Multiple scale simulations of stratocumulus clouds

This study introduces a flexible multiple nested modeling framework developed from the Weather Research & Forecasting (WRF) model for boundary layer cloud research. It features a nested large eddy simulation (LES) in a hindcasting mode that allows a direct comparison between simulations and observations. Using this approach, we simulated two stratocumulus cases observed at the southern Great Plains (SGP) observing site. The simulation-observation comparisons demonstrate that the multiple nested WRF well reproduces the cloud properties and vertical velocity fields detected by cloud radar and other remote sensing instruments. The simulations show that (1) the cloud fields can be strongly modulated by mesoscale organizations; (2) the vertical velocity variance and skewness in the boundary layer have well-defined diurnal variations and are strongly dependent on external forcings. But the in-cloud turbulent statistics appears to be more controlled by the internal turbulent processes; (3) the organized structures of stratocumulus are only slightly skewed, which is consistent with the cloud radar observations that the updraft fraction is close to 50%; and (4) the vertical distribution of cloud water and the associated high-order moments of continental stratocumulus can be substantially different from its marine counterpart due to different ambient thermodynamic structures. This study suggests that the multiple-scale WRF modeling system has a potential to be developed into a parameterization testbed in a hindcasting mode that can be used to address various issues regarding the parameterization of boundary layer clouds.