Earth Planets Space, Vol. 51 (Nos. 7, 8), pp. 475-484, 1999
Christopher D. Warner and Michael E. McIntyre
Centre for Atmospheric Science at the Department of Applied Mathematics and Theoretical Physics, University of Cambridge, U.K.
(Received August 13, 1998; Revised January 5, 1999; Accepted January 5, 1999)
Abstract: This paper reports first steps toward a computationally inexpensive spectral gravity wave para scheme whose predictions approximate those of a full three-dimensional (in spectral space) spectral model of atmospheric gw s. A reduction to two dimensions, as proposed by Hines, requiring the neglect of Coriolis and non-hydrostatic effects, is explored on the basis of comparisons with a full three-dimensional psm that includes Coriolis and non-hydrostatic effects. The reduction tries to be more realistic in terms of spectral shapes, though simpler in terms of wave-breaking criteria. It works remarkably well in the absence of, but less well in the presence of, background shear. The reasons for the discrepancies are under investigation, as are the implications for two-dimensional schemes, including Hines' as well as ours.