Modeling aerosol growth by aqueous chemistry in a nonprecipitating stratiform cloud

Abstract

The modification of sulfate aerosol by the activation, aqueous chemistry, resuspension cycle in a shallow marine stratiform cloud is modeled using a large‐eddy simulation (LES) model and a trajectory ensemble model (TEM). Both dynamical frameworks are coupled with a new microphysics module based on a 2‐D joint size distribution function representing both interstitial and cloud particles. Particle mixing, which is represented in LES but not TEM, leads to slightly broader droplet spectra and 6% lower droplet number on average. TEM simulations with a 1‐D moving‐bin microphysics module, also included in the study, predict far narrower droplet spectra and 15% higher droplet number that are closer to observations except near the cloud boundaries where mixing appears to be important. Despite these differences in simulated droplet spectra, predicted changes in aerosol spectra due to aqueous chemistry are consistent among all three model configurations. The TEM model is used to evaluate assumptions about liquid water partitioning among activated cloud condensation nuclei (CCN). These assumptions are used in large‐scale models to map the bulk aqueous chemistry sulfate production to the changes in the aerosol size distribution. Previously used assumptions, such as droplet mass being independent of CCN size or droplet mass being proportional to CCN mass, do not perform well in the considered case. The aerosol spectra changes from aqueous chemistry using these two assumptions differ markedly from the spectra changes with the explicitly predicted water partitioning and have root‐mean‐squared deviations normalized by the mean (RMSDnorm) of 0.51 and 1.17. These deviations greatly exceed the uncertainties because of the treatment of mixing in the TEM and numerical diffusion in the fixed‐bin representation for which the RMSDnorm is about 0.15. Instead, the explicitly predicted water partitioning suggests that in the considered case the mean droplet mass is proportional to CCN dry size.