Limited-area modelling of stratocumulus over South-Eastern Pacific

This discussion paper is/has been under review for the journal Atmospheric Chemistry and Physics (ACP). Please refer to the corresponding final paper in ACP if available. Abstract This paper discusses the application of the Weather Research and Forecasting (WRF) model to limited-area modeling of atmospheric processes over the subtropical southeastern Pacific, with the emphasis on the stratocumulus-topped boundary layer. The simulations cover a period of 42 h selected from the VAMOS (Variability of the Amer-5 ican Monsoon Systems) Ocean-Cloud-Atmosphere-Land Study Regional Experiment (VOCALS-REx) field project conducted in the subtropical southeastern Pacific in Oc-tober and November 2008. Modeling results are compared with aircraft observations with the main conclusion that the simulated stratocumulus-topped boundary layer is significantly too shallow. This appears to be a combination of an already too shallow 10 boundary layer in the dataset used to provide initial and lateral boundary conditions, and deficiencies of the boundary-layer scheme in the WRF model. We suggest that the latter comes from the scheme confusing the cloud base change of the vertical temperature and moisture gradients with the change at the boundary-layer inversion. The model does simulate the formation of mesoscale cloud-free regions, arguably similar 15 to Pockets of Open Cells observed in nature. In the model, formation of these regions does not seem to be related to drizzle-induced transition from open-to closed-cell circulations as simulated by LES models. Instead, the cloud-free regions appear to result from mesoscale variations of the lower-tropspheric vertical velocity. Areas of negative vertical velocity with minima near the boundary layer top (a few cm s −1) seem to induce 20 direct evaporation of the cloud layer. Parameterized boundary entrainment may play some role as well. It remains to be seen in LES studies whether the mechanism seen in the limited-area model is realistic or if it is simply an artifact of interactions between resolved and parameterized processes.