Key points are not available for this paper at this time.
A high-order model is proposed for the study of the 24 h evolution of clear planetary boundary layers. The model includes the rate equations of correlations up to the third order, as required for an accurate description of daytime convective phenomena, but it also takes into account interactions between radiative transfer and turbulence in order to achieve a physically reasonable description of the nocturnal structure of the boundary layer. This numerical model is tested against the Wangara boundary layer data of Day 33 and Night 33–34 (Clarke et al., 197l). The computed daytime mean structure of the boundary layer compares favorably with the Wangara data, while the daytime turbulent structure, expressed in the framework of the convective similarity theory, is in particularly good quantitative agreement with a number of experimental and numerical data concerning convection in the boundary layer, with particular concern to the production of turbulence at the top of the mixed layer. The computed nocturnal mean structure is shown to be driven principally by radiative transfer and the mesoscale pressure gradient. It agrees with the observed nocturnal structure with the exception that the height of the turbulent surface layer is underestimated in the model, but it is shown that this height is very sensitive to the imposed boundary conditions. The computed nocturnal turbulent structure is explained on a qualitative basis by the interactions between shear generation of turbulence, thermal stratification and radiative phenomena. It is also shown that generation and vertical propagation of nocturnal turbulence, which are of primary interest for environmental purposes, are strongly influenced by turbulence and radiative transfer interactions.
André et al. (Sun,) studied this question.