A parameter sweep experiment on topographic effects on the annular variability

Abstract

Effects of surface topography on annular variability of the extratropical troposphere are examined by parameter sweep experiments with a simplified global circulation model. Amplitude of the sinusoidal surface topography hm of zonal wavenumber m is swept as an experimental parameter in two series of experiments; a single zonal wavenumber, two (m = 2) or one (m = 1), is assumed in the series of WN2 or WN1, respectively. In additional series of WN2-1, the ratio of superposition of these two components is swept as an experimental parameter. In each run, long time integrations for 4300 days are done under a perpetual winter condition. Characteristics of the leading mode of empirical orthogonal function (EOF) of the zonal-mean zonal wind, and the surface pressure (Ps), depend on the amplitude and zonal wavenumber of the surface topography. In the WN2 experiment, characteristics of EOF1 of zonal-mean zonal wind and Ps change dramatically around h2 - 450 m, and the annular variability is divided into two types for large and small h2. In the WN1 experiment, on the other hand, characteristics of the annular variability do not show such drastic changes as those in the WN2 experiment, and EOF1 of Ps shows annular pattern for all h1 from 0 m to 1000 m. Three typical cases are analyzed in detail; h2 = 0 m (FLAT), h2 = 1000 m (HWN2), and h1 - 1000 m (HWNI). In the HWN2 experiment, the number of the storm tracks is two, and correlation map with the index that represents the oscillatory variability at that region shows a pattern localized in longitudes around the region with high teleconnectivity, even though EOF1 of Ps shows annular pattern. On the other hand, the annular variability has a sound physical basis in the FLAT and HWN1 experiments. Importance of the number and spatial structure of storm tracks which exist at the exits of jet streams is also confirmed in the WN2-1 experiment. © 2004, Meteorological Society Japan.