Content area
Full text
ABSTRACT
A number of large eddy simulations with the Regional Atmospheric Modeling System have been made to study the sensitivity of shallow marine cumulus convection to different microphysics and radiation schemes. In particular, the sensitivity of shallow marine cumulus convection to drizzle and radiation effects, and how drizzle and radiation modify turbulent fluxes, are investigated.
It is shown that for the case of prescribed radiative heating, drizzle-albeit very slight-leads to reduced buoyancy fluxes and less turbulence. Consequently, drizzling boundary layers appear to entrain less than their nondrizzling counterpart. Heavy drizzle events are simulated in association with deeper clouds as high as 2 km, even though the majority of clouds are only a few hundred meters deep. A heavier and longer lasting drizzle episode associated with a deeper boundary layer is produced when a two-stream radiative parameterization replaces the prescribed radiative heating in the simulation. Simulated surface precipitation rates agree reasonably well with observations. The greatest alteration in boundary layer structure is obtained when radiative heating interacts explicitly with the broadened drop distribution associated with drizzle formation.
1. Introduction
Accurate representation of cloud-related processes in general circulation and weather prediction models is a fundamental and challenging problem in atmospheric research, and the major objective of the Global Energy and Water Cycle Experiment (GEWEX) Cloud Systems Study (GCSS) program (Browning 1994).
As a participant of the GCSS intercomparison project Working Group 1, or boundary layer group, we have conducted large eddy simulations (LESs) of shallow cumulus convection that occurred during the Barbados Oceanographic and Meteorological Experiment (BOMEX) (Davidson 1968). Data collected during phase 3 (22-26 June 1969) of BOMEX are chosen for the shallow cumulus convection simulation. This period is marked by relatively undisturbed trade wind weather. It is a simple case that has no mesoscale complications and no transitions from cumulus to stratocumulus cloud. Nonetheless, radar observations provided a mean precipitation rate of 0.2 mm day^sup -1^ during the period of interest (Holland and Rasmusson 1973). As a rule of thumb, one would expect precipitation from trade wind cumuli in the region when they are greater than 2000 m deep (J. Simpson 1989, personal communication).
Data collected during BOMEX have been used in a number of studies. Siebesma and Cuijpers (1995) conducted LES of...