A numerical study of tropical cyclone motion using a barotropic model. II: Motion in spatially‐varying large‐scale flows

Abstract

The motion of an initially symmetric vortex in a spatially‐varying large‐scale flow on a beta plane is investigated using a nondivergent, barotropic numerical model. The calculations extend those carried out for the case of zero basic flow in Part I. The large‐scale flows are provided by meridionally‐varying zonal flows, or single‐mode, stationary, finite‐amplitude planetary waves in a channel. Interest is focused on the evolution of vortex asymmetries and their role in determining vortex motion relative to the basic large‐scale flow. As in Part I, the calculations are used to assess averaging procedures for computing the environmental wind field of a tropical cyclone from observed wind data. It is shown that averaging over an annular region centred on the vortex is, in principle, more accurate when there is a basic flow, in comparison with the case of zero basic flow.

Observational relationship of climatologic beta drift with large‐scale environmental flows

“…The correlation coefficients of the fitted and observed beta drift are 0.70 and 0.77 for C x and C y , respectively. By setting the environmental factors to zero, it can be seen that in a resting environment, the beta drift is westward and northward with a magnitude of 1.94 m s À1 and an orientation of 336°, which is consistent with most numerical results [Carr and Elsberry, 1990;Ulrich and Smith, 1991;Smith, 1991;Williams and Chan, 1994;Wang and Li, 1995;Li and Wang, 1996;Wang et al, 1997]. [11] The influences of the environmental factors in equations (1) and (2) can be compared with previous numerical results.…”

“…[8] The impact of large-scale environmental flows on beta drift has been extensively investigated numerically in the 1980s and 1990s without observational verification. Previous studies suggested that environmental factors such as latitude change, vorticity gradient and horizontal and vertical wind shears can interact with the primary TC circulation and then significantly affect the magnitude and orientation of beta drift [Ulrich and Smith, 1991;Smith, 1991;Williams and Chan, 1994;Wang and Li, 1995;Li and Wang, 1996;Wang et al, 1997]. Based on the climatologic beta drift obtained in the previous section, the influences of the environmental factors are examined by calculating vertical shears as the difference of wind speeds (u and v) between 200 hPa and 850 hPa, and horizontal shears and the gradient of relative vorticity (z) from the mean winds averaged over the layer between 850 hPa and 300 hPa.…”

“…[11] The influences of the environmental factors in equations (1) and (2) can be compared with previous numerical results. Ulrich and Smith [1991] and Smith [1991] found that the poleward component of beta drift (C y ) in an anticyclonic shear (@u/@y > 0) is significantly greater than that in a cyclonic shear (@u/@y < 0). Their result is consistent with (2).…”

“…Wu and Wang [2004] used the climatologic mean beta drift while Emanuel et al [2008] took the beta drift as a constant vector. Since previous studies indicated that large-scale environmental flows can affect beta drift [Ulrich and Smith, 1991;Smith, 1991;Williams and Chan, 1994;Wang and Li, 1995;Li and Wang, 1996;Wang et al, 1997], neglect of their influences may lead to considerable uncertainty in assessing possible changes in TC tracks under the background of global warming. The main objective of this study is to observationally verify the influence of large-scale environmental flows on beta drift by comparing with previous numerical results.…”

“…[3] Beta drift arises from the interaction between the gradient of earth's vorticity (b) and the primary TC circulation, which is generally westward and poleward [Adem and Lezama, 1960;Kasahara and Platzman, 1963;Holland, 1983;DeMaria, 1985;Chan and Williams, 1987;Fiorino and Elsberry, 1989]. Numerical studies suggested that the horizontal and vertical shears and gradient of relative vorticity of environmental flows can significantly affect beta drift [Ulrich and Smith, 1991;Smith, 1991;Williams and Chan, 1994;Wang and Li, 1995;Li and Wang, 1996;Wang et al, 1997], but observational studies were hindered due to insufficient observation and the extreme difficulty in separating environmental flows from the TC circulation.…”