New Community Atmosphere Model Underestimates Low-Level Cloud Water in the Arctic


Climate models have been used to predict future climate changes, including Arctic sea ice loss under future warming climate and Arctic processes that are highly sensitive to feedbacks between clouds and the surface. The representation of Arctic clouds in the newly released Community Atmospheric Model version 5 (CAM5) was examined and tested by a team of researchers, including Department of Energy (DOE) scientists from Pacific Northwest National Laboratory, Lawrence Livermore National Laboratory, and Brookhaven National Laboratory. The model was run in forecast mode using the DOE-supported Cloud-Associated Parameterizations Testbed (CAPT) framework to facilitate comparison with observations from the DOE Atmospheric Radiation Measurement (ARM) Indirect and Semi-Direct Aerosol Campaign (ISDAC) and Mixed-Phase Arctic Cloud Experiment (M-PACE). ISDAC and M-PACE were conducted at the North Slope of Alaska site in April 2008 and October 2004, respectively. The team found that CAM5 generally simulates cloud cover in the Arctic successfully; however, it underestimates the observed cloud liquid water content in low-level stratocumulus. The underestimate of low-level clouds causes CAM5 to significantly underestimate the surface downward longwave radiative fluxes by 20-40 W m-2, which would in turn compromise the model’s ability to accurately simulate Arctic climate. Model improvements on cloud microphysics such as the processes controlling conversion of liquid to frozen precipitation and on aerosol parameterizations are needed and highlighted in this research.


Liu, X., S. Xie, J. Boyle, S. A. Klein, X. Shi, Z. Wang, W. Lin, S. J. Ghan, M. Earle, P. S. K. Liu, and A. Zelenyuk. 2011. “Testing Cloud Microphysics Parameterizations in NCAR CAM5 with ISDAC and M-PACE Observations,” Journal of Geophysical Research 116, D00T11, DOI:10.1029/2011JD015889.