10C.3 Tropical-cyclone flow asymmetries induced by a uniform flow revisited

Wednesday, 18 April 2012: 2:00 PM
Champions FG (Sawgrass Marriott)
Gerald L. Thomsen, Ludwig Maximilian's University of Munich, Munich, Germany; and R. K. Smith and M. T. Montgomery

Idealized, convection-permitting, numerical model simulations are carried out to investigate the hypothesized effects of a uniform flow on the intensification, structural evolution, and maximum intensity of a tropical cyclone. The model used is a modified version of the Pennsylvania State University-National Center for Atmospheric Research fifth-generation Mesoscale Model. For the relatively intense vortices studied here, the differences in intensification rate and mature intensity resulting from changes in the background flow are not significant. As found in previous studies, a uniform flow in conjunction with surface friction leads to an asymmetry in the vertical motion exiting the boundary layer, but in the present experiments, this asymmetry is dominated by transient structures associated with rotating deep convection. Using sets of ensemble experiments in which the initial low-level moisture field is randomly perturbed below a value that can be measured by dropsonde observations, a coherent asymmetry in vertical velocity is evident only for the largest translation speed studied (7.5 m/s). The maximum vertical motion in the mature stage occurs about 45 degrees to the left of the motion vector, a result that differs from those of previous theoretical studies, which do not explicitly account for deep convection.

The time-mean maximum tangential wind speed occurs on the left side of the storm track as is found in recent observations of boundary-layer flow asymmetries in translating storms. However, the flow variability associated with transient convection raises questions concerning the ability to be able to adequately represent vortex-scale flow asymmetries (especially in the radial direction) from dropwindsonde observations spread over several hours.

The findings are broadly unchanged when using the relatively diffusive Gayno-Seaman boundary-layer parameterization scheme instead of the simpler bulk scheme used for the other calculations.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner