Representation of marine lowlevel clouds in global-coupled kilometer-scale simulations

Marine boundary layer clouds stand out because of their importance for Earth's planetary albedo and their central role in determining Earth's sensitivity to forcing. The new global-coupled simulations at kilometer-scale resolution in both the atmosphere and the ocean in the framework of the H2020 nextGEMS project offer new opportunities to study cloud processes and their environmental factors, as well as provide unprecedented realism and new opportunities for comparison to observations. We examine the representation of (sub)tropical stratocumulus and trade-wind cumulus clouds by the IFS and ICON models configured with kilometer-scale resolution and global domains. The simultaneous consideration of ICON and IFS allows us to compare two strategies. The former simplifies parameterizations to understand process interactions better, sacrificing degrees of freedom to tune the model. The latter considers more sophisticated parameterizations, which allow for better tuning. The results of this study show the value of both. The performance of the four-year simulations is assessed in terms of the top-of-atmosphere (TOA) albedo and the vertical structure of the atmospheric boundary layer in eight regions where low-clouds are climatologically found. The stratocumulus regions are located in the eastern subtropical ocean basins, and the trade-wind cumulus regions are located west and equatorward from the stratocumulus ones. As an observational reference for the TOA albedo, we used satellite data from the CERES-EBAF TOA dataset.Both models captured the mean horizontal distribution and seasonal cycle of TOA albedo and the typical vertical structure of the low atmosphere over the stratocumulus regions. Despite its relatively simplistic approach to sub-grid parameterizations, particularly turbulence mixing treated with the Smagorinsky scheme, ICON performed comparably well to IFS, which employs more sophisticated solutions, including eddy-diffusivity mass flux and convection schemes. Regarding trade-wind cumulus, both models overestimate the mean TOA albedo. To validate the simulated vertical structure of the atmospheric boundary layer in the northwestern Atlantic trade-wind regime, we used the radiosondes launched at the Barbados Cloud Observatory (BCO) during the EUREC4A field campaign. The ICON and IFS models represent the main characteristics of the vertical structure of wind speed, temperature, and moisture observed at the BCO. We also find some discrepancies between the model representation and the observations. The simulations represented a colder (1 K) vertical profile than the observations. The ICON represented a drier cloud layer between 12 km and a moister layer above it, which is attributed to too much vertical mixing across the top of the cloud layer and suggests some revision of the stability correction function. The IFS model represented this region better than ICON, which was expected because IFS uses a shallow convection scheme, which allows better control of this region. However, IFS represented slightly drier the lowest 500 m.