The objective of this project is to reduce the subtropical boundary layer cloud deficiencies and biases in oceanic upwelling regions by implementing, and evaluating, in E3SM a new unified boundary layer and convection parameterization based on the multi-plume Eddy-Diffusivity/Mass-Flux (EDMF) approach.

The Intergovernmental Panel on Climate Change (IPCC) has reiterated that clouds remain the largest source of uncertainty in climate projections. Studies have shown that subtropical boundary layer clouds are a primary reason for this model spread. Over the subtropical oceans, the prevalent cloud regimes are stratocumulus and shallow cumulus, capped by subsiding warm and dry air in the free troposphere. Stratocumulus is prevalent over the cool upwelling waters off the west coast of continents while cumulus is common over warmer subtropical waters. What controls the existence of these cloud regimes and the transition between them is still a matter of debate. This transition from stratocumulus to cumulus plays a key role in cloud-climate feedbacks because of the essential difference in terms of albedo between the two regimes. Current climate and weather prediction models (including E3SM) have difficulties in realistically simulating stratocumulus and the transition to cumulus, and its oceanic and climate impact in terms of cloud albedo and sea surface temperature (SST). The EDMF parameterization has been shown to improve the representation of the subtropical cloudy boundary layer, including cumulus and stratocumulus.

Eddy-Diffusivity/Mass-Flux (EDMF): The EDMF approach is based on the unification of concepts generally used for the parameterization of turbulence in the boundary layer (ED) and of moist convection (MF). The EDMF approach was first proposed by Siebesma and Teixeira (Proc. PBL AMS, 2000) and has been shown to represent well dry (Siebesma et al., J. Atmos. Sci., 2007) and moist convective boundary layers (Soares et al., Quart. J. Roy. Meteor. Soc., 2004). In the last few years the PI’s group has developed a new version of EDMF that is particularly well suited to simulate moist convective boundary layers and is able to represent in a realistic manner the dry boundary layer, stratocumulus, shallow and deep cumulus convection, and the transitions between these different regimes. This new version (Suselj et al.,J. Atmos. Sci., 2013) uses a multi-plume approach where (1) cloud base or surface variability of updraft properties is parameterized using a simple diagnostic variance equation; (2) these probability density functions (PDFs) of updraft properties are sampled with a Monte-Carlo method; and (3) lateral entrainment is parameterized in a stochastic manner (Romps and Kuang, J. Atmos. Sci., 2011). This new EDMF parameterization is scale-adaptive: (i) the updraft area term in the full EDMF decomposition of the subgrid vertical fluxes is not neglected (as is often done); and (ii) the number of plumes can be made dependent on the grid size by taking into account the typical convective cell size versus grid size.

E3SM implementation: In this project, we will implement and evaluate the new EDMF parameterization in the E3SM model. During the first half of the project, we will implement, and evaluate, the new EDMF into the E3SM Single Column Model (SCM), versus a variety of LES case-studies and observations from the ARM MAGIC field experiment. For the full 3D implementation, we will focus our evaluation on cloud and convection variables as observed by satellite instruments during present climate. In particular, we will evaluate how the new EDMF version of E3SM is able to simulate stratocumulus and the transition to cumulus clouds. We will also investigate the impact, and feedback, of the new EDMF E3SM simulations in the sea surface temperature off the west coast of continents (in a coupled model context).