Latitudinal-Wavenumber Power Spectra of Stratospheric Temperature Fluctuations

John L. Stanford Department of Physics, Iowa State University, Ames 50011

Search for other papers by John L. Stanford in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Daily global grids of satellite radiance data over a 24-month period (1973–74) are decomposed into spherical harmonic functions. It is shown that a certain function of the expansion coefficients can be interpreted as the latitudinal wavenumber (k) dependence for global stratospheric temperature variance spectra. Beyond a forcing region, the spectra can be fit to the form km, with m=−4.1±0.6 for the upper and lower stratosphere, respectively. To within the data scatter, no seasonal dependence is found for m. The onset of the inertial subrange occurs at lowers k for the upper stratosphere than for the lower stratosphere. From a theoretical standpoint it is shown that a sufficient condition for isotropy/homogeneity of the fluctuation spectrum for a two-dimensional scalar on the surface of a sphere is that the variance matrix elements be independent of zonal wavenumber. This leads to a prediction of m=−3.6 for the parameters appropriate to the present investigation. The detailed results suggest that the discrepancy between observation and theory is due in part to significant anisotropy effects on large-scale fluctuations. These results are thought to be the first measurements of latitudinal power spectra for large-scale atmospheric motions and, in addition, to represent the first spectral measurements of any kind for the inertial region of large-scale motions in the stratosphere.

Abstract

Daily global grids of satellite radiance data over a 24-month period (1973–74) are decomposed into spherical harmonic functions. It is shown that a certain function of the expansion coefficients can be interpreted as the latitudinal wavenumber (k) dependence for global stratospheric temperature variance spectra. Beyond a forcing region, the spectra can be fit to the form km, with m=−4.1±0.6 for the upper and lower stratosphere, respectively. To within the data scatter, no seasonal dependence is found for m. The onset of the inertial subrange occurs at lowers k for the upper stratosphere than for the lower stratosphere. From a theoretical standpoint it is shown that a sufficient condition for isotropy/homogeneity of the fluctuation spectrum for a two-dimensional scalar on the surface of a sphere is that the variance matrix elements be independent of zonal wavenumber. This leads to a prediction of m=−3.6 for the parameters appropriate to the present investigation. The detailed results suggest that the discrepancy between observation and theory is due in part to significant anisotropy effects on large-scale fluctuations. These results are thought to be the first measurements of latitudinal power spectra for large-scale atmospheric motions and, in addition, to represent the first spectral measurements of any kind for the inertial region of large-scale motions in the stratosphere.

Save