Near-LES modeling of Eastern Pacific Stratocumulus Drizzle and Cloud Variability in VOCALS

Recent studies have focused on the representation of low clouds as a critical component driving uncertainties in future climate change scenarios simulated by large-scale models. The southeast Pacific (SEP) region off the coast of Peru and Chile is home to one such extensive region of low clouds, set against a complicated environment of land-atmosphere-ocean interactions. The recent VOCALS Regional Experiment (VOCALS-Rex), which took place from 6 October to 2 December 2008, sampled properties of aerosol, cloud, and precipitation in order to understand better these interactions in marine stratocumulus. As part of the field campaign, the NOAA R/V Ronald H. Brown (RHB) fielded a suite of remote sensing and in situ instruments. Observations from C-band radar captured frequent incidences of unexpectedly strong convection, with radar reflectivity values as high as 42 dBZ. Preliminary analysis of the upper air soundings launched from the RHB reveals that these strongly drizzling cases were generally associated with boundary layers that were both moist and deep (1.4-2 km in height). Guided by preliminary analyses of the ship data, a number of possible factors affecting drizzle and cloud system variability were identified. We conducted numerical simulations employing a "near-LES" framework, with horizontal grid spacings of 150 m and a stretched vertical grid that was ~25 m near the surface and at the inversion. This near-LES configuration, together with a size-resolved formulation of microphysical processes, is able to capture mesoscale aspects of a precipitating cellular ensemble while also resolving turbulent boundary layer flow. We performed simulations to assess the relative importance of two of the identified factors: boundary layer depth and CCN concentration. The control simulation is loosely based on one of the unexpectedly strong events. We present results from this control simulation, as well as from sensitivity experiments in which the boundary layer depth is reduced and CCN concentration is decreased, and the joint simulation where both depth and CCN are varied. The simulation suite is consistent with the observations that suggest drizzle production is to a first order thermodynamically controlled by factors such as entrainment and large-scale vertical velocity that influence boundary layer depth and moisture The increase in drizzle associated with small changes (+200 m) in boundary layer thickness swamps changes to precipitation associated with CCN variations (40% decrease from 105 to 63 cm-3) These results are an important first step in attempting to detangle cause-and-effect for aerosol cloud-drizzle interactions over the SEP.