Analysis of CCCma radiative transfer calculations for low level overcast clouds over ENA and SGP
Jordann Brendecke1, Xiquan Dong1, Baike Xi1, Xiang Zhong1, Jiangnan Li2, Howard W. Barker2, and Peter Pilewskie3
1Department of Hydrology and Atmospheric Sciences, University of Arizona, Tucson, AZ, 85721, USA
2Canadian Center for Climate Modeling and Analysis, Meteorological Service of Canada, Victoria, British Columbia, Canada
3Department of Atmospheric and Oceanic Sciences, University of Colorado, Boulder, CO 80309, USA
In this study, the Canadian Centre for Climate Modeling and Analysis (CCCma) radiative transfer model (RTM) is used to calculate surface and top of atmosphere (TOA) SW fluxes for low-level overcast liquid clouds. To verify the CCCma-calculated surface and TOA SW fluxes, observed surface and TOA SW fluxes at the DOE ARM Southern Great Plains (SGP, land) and Eastern North Atlantic (ENA, ocean) sites are used. MODIS-retrieved low-level cloud microphysical properties are input into the CCCma assuming three profiles: (a) cloud-droplet effect radius (re) and liquid water content (LWC) are constant with height; (b) LWC and re increase linearly from cloud base to cloud top, and (c) LWC and re increase linearly from cloud base to ¾ of the way up from cloud base and then decrease linearly towards cloud top. At the SGP site, surface SW flux Mean Bias Errors (MBEs) between CCCma calculations and surface observations are 5.3, 2.5 and -0.8 Wm-2, respectively, for cloud profiles (a) to (c), while they are -6.3, -8.9 and -4.5 Wm-2, for TOA SW flux differences. Comparisons at the ARM ENA site mimic their SGP counterparts with calculations using cloud profile (c) showing the best agreement at surface and TOA. All CCCma calculations at both SGP and ENA sites agree with observations much better than CERES Fu-Liou calculations. This profile results in minimal differences and RMSEs when compared with both surface and TOA observations, making it the most accurate for representing low-level cloud properties in radiative transfer modeling.