ICTS

This talk will do a detailed scale analysis of the tropical atmosphere.

It is well-known that 'dry' atmospheric dynamics (of velocity, temperature, etc.) can be decomposed into slow and fast, or balanced and unbalanced, components. However, moisture is typically neglected. I will show that water vapor and clouds can also be decomposed into slow and fast components. As one application, water vapor and cloud dynamics may need to be added onto the long history of the importance of balanced-unbalanced concepts for data assimilation and forecasting.
 

The intertropical convergence zone (ITCZ), as the name suggests, is the region within the tropics where the northerly and southerly trade winds converge, resulting in high cloudiness and intense precipitation. As the ascending branch of the Hadley circulation, the ITCZ is also a key component of the general circulation of the atmosphere. The position, area and intensity of the ITCZ are determined by thermodynamical and dynamical processes that span a wide range of spatial and temporal scales.

In this lecture, I will focus on the dynamics of the ITCZ, and present the various momentum balance models that are used to understand the winds and convergence in the trade wind boundary layer. While linear balance models have had considerable success in explaining the winds and convergence in the tropics, observations suggest that the nonlinear horizontal advection is a key process in the boundary layer of a cross-equatorial flow. Recent studies using dry dynamical models have supported this view and showed that high horizontal resolution is needed to accurately capture the magnitude of zonal and meridional advection. Finally, I will present the findings of my recent work where a global, coupled atmosphere-ocean-land simulation at kilometre-scale horizontal resolution is used to evaluate the momentum balances in the off-equatorial ITCZ in East Pacific and Atlantic basins.

Reversal of the sign of the Coriolis force across the equator creates a waveguide along the equator home to a wide spectrum of waves. Initially found as solutions to the shallow water model on the equatorial beta plane, they include Rossby waves, gravity waves, and mixed forms. I discuss the leading solution forms and their propagation mechanisms. Then, I give special focus to the Kelvin wave, a non dispersive disturbance of zonal wind. I show how Kelvin waves interacting with the mean atmospheric flow gives rise to the phenomenon known as the Madden Julian Oscillation.

Understanding cloud microphysical processes remains a fundamental challenge in atmospheric science, owing to the vast range of spatial and temporal scales involved within a cloud system. A rigorous investigation of these processes is essential for improving our understanding of the role of clouds in the Earth's climate, with direct implications for monsoon prediction and weather forecasting. Conventional numerical models struggle to accurately represent small-scale phenomena such as droplet dynamics, aerosol activation, and turbulent interactions at the Kolmogorov scale. Furthermore, processes such as entrainment and mixing within clouds critically govern the microphysical composition, including droplet number concentration and size distribution, which in turn influence the cloud's radiative properties and its broader role in the climate system. Direct Numerical Simulation (DNS), by resolving all physically relevant scales of turbulent flow without parameterisation, serves as a powerful virtual laboratory for investigating these processes from first principles. In this talk, the application of DNS to study droplet dynamics, aerosol activation, and turbulent characteristics in cloud-like environments will be presented. Additionally, the integration of machine learning techniques for extracting turbulent flow features from DNS data will be discussed. Additionally, a framework termed "Scaled-up DNS" will be discussed, which aims to bridge the resolution gap between DNS and Large Eddy Simulation (LES) grids, offering a promising pathway towards improved sub-grid parameterisation in cloud-resolving models.

Keywords: Direct Numerical Simulation, droplet dynamics, aerosol activation, cloudmicrophysics, turbulence, machine learning, Scaled-up DNS, LES parameterisation

The study of cloud processes at the micro-scale poses challenges due to the broad spectrum of scales involved within the cloud system. A thorough investigation of these processes is crucial for obtaining a deeper understanding of the importance of clouds in human affairs, particularly in fields such as monsoon prediction. In general, numerical models that aim to replicate cloud phenomena encounter difficulties accurately representing complex small-scale phenomena, such as droplet dynamics and turbulent features at the Kolmogorov scale. Also, it is important to keep in mind that the processes of entrainment and mixing that happen inside clouds has a big effect on the cloud's microphysical composition, which includes the number and size of cloud droplets. These parameters have a significant impact on the cloud's macroscopic radiative properties and its overall function within the climate system. Direct numerical simulation (DNS), a computationally demanding method, is the one the best way to study these processes. In this talk, the use of DNS to investigate droplet dynamics, aerosol activation, and turbulent characteristics will be discussed. It will also go through some machine learning techniques for analyzing turbulent properties using DNS data, as well as the introduction of a novel strategy termed "Scaled-up DNS" and how it may be helpful in providing the information to LES grids.
Keywords: Direct Numerical Simulation for Droplet and Aerosol dynamics: Bird eye view

In this talk, we will first discuss some of the fundamental properties of moist convection. A question of interest is what sets vertical velocities in clouds, and we will investigate this question in idealized sensitivity simulations. We will then investigate the physical processes responsible for the spatial organization of moist convection into large storms.