Numerical diffusion and turbulent mixing in convective self-aggregation

What is it about?

Convection acts to transport moisture from the surface to the free troposphere and to form clouds. When clouds develop in a dry environment, they can be diluted by turbulent mixing. On the contrary, their deepening is favored by mixing within a moist environment. Therefore, turbulent mixing will favor convection on moist regions making them moister, while the opposite is true for dry regions. This is called the water-vapor convection feedback (or entrainment feedback). This feedback plays a relevant role in the creation of humidity perturbations in the free troposphere, destabilizing the radiative-convective equilibrium state, where convection is homogeneously distributed, into a stable state where convection clumps into a single cluster. This work analyzes this phenomenon, called Convective Self-Aggregation, and its sensitivity to turbulence models and implicit numerical dissipation, the main sources of mixing when grid resolution of O(1 km) are employed. In particular, it is found that large dissipation at small-scales, is necessary to develop large humidity fluctuations in the free-troposphere and favors the creation of large dry areas free of convection. Therefore a correct representation of sub-grid scale mixing processes is necessary to capture the convective self-aggregation phenomenon in high-resolution climate models.

Featured Image

Why is it important?

Turbulence is parametrized in high-resolution climate models, then mixing is resolved through parametrized terms and numerical ones at these scales O(1 km). Since convection is sensitive when aggregates to mixing processes at sub-grid scale, this paper underlines the importance of proper resolution of mixing provcesses when high-resolution climate models are concerned.

Perspectives