Abstract/Summary

Representing convection at kilometre-scale resolution (also known as the ‘grey-zone’) has been a challenge for numerical modellers. Conventional parameterisation approaches rely on assumptions only valid for very coarse resolution simulation and so cannot be applied at this resolution. However, the ‘grey zone’ is not sufficiently fine to fully resolve convective updraughts. A new approach towards solving the problem is proposed. The new Stochastic Explicitly Prognostic Mass Flux (SEP-MF) scheme features a stochastic triggering function based on the occurrence and strength of boundary layer thermals, as well as a fully prognostic (growth) plume model in which prognostic equations representing the evolution of sub-grid convection are solved explicitly. The SEP-MF scheme was tested in a case featuring a diurnal cycle of convection, a phenomenon which is known to be difficult to simulate. Results from this study suggest that the SEP-MF scheme is capable of producing scale-independent solutions of the diurnal cycle of convection, as well as improving the overall performance of kilometre-scale models in representing convection.