Simulated Impacts of a Mesoscale Convective System on the Track of Typhoon Robyn during TCM-93

Abstract

The contribution that a mesoscale convective vortex that developed within the circulation of Typhoon Robyn (1993) may have had to a significant tropical cyclone track change is simulated with a mesoscale model using initial conditions that approximate the circulations measured by aircraft during the Tropical Cyclone Motion (TCM-93) field experiment. A dry version of the PSU?NCAR Mesoscale Model is used to first investigate the dynamic aspects of the interaction. A deceleration of about 2 m s?1 and then more northward movement, similar to that observed for Typhoon Robyn, could have been produced by an interaction with the mesoscale convective vortex of the type modeled in the control run. Sensitivity of the simulated track change is tested for various aspects of the mesoscale vortex and tropical cyclone. It is found that track deflections between 67 and 130 km in 18?24 h could be produced under a variety of realistic scenarios. As the mesoscale vortex is advected around the tropical cyclone by the cyclonic winds, it is also being filamented by the horizontal shear of the tropical cyclone outer winds. Whereas the rate at which the vortex is advected about the tropical cyclone is critical to the amount of curvature of the tropical cyclone track deflection, the timescale of the filamentation of the mesoscale vortex is critical to the longevity of the track deflection, and thus maintenance of the vortex is a crucial factor. For the same tropical cyclone and separation distance, the track deflections are greater for a larger, deeper, and more intense mesoscale vortex. For the same mesoscale vortex, a variety of track deflections is possible, depending on the outer wind structure of the tropical cyclone due to both the advective effect and to the change in the gradient of vorticity. For a large tropical cyclone, positive vorticity extends farther from the core region, and the curvature of the track deflection is greater than for a smaller tropical cyclone where the vorticity becomes anticyclonic at smaller radii. Although a large initial effect on the tropical cyclone path occurs for small separation distances, the mesoscale vortex is rapidly filamented so that effects on the tropical cyclone track are negligible by 12 h. For larger separation distances, the mesoscale vortex does not filament as rapidly, and a smaller but longer lasting track deflection is simulated. When the control simulation is extended to a ? plane, it is found that the primary contribution to the tropical cyclone track change is still due to the interaction with the mesoscale convective vortex. However, a secondary effect due to a nonlinear interaction between the ? gyres and the mesoscale convective vortex adds a small component of propagation to the tropical cyclone that becomes significant after about 15 h of simulation.