Investigating convective aggregation with increasingly realistic conditions

Moist deep convection is important in the tropics for distributing heat, moisture and momentum. However its intensity and lifetime can vary greatly depending on its environment. This convection can be organised in a number of different ways, such as squall lines, mesoscale convective systems (MCS) or supercell clusters which can develop into planetary envelopes such as the Madden-Julien Oscillation. Convection that aggregates, such as MCS, contributes a signifcant amount of the total tropical cloudiness and roughly half of tropical precipitation (Tobin et al., 2012). Whilst the physical processes underlying the organisation of convection have been investigated thoroughly, there remains a problem in representing it accurately in models (Tobin et al., 2013).The interactions between aerosols and clouds are the focus of a great deal of research. Aerosols have the potential to increase the number of cloud droplets, hence brightening the cloud (the 'Twomey effect')(Twomey, 1977). They can also suppress precipitation and therefore extend the lifetime of a cloud (Albrecht,1989).These and other interactions affect different cloud types in different ways, making it difficult to constrain an exact aerosol effect on clouds. This is one of the key reasons why clouds and aerosols are so uncertain in models (Flato et al., 2013).An idealised version of the tropical atmosphere can be modelled in radiative-convection equilibrium (RCE).This is the equilibrium state the atmosphere would be in in the absence of lateral energy transport.Due to this, many studies of tropical deep convection are conducted in this simplified set-up. Typical studies of RCE often use simplifications, such as a lack of rotation, constant insolation and a constant sea-surface temperature. This idealised set-up is useful for focussing on the key physical drivers of convection such as radiation, humidity, surface uxes and convectively driven circulations (Becker, 2018).One type of organisation that occurs in RCE simulations is convective self-aggregation. This is the spontaneous clumping of several convective cells into one supercluster, whilst leaving the remainder of the domain extremely dry (Holloway et al., 2017), which causes a drying of the mean state. This makes the domain more effective at radiating heat which has large impacts for global circulations and energy budgets (Coppin and Bony, 2017). It also affects cloud coverage with a decrease in high-clouds.The combination of these effects show that self-aggregation has the potential to change the climate, thus emphasising the need for a better understanding (Wing, 2019). The role of convective aggregation in climate was identified by the World Climate Research Program as a grand challenge due to its inherent implications for climate sensitivity and the hydrological cycle (Bony et al., 2015). This self-aggregation has similar properties to observed aggregated convection (such as MCS) however the link between the simulated and observed aggregation remains unclear.Research has shown some dependence of convective aggregation on sea-surface temperature (SST),however the exact nature of this relationship is currently unknown. Some studies found aggregation occurs only at warm SSTs (Held et al., 1993; Khairoutdinov and Emanuel, 2010; Wing and Emanuel, 2014), yet other studies since these have found the relationship to be more complex than this (Wing and Cronin,2016; Coppin and Bony, 2015, 2017; Cronin and Wing, 2017).This is a key area of research in the field currently.Due to its interactions with humidity distributions and radiation budgets, convective aggregation also has the ability to affect climate sensitivity. Concrete conclusions on this are hindered by the lack of synthesis in temperature dependence results.Evidence suggests that increased aggregation reduces the climate sensitivity however it is unclear how much of this is represented in models, whether it changes with an interactive ocean surface.