Quantifying and parameterizing the transport of sub-cloud layer moisture and reactants by shallow cumulus clouds over land

We investigate the vertical transport of moisture and atmospheric chemical reactants from the sub-cloud layer to the cumulus cloud layer that is related to the kinematic mass flux driven by shallow convection over land. The dynamical and chemical assumptions needed for mesoscale and global chemistry-transport model parameterizations are systematically analysed using numerical experiments performed by a Large-Eddy Simulation (LES) model.

First, we identify and discuss the four primary feedback mechanisms between sub-cloud layer dynamics and mass-flux transport by shallow cumulus clouds for typical mid-latitude conditions. These mechanisms link shallow cumulus properties, such as cloud cover and kinematic mass flux, to mixed-layer drying and heating, to changing the moisture variability at the sub-cloud layer top and to adjusting the entrainment velocity. Based on this analysis and LES experiments, we design parameterizations for cloud properties and mass-flux transport of air and moisture that can be applied to large-scale models. As an intermediate step, we incorporate these parameterizations in a conceptual mixed-layer model, which enables us to study these interplays in more detail. By comparing the results of this model with LES case studies that are based on observations, we show for a wide range of conditions that the new parameterizations enable the conceptual model to reproduce the sub-cloud layer dynamics and the four aforementioned feedbacks. However, by considering heterogeneous sensible and latent heat fluxes at the surface, we further demonstrate that the parameterizations are sensitive to specific boundary conditions due to changes in the boundary-layer dynamics. Second, we extend the investigation to determine whether the parameterizations are suitable for tropical conditions and to represent the transport of reactants. The numerical experiments in this analysis are inspired by observations over the Amazon during the dry season. We show that the mass-flux induced removal can significantly affect reactants. For example, isoprene, a key atmospheric compound over the tropical rain forest, decreases by 8.5 % hr^-1 on average and by 15 % hr^-1 at maximum due to this process. The new general mass-flux parameterizations for the transport of chemical reactants agree satisfactorily with the transport that is numerically resolved by LES for individual species, except for some reactants like O₃, NO and NO₂. The latter is caused by the local partitioning of reactants, influenced by UV radiation extinction by clouds and small-scale variability of ambient atmospheric compounds. By considering the longer lived NO_x (NO + NO₂), the transport of these species is well represented by the parameterization as well.