Scale-aware Parameterization of Liquid Cloud Inhomogeneity and Its Effects on Simulated Climate in CESM

Abstract:

Cloud inhomogeneity is an important subject of physical parameterizations in in climate models. Subgrid-scale cloud variability can impact cloud microphysics, precipitation, and radiation. In the CESM, the subgrid-scale cloud liquid water concentration is represented by a gamma distribution function. The key shape parameter of the gamma distribution is fixed to an observation-based value. In reality, this shape parameter should depend on model grid scale and other atmosphere conditions. The objective of this study is to estimate and parameterize the shape parameter suitable for different model resolutions. Cloud measurements from the Atmospheric Measurement Program (ARM) are used. These include the ground-based long-term liquid water content data from Continuous Baseline Microphysical Retrieval (MICROBASE) product and Best Estimate Data Products. Larger grid scale and more unstable atmosphere are found to increase inhomogeneity and therefore have smaller shape parameter.

A scale-dependent parameterization of the shape parameter is presented based on the above observation information. When it is implemented in the CESM, spatially and temporally varying shape parameters are simulated with larger values in high latitudes and smaller values in low latitudes. Single column model and general circulation model (GCM) sensitivity experiments are conducted to understand the effects of varying shape parameter on climate simulations. Compared with fixed shape parameter setting, varying shape parameter simulation tends to increase liquid water path in polar regions and decrease liquid water path in low latitudes. The proposed parameterization can be used for all spatial resolutions of the CESM without special tuning of the shape parameter.