A subgrid model for chemical transformations in LES of the atmospheric boundary layer

The chemical lifetime of reactants in the atmosphere can vary within a wide range of time scales. Some highly reactive compounds such as OH and HO2 radicals have typical time scales smaller than a second. For such species, chemistry can be so active that the chemical compounds react in situ and are hardly transported by the flow. In large eddy simulations (LES) of atmospheric reacting flows, homogeneous and instantaneous mixing of reactants within a grid-cell is normally assumed. However, this assumption can result in large errors in the estimation of the reaction rates due to the fact that highly reactive species can be segregated or pre-mixed at small scales. Since this process occurs at scales smaller than the grid length (sub-grid process), it requires a subgrid parameterization.

In this paper, we propose a parameterization for the subgrid chemical transformations. Its formulation relies on the description of the subgrid covariance, i.e. the quantity that accounts for the mixing within a grid-cell, by using similarity arguments. The model is tested in large eddy simulations of a convective atmospheric boundary layer with reactive chemical species at different resolutions. The new model is able to capture the expected changes of magnitude of the subgrid covariance associated with changes in resolution. As a result, the simulations yield resolution-independent overall reaction rates (resolved plus subgrid reaction rates) and concentrations of reactants.