SIMULATIONS OF TROPICAL CYCLONE CONNIE FROM THE AUSTRALIAN MONSOON EXPERIMENT

Simulations have been made of tropical cyclone Connie which developed during the Australian Monsoon Experiment (AMEX) Phase 2. The aims are to investigate factors which affected the cyclone's movement. Experime nts were made using a primitive equation model with parametrised physi cal processes, which has been developed at JMA (Japan Meteorological A gency). Reprocessed objective analyses from the European Centre for Me dium Range Weather Forecasts (ECMWF), and the JMA's typhoon implantati on scheme were used to generate the initial conditions. Analysis of th e vorticity equation using data from the 24-hour simulations shows tha t the horizontal advection of absolute vorticity and the divergence te rm were dominant effects on the motion. This suggests that the track w as determined by (a) steering flow, (b) the beta effect and (c) lower layer convergence which was generated and intensified by parametrised condensational heating. With regard to the beta effect, we illustrate that the motion in the three-dimensional atmosphere had similar charac teristics to the motion of a vortex in a nondivergent barotropic flow. That is, the track was sensitive to tropical cyclone size, but insens itive to its inner core structure. Further, the track differences simu lated from initial data containing circulations of different size beca me larger with simulation time. We conclude that to a first approximat ion, tropical cyclone Connie moved in a similar way to a vortex in a n ondivergent barotropic flow, but its movement was modified by condensa tional heating.