Conceptual model of organized thunderstorm clusters under wind shear

Mesoscale convective systems (MCSs) are organized thunderstorm clusters which span over 100 km horizontally. They are responsible for producing the majority of rainfall in the tropics and can cause extreme precipitation events. Over the tropical ocean, MCSs can develop into tropical cyclones.

Recent work found that including a diurnal cycle -  to mimic surface temperature variations between night and day over land - enables convective self-aggregation (CSA) in idealised cloud-resolving atmospheric simulations, which means that there is a persistent spatial separation into dry patches and patches with a lot of rainfall. In simulations with a constant surface temperature – which mimic the situation over the ocean – no strong aggregation is observed (Jensen et al., JAMES, 2022).  

This study investigates the effects of wind shear in simulations with and without a diurnal cycle. Including wind shear as an imposed idealised wind profile that was derived from ERA5 data makes the simulations more realistic. Similar to the simulations without wind shear, in the case of a diurnal cycle and wind shear, self-aggregation is observed. For the constant surface temperature simulation, the aggregation stays low in comparison to the diurnal case but including wind shear increases the reached normalized spatial variance level by one order of magnitude (Kruse, University of Copenhagen, 2024).

We observe that in the simulation with wind shear and a diurnal cycle (DIU Wind), stripes with either abundant or no precipitation form along the imposed wind direction. In contrast, in the simulation with wind shear and constant surface temperature (OCEAN Wind), the clouds - less pronounced - tend to form stripes perpendicular to the wind direction which are advected by the wind.  In the DIU Wind simulation, the advection velocity of the outgoing longwave radiation at the top of the atmosphere slows down simultaneously with the formation of CSA, and converges with the wind velocity and advection velocity of the moisture field at 3500 m, indicating that the onset of CSA and the slowdown are related.  We also observe multi-day oscillations in the advection velocities in the DIU Wind simulation that will be further explored.

Our work has implications for the understanding of the organization of the convectively-driven moisture field over continent and its advection over the ocean - where it can pre-condition tropical cyclogenesis.