New J. Phys. 20, 023001 (2018)

One reason that weather forecasting is so frustrating is the breadth of scales involved, from global weather patterns to microscale processes affecting, for example, raindrop formation. Connecting these domains remains a substantial challenge in numerical weather prediction. But progress is underway, including the development of a ‘cloud microphysics simulator’, reported by Izumi Saito and Toshiyuki Gotoh.

Credit: Matthias Scholz/Alamy Stock Photo

Saito and Gotoh introduced a computer model that simultaneously resolves turbulent motion in cumulus clouds (pictured) and the evolution of droplets to raindrops therein. To do so, they had to take into account a broad range of small-scale processes, from condensation and evaporation to collisions and coalescence of droplets. By covering large spatial and long temporal scales simultaneously, the pair ensured that they could capture the interplay between turbulence and droplets.

The model reproduced a number of known features, such as turbulence-induced collisions accelerating rain initiation and droplets modifying turbulence. The ability to see such effects in direct numerical simulations — rather than simply relying on general assumptions — should help to refine the tools that determine whether or not we need to carry an umbrella.