Extended Abstract (100K)5.2
Representation of marine stratocumulus in regional forecast models: Role of subgrid inhomogeneity
David B. Mechem, CIMMS/Univ. of Oklahoma, Norman, OK; and Y. L. Kogan
A simple treatment of subgrid scale variability has been added to a bulk drizzle microphysics parameterization incorporated into the US Navy Coupled Ocean-Atmosphere Mesoscale Prediction System (COAMPS). The bulk drizzle parameterization provides a more physically based treatment for marine stratocumulus, but microphysical process rates computed from grid point variable values assume the entire grid volume is characterized by that value. Grid volume average process rates, which account for subgrid inhomogeneity, arise by integrating the rate over the probability distribution function (PDF) of the microphysical variables.
A bias factor is derived for the autoconversion rate by assuming the subgrid PDFs of cloud water and number concentration may be represented as gamma distributions. Distribution shape parameters are a function of the grid point values and are constrained using LES data for unbroken and broken stratocumulus cloud fields. The bias factor is extremely sensitive to the nature of the distributions, particularly in a situation where the process rate exponents are large in magnitude. Though not directly addressed, cloud fraction information is implicitly present in the LES PDFs.
COAMPS was run for a typical summer stratocumulus cloud system off the California coast. Including information about subgrid heterogeneity for the 2 km mesh results in larger autoconversion rates that lead to increased drizzle production and a reduction in liquid water path. Cloud breakup associated with the decoupled boundary layer and its subsequent transition from a stratocumulus to a boundary layer cumulus regime is enhanced.
Session 5, Atmospheric Modeling
Friday, 9 November 2001, 10:45 AM-12:15 PM
Previous paper Next paper