Convective Dynamics of the Tropical Atmosphere in Three Idealized Approaches

Abstract

Atmospheric convection organized at large spatial scales significantly influences precipitation patterns and weather events in tropical and subtropical regions, and has a rich, two-way interaction with Earth's climate. Tropical cyclones, mesoscale convective complexes, heat lows, and the rainbands of the Intertropical Convergence Zone are all examples of such large-scale convective organization, but despite their relevance for human life and ecosystems, the mechanisms that govern their formation and many of their characteristics are not fully understood. This work presents three studies of convective dynamics in the tropical atmosphere using idealized frameworks. In the first part, we use direct numerical simulations of simple setups of rotating dry convection based on the Rayleigh-Bénard system to study minimal conditions that produce large-scale convective organization, and the spontaneous formation of tropical-cyclone-like structures. We find that the latter form more readily for a particular set of controlling parameters and thermal boundary conditions, characterized by a slow enough rotation rate, asymmetry of the heat fluxes at the boundaries, effective internal cooling, and a dependence of the low-level heat flux on the overlying flow. In the second part, we use rotating tank experiments of turbulent convection to probe further some of the findings of the first part, particularly the formation of large-scale cyclonic flows with top-bottom asymmetric, flux-based thermal boundary conditions, in a setup with hot water insulated at the bottom and sides, and cooling freely to the air above. We find large, persistent cyclonic vortices in experiments with a similar range of governing parameters as the results from the numerical simulations in the first part, particularly for cases where the convective time scale is shorter than the rotational time scale, that is, for convective Rossby numbers greater than about 1. Our approach in the first two parts seeks to narrow the disciplinary gap between traditional turbulence research and the physics of atmospheric convective organization and tropical cyclones. In the third part, we turn to the topic of the mechanisms that drive precipitation in the tropics at scales of tens to a few hundred kilometers. Using cloud-resolving model simulations, we study how rainfall responds to imposed anomalies in the surface temperature, the atmospheric heating rate at different heights, and the pressure gradients that drive the winds near the surface. We find that such forcings lead to self-consistent but different relationships between the amount of rainfall produced and the net heating of the atmosphere, quantified by the Normalized Gross Moist Stability. We show that the spatial extent of the forcings affects how well this relationship can be inferred from horizontally-averaged atmospheric properties. In contrast, we find that the relationship between rainfall and the average relative humidity in the atmosphere falls onto the same curve for all types of environmental forcings considered. As a general contribution, the three parts of this work highlight the fruitfulness of a diverse set of idealized approaches in deriving a more mechanistic understanding of the convective dynamics of the atmosphere.