Statistical representation ("parametrization") of convective clouds, their precipitation, phase changes and transports of heat, water and momentum is a critical component of weather and climate prediction.
The APP convection group conducts research and development in convection parametrization as well as owning and maintaining the Unified Model convection code. Currently we are evaluating different approaches to convection parametrization and specifically a new turbulence-based approach. This involves both a warm scheme for shallow and congestus clouds, and a deep convection scheme suitable for glaciated convection. At the same time we are working to improve incrementally the existing mass-flux scheme. All this work is in partnership with groups responsible for Unified Model assessment in weather and climate prediction.
We are also involved in high-resolution modelling through the Cascade project (collaborative with universities).
Incremental improvement of our mass-flux convection scheme
Evaluation of new approaches to convection parametrization
Research on stochastic physics
Improvements to convective diurnal cycle
Testing of a turbulence-based convection scheme, including both warm
clouds (shallow cumulus/congestus) and deep convection components
Assessment of stochastic physics development and strategy