Numerical Exploration of the Stable Boundary Layer

Simulating the stable atmospheric boundary-layer (SABL) presents a significant challenge to numerical models due to the interactions of several processes with widely varying scales. For example, at larger scales, gravity waves can extract energy and momentum from the mean flow and efficiently transport them between regions in the atmosphere. Simulation of gravity waves requires a large spatial domain. Meanwhile, at small-scales, turbulent motions exhibit a complicated, time-dependent structure requiring sophisticated numerical treatment and a very small grid spacing. How these two processes interact to generate relatively strong turbulent mixing in environments ostensibly hostile to turbulent production remains a mystery. While focusing on idealized simulations to make the problem more tractable, a Reynolds stress turbulence closure scheme was incorporated into the National Taiwan University/Purdue University non-hydrostatic model, in order to more accurately simulate shear instability, one of the most important processes in the SABL. Shear instability is the most common mechanism for gravity wave generation, a mechanism for gravity wave absorption, and the dominant turbulence production process. The primary instability and cross-flow vorticity are accurately reproduced. In addition, secondary instabilities appear to be occurring, such as vortex pairing and knotting, which leads to the generation of stream-wise and vertical vorticity. In addition, gravity waves are generated via non-linear interactions between the Kelvin-Helmholtz modes directly produced by the shear. The resulting waves have an order of magnitude longer wavelength than the Kelvin-Helmholtz waves and a significantly greater phase speed in agreement with the literature.