03 April 2018

An Evaluation of Marine Boundary Layer Cloud Property Simulations in the Community Atmosphere Model Using Satellite Observations: Conventional Subgrid Parameterization Versus CLUBB

Science

Low clouds over the ocean surface—known as marine boundary layer clouds—play a critical role in regulating the Earth’s energy balance. However, accurately representing these clouds and related physical processes, such as turbulence and convection, in climate models is difficult. These processes occur at much smaller scales than a typical model grid size of about 100 kilometers, so “subgrid” cloud representations are necessary to reliably represent and connect small-scale cloud processes with larger grid-level variables in climate models. Researchers from the U.S. Department of Energy’s Pacific Northwest National Laboratory participated in a study showing a direct relationship between model biases—offsets from satellite observations—in the simulated frequency of very thin, low clouds and of precipitation from those clouds. 

Impact

Model deficiencies identified in this study related to low-cloud frequency and precipitation will guide improvements in representing low clouds and simulating their effects in Earth system models.

Summary

Clouds affect Earth’s climate in a variety of ways, including scattering and absorbing energy coming from sunlight and the surface and through precipitation. This study used global satellite data to evaluate two aspects of low clouds (within 1,000 meters of the surface) that are very important for climate: their influence on sunlight and the likelihood that they will produce drizzle or rain. Researchers evaluated and compared two different model representations of clouds within the Community Atmospheric Model Version 5 (CAM5). Comparing the representations with the satellite observations, researchers found that both representations simulated too many thin, wispy clouds that scatter little sunlight. They also noted that both cloud treatments simulated drizzle far too often. These findings suggest that the excessive drizzle in both cloud representations depletes the water in low clouds and that correcting the drizzle problem will result in more realistic simulations of sunlight scattering.

Contact
Hua Song
University of Maryland Baltimore County (UMBC)
Publications
Song, H, Z Zhang, P Ma, S Ghan, and M Wang.  2018.  "An Evaluation of Marine Boundary Layer Cloud Property Simulations in the Community Atmosphere Model Using Satellite Observations: Conventional Subgrid Parameterization versus CLUBB."  Journal of Climate 31(6): 2299-2320, doi:10.1175/jcli-d-17-0277.1.