Abstract
Surface longwave emissivity can be less than unity and vary significantly with frequency. However, most ESMs still assume a blackbody surface in the longwave (LW) radiation scheme of their atmosphere models. Moreover, ice cloud in the far-IR can have a single scattering albedo as large as 0.8, while most ESMs still assume non-scattering clouds in the LW. The two issues can manifest in the high latitudes where atmospheric column water vapor is much less than in the mid-latitudes and tropics. The study is motived by such issues and aimed at improving the representation of surface-atmosphere radiative coupling in the models.
This first part of the study describes our work of incorporating realistic surface spectral emissivity into the atmospheric component of the CESM 1.1.1 and evaluates its impact on simulated climate. By ensuring consistency of the broadband surface longwave flux across different components of the models, the top of atmosphere (TOA) energy balance in the modified model can be attained without re-tuning the model. The inclusion of surface spectral emissivity, however, leads to a decrease of net upward longwave flux at the surface and a comparable increase of latent heat flux. The global-mean surface temperature difference between the modified and standard CESM simulation is 0.20 K for the fully coupled run and 0.45 K for the slab-ocean run. Noticeable surface temperature differences between the modified and standard CESM simulations are seen over the Sahara desert and polar regions. Accordingly, the climatological mean sea ice fraction in the modified model simulation can be less than that in the standard simulation by as much as 0.1 in some regions. Therefore, the inclusion of surface emissivity partly addressed the cold biases in polar surface air temperature and the excessive freezing bias in sea ice coverage of both polar regions. The second part of the study describes our implementation of surface spectral emissivity and longwave cloud scattering into the E3SM, and the assessment of its impact on the simulated climate mean state in high latitudes, as well as comparisons with the simulation results from the first part.