Date of Award

Spring 1-1-2015

Document Type


Degree Name

Doctor of Philosophy (PhD)


Atmospheric & Oceanic Sciences

First Advisor

Joan Alexander

Second Advisor

Katja Friedrich

Third Advisor

Cora Randall

Fourth Advisor

Jeffrey P. Thayer

Fifth Advisor

Robert Sharman


Small-scale gravity waves (GWs) with horizontal wavelengths of tens up to several hundred kilometers have demonstrated importance for driving the general circulation of the atmosphere, which affects many climate processes. GWs that propagate vertically from the troposphere into the middle atmosphere eventually dissipate and deposit momentum to the mean flow. Through this process they influence the timing of the transition in springtime from winter westerlies to summer easterlies in the stratosphere. They also play an important role in driving the mean-meridional transport circulation, the Brewer-Dobson circulation, and in the tropics help drive the Quasi-Biennial Oscillation and the Semi-Annual Oscillation. GWs with scales on the order of the size of a model grid box or smaller remain unresolved in Global Circulation Models (GCMs) and therefore need to be parameterized.

GWs are generated by a variety of sources including orography, convection, and geostrophic adjustment in regions of baroclinic instability. We focus here in particular on convectively-generated GWs, which are prevalent in the tropics and summer mid-latitudes. Their parameterizations in climate models range in complexity from simple assumptions of uniform sources to more complex methods that relate the spectrum of GWs to properties of convection in the climate model. The parameter settings that must be chosen to apply these GW parameterizations are poorly constrained by observations, so they are instead based largely on cloud-resolving model results.

Cloud-resolving model studies themselves use parameterized physics for the microphysics of precipitation particle formation. We first explore the sensitivity of the waves generated in cloud-resolving models to these physics parameterizations and show that knowledge of large-scale storm conditions is sufficient to predict the large-area and time-average spectrum of GW momentum flux above storms, irrespective of the convective details that coarse-resolution models cannot capture. However, aside from average spectral properties, information about local and instantaneous wave amplitudes is required for a realistic parameterization of GWs, as these determine the breaking levels of waves. We next develop a novel approach for simulating GWs that permits direct validation of modeled waves, including their amplitudes. A three-dimensional and time-varying heating field is derived from high-resolution observations of precipitation and used for forcing an idealized dry version of the Weather Research and Forecasting (WRF) model. Wave patterns and amplitudes observed in individual satellite overpasses are reproduced with remarkable quantitative agreement. The relative simplicity of the new model permits longer simulations with much larger and deeper domains needed to simulate wave horizontal/vertical propagation.

We simulate the entire month of June 2014 and an area covering most of the Continental U.S. at a high horizontal resolution of 4 km to determine characteristic properties of GWs from convection, in particular those quantities needed for their parameterization in large-scale models. In comparing to the stratospheric GW drag in the Community Atmosphere Model (CAM) and Modern-ERa Retrospective Analysis for Research and Applications (MERRA) we find that convectively-generated waves can give forces in the lower stratosphere that at times rival orographic wave forcing. Furthermore, their contribution to the summer branch of the stratospheric Brewer-Dobson circulation is in fact much larger than models predict. We analyze properties of radar precipitation and GWs from this model study and propose a change for future parameterization methods that would give more realistic drag forces in the stratosphere and may help alleviate model biases on synoptic scales.