Scale interactions of organised convection in West Africa

Moist convection drives energy and moisture transfer across a range of scales, from microscopic phase changes within clouds, to latent heat release over hundreds of kilometres in tropical cyclones and thunderstorms. The complex feedbacks arising from scale interactions between moist convective processes are a source of major uncertainty in global climate models, and moist convection often drives high-impact weather with devastating effects.

Mesoscale convective systems (MCSs) are an example of high-impact organised convection, large enough to generate their own mesoscale circulations which can modulate synoptic-scale weather patterns, which in turn can modulate initiation and development of MCSs, leading to complex feedbacks. In West Africa, MCSs are the primary source of seasonal precipitation variability. The region is also home to well-characterised synoptic systems such as the African Easterly Jet and African Easterly Waves, which makes it a perfect natural laboratory for investigating convection-circulation feedbacks.

The thesis focuses on the use of regional high-resolution simulations which can, to some extent, represent mesoscale convection explicitly without a deep convection parameterisation, referred to as ‘convection-permitting’ (CP) simulations. Throughout the thesis, several techniques are used to explore moist convective scale interactions using such CP simulations. The first results chapter explores one such technique to diagnose scale interactions: the circulation budget. The
chapter examines how the closure of the circulation budget varies under changing environmental conditions, output timesteps, and spatial resolution. Having confirmed that instantaneous circulation budget terms are a valid quantification of atmospheric vorticity accumulation and tilting, the following chapter divides the terms into mean and anomalous components to understand the effects of different scales of atmospheric processes on synoptic-scale circulation in West Africa. It also highlights the representation of the dynamics of MCSs in the region and the vortex structures associated with them in CP and parameterised simulations, and how such
dynamical structures feed back on larger circulations. A third results chapter examines the statistics of the vortex structures and their potential for undergoing tropical cyclogenesis based on their size, vorticity intensity, and thermodynamic properties, highlighting another route for mesoscale convection to evolve into synoptic-scale systems.

The fourth results chapter utilises a dataset of tracked MCSs in a CP simulation and a parameterised simulation to examine the environments where MCSs are likely to be initiated, and how environments have been modulated by MCSs when they dissipate. Mechanisms for feedbacks between MCSs and themselves are explored, highlighting the roles of soil moisture feedbacks and low-level moisture in MCS initiation. The role of these mechanisms is further emphasised in the final results chapter, which compares CP simulations initialised with and without convective-scale atmospheric structures and evaluates how differences between their forecasts emerge. The comparison also demonstrates differences in the evolution of kinetic energy at different scales, and illustrates an transfer of energy upscale associated with the diurnal cycle of convection.

The work in this thesis demonstrates the capacity of convective-scale models for representing mesoscale-to-synoptic convective scale interactions, justifying further exploration of convective-scale models on larger domains despite the associated computational cost. Results could also inform development of parameterisations which account for the upscale impacts of organised mesoscale convection, thus improve atmospheric modelling across temporal and spatial scales.