One of the most challenging aspects of tropical climate prediction is changes in clouds, and their spatial and vertical distributions. Deep convection is a fundamental part of this challenge. It is key in driving both the extreme weather and rainfall in the tropics, as well as influencing moisture and radiation budgets.
Organized convection exists across all scales in the tropics. This from mesoscale convective systems to tropical cyclones, all the way to planetary-scale features like the Madden-Julian Oscillation and Walker circulation.
With advances in computing power, and simple approximations for the behavior of the tropical atmosphere, we can study the behavior of convection "on its own free will". A fascinating consequence of this idealized modeling framework is convective "self-aggregation", the spontaneous organization of previously random clouds into one or several coherent clusters, surrounded by comparatively dry, sinking air. This is driven by feedbacks involving clouds, water vapor, radiation, and circulation, making dry areas drier and moist areas moister. In other words, the spatial variability of moisture increases, which can be used to quantify the physical feedbacks helping to organize convection!
I study how this process depends on the background rotation we impose in our model. In some cases, this "self-aggregated" region exists in the form of a tropical cyclone. I am thus able to study these as well, simulating the unique and critical interplay between convection and hurricanes in one space. While it falls largely outside the scope of my current work, studying convective organization in this way is also quite useful in diagnosing cloud feedbacks on climate, and how we might expect this to change with warming.