The atmospheric boundary layer is the lowest part of the atmosphere, which is in direct contact with the surface. It is here, in this turbulent layer, that the exchange of heat, moisture and momentum between the surface and the atmosphere takes place. This thesis examines how well the boundary layer is described in global climate models with particular focus on the representation of the diurnal cycle. Two versions of the Community Atmosphere Model (CAM) that employ different turbulence parameterizations are evaluated in the same model framework. It is found that both overestimate the amplitude of the diurnal cycles of near-surface variables compared to observations from flux tower sites. The 10-m wind is much lower in CAM5 than in CAM4 due to the Turbulent Mountain Stress (TMS) parameterization of subgrid orography in CAM5. Additionally, the diurnal temperature range (DTR) is studied in a large set of models participating in the fifth phase of the Coupled Model Intercomparison Project (CMIP5). The model discrepancies are large both in simulations of the present day and in projections of the future. A correlative approach is used to assess which parameters are important for the model differences in DTR. No single parameter is found to be responsible, but clouds play an important role in all seasons and so do the evaporative fraction in summer. The diurnal cycles of these CMIP5 models are also evaluated against flux tower observations. The diurnal cycle of temperature is well captured, while most variables show a large inter-model spread. A subset of the models are analyzed deeper regarding their vertical boundary layer structure for a flux site in Oklahoma. A substantial warm summer bias is revealed in the models. Finally, the impact of TMS and changes in the vertical diffusion in CAM5 is studied. It is found that, although the inclusion of TMS leads to an improved large-scale circulation, the wind turning becomes too strong, which adds to the overestimation of the ageostrophic flow in the boundary layer. Instead, increasing the diffusivity in stable conditions tends to both degrade the large-scale circulation and cause an underestimated wind turning in the boundary layer.