Abstract
This paper examines the characteristics of the asymmetric flow associated with tropical cyclone (TC) motion using the Final Analysis dataset produced after the Tropical Cyclone Motion Experiment (TCM-90). The wind data vertically-integrated between 850 and 300 hPa around a TC are first separated into an environment flow and a vortex circulation using the filtering algorithm of Kurihara et al. (1995). The latter is then Fourier-decomposed azimuthally to obtain the symmetric and asymmetric components. Nine TCs that occurred during the TCM-90 Experiment are examined.
For generally westward-moving TCs, the wavenumber-1 (WN-1) component is found to dominate the asymmetric flow. However, its pattern does not always exhibit a pair of counterrotating gyres as would be expected from previous modelling results (Fiorino and Elsberry, 1989). Further, the ventilation flow associated with WN-1 does not necessarily point towards the northwest. For a TC undergoing recurvature, the WN-2 flow becomes significant, and even has a larger magnitude than the WN-1 component, starting from about one day before recurvature. Consistent with the modelling results of Williams and Chan (1994), the WN-2 component also rotates counter-clockwise with time.
The growth and decay of the asymmetric components result from the interaction between the environmental flow and the symmetric flow of the TC through an energy exchange, in addition to such exchanges between the asymmetric components. Energy generally flows from the environment and the symmetric circulation of the TC to the WN-1 component during intensification but vice versa when the TC is weakening. The growth of the WN-2 component in recurving TCs is due to a transfer of energy from the environment, the symmetric circulation and the WN-1 flow. It is for this reason that the WN-1 flow becomes weaker than the WN-2 flow in such cases. The WN-1 component of fast-moving TCs is found to extract energy from the WN-2 component, in addition to those from the environment and the symmetric flow.
For generally westward-moving TCs, the wavenumber-1 (WN-1) component is found to dominate the asymmetric flow. However, its pattern does not always exhibit a pair of counterrotating gyres as would be expected from previous modelling results (Fiorino and Elsberry, 1989). Further, the ventilation flow associated with WN-1 does not necessarily point towards the northwest. For a TC undergoing recurvature, the WN-2 flow becomes significant, and even has a larger magnitude than the WN-1 component, starting from about one day before recurvature. Consistent with the modelling results of Williams and Chan (1994), the WN-2 component also rotates counter-clockwise with time.
The growth and decay of the asymmetric components result from the interaction between the environmental flow and the symmetric flow of the TC through an energy exchange, in addition to such exchanges between the asymmetric components. Energy generally flows from the environment and the symmetric circulation of the TC to the WN-1 component during intensification but vice versa when the TC is weakening. The growth of the WN-2 component in recurving TCs is due to a transfer of energy from the environment, the symmetric circulation and the WN-1 flow. It is for this reason that the WN-1 flow becomes weaker than the WN-2 flow in such cases. The WN-1 component of fast-moving TCs is found to extract energy from the WN-2 component, in addition to those from the environment and the symmetric flow.
| Original language | English |
|---|---|
| Pages (from-to) | 183-196 |
| Journal | Meteorology and Atmospheric Physics |
| Volume | 65 |
| Issue number | 3-4 |
| DOIs | |
| Publication status | Published - Sept 1998 |