This chapter is a review of recent developments in our understanding of the mesoscale spectrum of atmospheric motions. This topic has received considerable attention in the two decades since Doug Lilly's (1983) seminal paper on stratified turbulence. The subject has not been without controversy as atmospheric scientists and fluid dynamicists have debated the relative contributions of turbulent processes and internal waves to the spectrum of atmospheric motions. In this review we focus attention on the lower atmosphere, which is of primary interest to meteorologists.
Several papers that preceded Lilly's work are worth noting. Gage and Jasperson (1979) noted the variability in high-resolution sequential wind observations taken with a novel balloon sounding system. Gate (1979) placed these observations in a turbulence context and attributed much of the variability in these observations to two-dimensional turbulence arguing that the scales were too large to be associated with three-dimensional turbulence. Dewan (1979) examined stratospheric spectra and concluded that while the spectra had many of the features generally associated with turbulence they could also be explained by a spectrum of internal waves. Similar arguments were made by VanZandt (1982) who argued for a universal spectrum of internal waves analogous to the Garrett-Munk spectrum of internal waves in the ocean (Garrett and Munk, 1972).
The importance of an improved understanding of mesoscale variability has recently become evident as increasing importance is attached to the assimilation of diverse atmospheric data into numerical models. The current situation is summarized by Daley (1997) who argues that model forecasts depend critically on the assimilation of data with a specified error covariance. The error covariance has three components: measurement error and model error are quantifiable by observationalists and modelers, respectively, but the variability of the fields being measures is often unknown and must be estimated.
Sources of mesoscale variability have only recently come into focus. The advent of radar wind profiling has contributed substantially to our ability to observe rapidly changing wind fields and begin to resolve the space-time variability of mesoscale fields of motion. Profilers have also enabled resolution of short-period internal waves and inertia-gravity waves as well as continuous measurement of vertical motions (Gage, 1990; Gage and Gossard, 2003). Aircraft observations have also contributed substantially to our ability to resolve mesoscale atmospheric motion fields. Mesoscale spectra from aircraft motion sensors have provided a wealth of information on the spectrum of mesoscale motions (Nastrom el al., 1984; Nastrom and Gage, 1985; Cho et al., 1999a, b).
The aim of this review is to synthesize some of the most important developments in understanding the dynamics of mesoscale atmospheric variability since the publication of Lilly's work. The advances reported here have been made by many groups working in different disciplines.
The review begins with an examination of highly resolved samples of wind variability. Examples of horizontal and vertical velocities in Section 10.2 are shown to illustrate the variability intrinsic to these fields of motion even in the absence of extreme weather events. The observed spectrum of mesoscale wind variability is examined in Section 10.3, where spectra of horizontal and vertical velocities are considered separately. Dynamic processes that contribute to the observed mesoscale spectrum of motions are discussed in Section 10.4, followed by a consideration of some spectral models that have been developed for internal waves (Section 10.5) and stratified turbulence (Section 10.6). The consistency of the observed and modeled spectra is examined in Section 10.7. Contributions of topography and convection to meteorological variability on mesoscales are considered in Section 10.9. The review concludes with a summary (Section 10.9) of some recent developments and some thoughts about the direction of future research.