Impacts of a Revised Surface Roughness Parameterization in the Community Land Model 5.1

Abstract. The roughness of the land surface (z0) is a key property for the amount of turbulent activity above the land surface and through that for the turbulent exchange of energy, water, momentum, and chemical species between the land and the atmosphere. Variations in z0 are substantial across different types of land cover from typically less than 1 mm over fresh snow or sand deserts up to more than 1 m over urban areas or forests. In this study, we revise the parameterizations and parameter choices related to z0 in the Community Land Model 5.1 (CLM), the land component of the Community Earth System Model 2.1.2 (CESM). We propose a number modifications for z0 in CLM, which are guided by observational data. Most importantly, we increase the z0 for all types of forests, while we decrease the momentum z0 for bare soil, snow, glaciers, and crops. We then assess the effect of those modifications in land-only (CLM) and land-atmosphere coupled (CESM) simulations. Diurnal variations of the land surface temperature (LST) are dampened in regions with forests, while they are amplified over warm deserts. These changes mitigate model biases compared to MODIS remote sensing observations, which have been identified in several earlier studies. The alterations in LST are mostly stronger during the day than at night. For example, the LST at 13:30 increases by more than 4.80 K during boreal summer across the entire Sahara. The induced changes in the diurnal variability of air temperatures at the bottom of the atmosphere generally oppose changes in LST in sign and are of smaller magnitude. Further, winds close to the land surface accelerate in areas where the momentum z0 was lowered, such as the Sahara desert, the Middle East, or the Antarctica, and decelerate in regions with forests. Overall, this study highlights that the current representation of z0 in CLM is not in agreement with observational constraints for several types of land cover. The resultant model modifications are shown to considerably alter the simulated climate in terms of temperatures and wind speed at the land surface.