MAGIC Assessment of a Stochastic Edmf Boundary Layer Parameterization

Abstract:

The northeast Pacific is representative of subtropical ocean basins with high-albedo regions of persistent stratocumulus clouds that transition to low-albedo regions of shallow convection. The accurate modeling of this system is a longstanding and critical problem in climate science. We use data from the recent ship-based Marine ARM GPCI Investigation of Clouds (MAGIC) campaign in the northeast Pacific to evaluate the skill of a unified stochastic Eddy-Diffusivity/Mass Flux (EDMF) boundary layer model. The MAGIC campaign data provides a nearly ideal validation framework, as it samples marine stratocumulus and cumulus regimes and the transition between them. We classify MAGIC scenes by cloud type, and produce probability distributions for the key EDMF forcing/initialization and output variables by cloud type. We initialize the EDMF model with the MAGIC input distributions and compare the output to the MAGIC data. Through this assessment we demonstrate improvements in the unified model's ability to handle a variety of boundary layer regimes.