A surface layer similarity in the baroclinic atmospheric boundary layer

Similarity and scaling arguments for turbulent flows in the atmospheric surface layer (ASL) have traditionally relied on the simplifying assumptions of negligible Coriolis effects and a vertically uniform atmospheric pressure gradient (barotropic mesoscale environment). The ASL with these assumptions is then exactly analogous to the inertial overlap region of canonical wall-bounded flows. Townsend’s “attached” eddy hypothesis, with the distance from the wall () and friction velocity () as the characteristic scales, can then be invoked and, in the presence of buoyancy, generalized to the Monin-Obukhov Similarity Theory (MOST) that accounts for the effects of stratification. This canonical picture has however frustrated field measurements and numerical experiments in the ASL with contradicting results regarding the scaling laws predicted by the aforementioned theories. In this work, we revisit the problem and refocus the attention on the overlooked importance of wind turning (directional shear) associated with a ‘finite’ Rossby number, and on the role of baroclinicity (height-dependent pressure gradients resulting from large-scale horizontal temperature gradients) as likely causes of such disparities. We then combine a theoretical approach and a suite of high-resolution large eddy simulations (LES) to provide a revised formalism to model ASL flows that incorporates the effects of wind turning and baroclinicity in modeling the mean wind and Reynolds stress profiles. Our new model matches the LES significantly better than existing theories, and uncovers the implications of the departure from the logarithmic wind and constant stress profiles on the characteristics and scaling of turbulence in the ASL.