Abstract/Summary

Tropical weather is dominated by convection which is organised across a wide range of spatial and temporal scales. The degree of convective organisation has important consequences for weather and climate. The uncertainty in the response of convective organisation to a warming climate is one of the largest sources of uncertainty in climate sensitivity estimates. Self-aggregation is the process in which convection spontaneously clusters despite homogeneous initial conditions and forcing. It has been the focus of many recent studies because of its implications for real world weather and climate. Cloud-radiation interactions have been shown to be crucial drivers and maintainers of aggregation. Yet there remains uncertainty in their role in self-aggregation. In this thesis, we develop a framework to study aggregation and quantify the contributions of radiative interactions with different cloud types to aggregation. We study models that form part of the RadiativeConvective Equilibrium Model Intercomparison Project, comparing models with explicit and parameterised convection across a range of sea surface temperatures (SSTs). We find that longwave interactions with high-topped cloud and clear regions, as well as shortwave interactions with water vapour are key drivers and/or maintainers of aggregation. Their influence on aggregation tends to decrease with SST, but the rate of aggregation remains similar. We find the strength of these interactions strongly correlates with the rate of aggregation in parameterised convection simulations, yet the rate of aggregation in explicit simulations is more strongly influenced by circulations. Parameterised convection simulations often have stronger longwave interactions with high-topped cloud than explicit simulations, resulting in faster aggregation. We find that by artificially reducing this longwave feedback in parameterised simulations, the aggregation behaves more similarly to explicit simulations. This highlights that global weather and climate models may be able to model the effects of real-world aggregation more accurately given an accurate representation of cloud-radiation interactions.