Person:

Nie, Ji

Our limited knowledge of convection and its poor representation in climate models is one of the factors that most hamper our ability to understand and predict the climate system. In this thesis, the dynamics of shallow cumulus convection are probed using Large-eddy simulations (LES) and simple models.

Aqueous phase reactions are important, sometimes dominant (e.g., for SO2), pathways for the oxidation of air pollutants at the local and/or global scale. In many current chemical transport models (CTMs), the transport and aqueous reactions of chemical species are treated as split processes, and the subgrid-scale heterogeneity between cloudy and environmental air is not considered. Here using large eddy simulation (LES) with idealized aqueous reactions mimicking the oxidation of surface-originated SO2 by H2O2 in shallow cumuli, we show that the eddy diffusivity mass flux (EDMF) approach with a bulk plume can represent those processes quite well when entrainment/detrainment rates and eddy diffusivity are diagnosed using a conservative thermodynamic variable such as total water content. The reason is that a typical aqueous reaction such as SO2 aqueous oxidation is relatively slow compared to the in-cloud residence time of air parcels in shallow cumuli. As a result, the surface-originated SO2 is well correlated with and behaves like conservative thermodynamic variables that also have sources at the surface. Experiments with various reaction rate constants and relative abundances of SO2 and H2O2 indicate that when the reaction timescale approaches the in-cloud residence time of air parcels, the errors of the bulk plume approach start to increase. Treating chemical tracer transport and aqueous reaction as split processes leads to significant errors, especially when the reaction is fast compared to the in-cloud residence time. Overall, the EDMF approach shows large improvement over the CTM-like treatments in matching the LES results.