Most extant studies of tropical cyclone movement consider a barotropic vortex on a β plane. However, observations have shown that real tropical cyclones are strongly baroclinic, with broad anticyclones aloft. Also, the distribution of the large-scale potential vorticity gradient in the tropical atmosphere is very nonuniform. These properties may substantially influence the movement of such storms.
Note that the anticyclone above a hurricane will interact with the lower hurricane vortex and induce storm motion. Such interaction can be caused by both the direct effect of ambient vertical shear and the effect of vertical variation of the background potential vorticity gradient. In this paper, an attempt to isolate the effect of background vertical shear is made. The hurricane is represented in a two-layer quasigeostrophic model as a point source of mass and zero potential vorticity air in the upper layer, collocated with a point cyclone in the lower layer. The model is integrated by the method of contour dynamics and contour surgery.
The results show that Northern Hemisphere tropical cyclones should have a component of drift relative to the mean flow in a direction to the left of the background vertical shear. The effect of weak shear is also found to be at least as strong as the β effect, and the effect is maximized by a certain optimal ambient shear. The behavior of the model is sensitive to the thickness ratio of the two layers and is less sensitive to the ratio of the vortices' horizontal scale to the radius of deformation. Storms with stronger negative potential vorticity anomalies tend to exhibit more vortex drift.