A Numerical Study of Topographical Effects on Flow Regimes in the Lower Atmosphere

Abstract

Orographically generated gravity waves, or mountain waves, have been the focus of much research for decades because of their importance on the general mean atmospheric circulation. These waves affect the flow on scales which are too small to be resolved by global weather and climate models, and so their impact on the larger scale flow must be parameterized. Linear theory has proven useful for obtaining a quantitative understanding of wave processes and their effects on the background flow, though one must assume that the low level flow in mountainous regions is approximately linear. Numerical simulations and field experiments indicate that this is often not the case, however, as nonlinear effects can dominate the flow near the orography. These nonlinear effects, which include processes such as flow splitting around a mountain or upstream blocking of the flow, affect gravity wave generation and decrease the accuracy of predictions based on linear theory. The purpose of this dissertation is to investigate the extent that linear theory-based mountain wave predictions can be improved by using an alternative initialization scheme. Linear orographic gravity wave models traditionally have been initialized at the lower boundary assuming the orography is equivalent to the surface vertical displacement field. While this method works when wave-induced perturbations are small compared to the mean flow, this has been shown to fail in weak flow regimes and tall mountains. We introduce an initialization technique where the linear model is initialized on a horizontal plane with results from a corresponding simulation from a nonlinear numerical model. The height level of initialization must be in a region in which the flow can be approximated by linear theory, and in practice this occurs above the low level nonlinear processes in the vicinity of the mountain. We show that this method leads to greater accuracy in the solutions of the wavefield above the orography. This new method is tested for flow regimes of uniform background wind and stability, and for simple bell shaped hills and more complex and realistic orography. Parameters derived from linear theory which are useful for global weather models are shown to be significantly affected by the new initialization scheme. These results have the potential to quantitatively improve global weather model mountain wave parameterization schemes in the relatively common instance of orographically-induced nonlinear flows , as well as to provide quick and accurate forecasts of wave activity for the aviation community.