TY - JOUR

T1 - Evolution and Features of Dust Devil-Like Vortices in Turbulent Rayleigh-Bénard Convection - A Numerical Study Using Direct Numerical Simulation

AU - Giersch, S.

AU - Raasch, S.

N1 - Funding Information: This work was funded by the German Research Foundation (DFG) under Grant RA 617/31‐1. Open access funding enabled and organized by Projekt DEAL.

PY - 2021/4/8

Y1 - 2021/4/8

N2 - Dust devils are convective vortices with a vertical axis of rotation that are made visible by entrained soil particles. These soil particles contribute to the atmospheric aerosol input, influencing the Earth radiation budget. Quantifying this contribution requires reliable information about the statistics of dust devils, their formation process, and how they are maintained. In the past, this information was mainly derived from field experiments and large-eddy simulations (LESs). Field experiments suffer from the erratic occurrence of dust devils and the limited area that can be monitored reliably. In LESs, dust devils cannot be resolved completely, especially close to the ground. Additionally, they are affected by numerical features of surface boundary conditions, as well as subgrid-scale models in an unknown way. To mitigate these limitations, we employ direct numerical simulations (DNSs) to improve our understanding of dust devils. We comprehensively investigate the statistics and structure of dust devils for Rayleigh numbers up to 1011 using DNS of Rayleigh-Bénard convection between two plates for the first time. We find that dust devil-like structures occur in DNS with Rayleigh numbers much lower than in the atmosphere (≥107). These results support previous DNS studies in which vortices with vertical axes were observed but not further investigated. The dust devil statistics strongly depend on the Rayleigh number and velocity boundary conditions, but depend little on the aspect ratio of the model domain. Simulated dust devils show very similar properties to convective vortices analyzed in LESs of the atmospheric boundary layer.

AB - Dust devils are convective vortices with a vertical axis of rotation that are made visible by entrained soil particles. These soil particles contribute to the atmospheric aerosol input, influencing the Earth radiation budget. Quantifying this contribution requires reliable information about the statistics of dust devils, their formation process, and how they are maintained. In the past, this information was mainly derived from field experiments and large-eddy simulations (LESs). Field experiments suffer from the erratic occurrence of dust devils and the limited area that can be monitored reliably. In LESs, dust devils cannot be resolved completely, especially close to the ground. Additionally, they are affected by numerical features of surface boundary conditions, as well as subgrid-scale models in an unknown way. To mitigate these limitations, we employ direct numerical simulations (DNSs) to improve our understanding of dust devils. We comprehensively investigate the statistics and structure of dust devils for Rayleigh numbers up to 1011 using DNS of Rayleigh-Bénard convection between two plates for the first time. We find that dust devil-like structures occur in DNS with Rayleigh numbers much lower than in the atmosphere (≥107). These results support previous DNS studies in which vortices with vertical axes were observed but not further investigated. The dust devil statistics strongly depend on the Rayleigh number and velocity boundary conditions, but depend little on the aspect ratio of the model domain. Simulated dust devils show very similar properties to convective vortices analyzed in LESs of the atmospheric boundary layer.

KW - direct numerical simulation

KW - dust devils

KW - Rayleigh-Bénard convection

UR - http://www.scopus.com/inward/record.url?scp=85104272164&partnerID=8YFLogxK

U2 - 10.1029/2020JD034334

DO - 10.1029/2020JD034334

M3 - Article

AN - SCOPUS:85104272164

VL - 126

JO - Journal of Geophysical Research: Atmospheres

JF - Journal of Geophysical Research: Atmospheres

SN - 2169-897X

IS - 7

M1 - e2020JD034334

ER -