional differences in the importance of marine biogenic INPs relative to dust INPs

Objectives

Results

Conclusion