A numerical investigation of surface-induced mesocyclogenesis near the Gulf Stream

A series of numerical experiments designed to simulate the initial development stages of low-level coastal mesocyclogenesis near the Gulf Stream was recently conducted. Under initially quiescent conditions, surface cyclogenesis in the control simulation occurs along a Gulf Stream meander in a region where the gradients in sea surface temperature (SST) are maximized. A low-level mesovortex on the order of 140 km forms approximately 12 h into the simulation and continues to intensify through 42 h. During the 24—48 h time period, a mesoscale frontal feature develops in direct response to strong diabatic forcing associated with sustained surface latent and sensible heating near the Gulf Stream frontal zone south of the main circulation center. Due to the non-linear advection of the frontal feature during this time period, the previously quasi-stationary circulation center drifts eastward (and away) from the thermal forcing associated with the large SST gradients found to the west. This eastward frontal propagation acts to decrease the magnitude of the low level horizontal air temperature gradient near the center of circulation throughout the 24—42 h development period. During the 42—48-h period, the relatively quick moving frontal feature acts to severely shear the nearly stationary center of circulation in the east—west direction. As a result, the mesoscale system begins to fill during the final 6 h of integration. In addition to the control simulation, additional sensitivity experiments were conducted. These experiments were specifically designed to: (1) investigate how the magnitude of the Gulf Stream SST gradients affect the timing and degree of cyclonic development; (2) address the impact surface moisture fluxes and moist convection each have on the simulated low level mesocyclogenesis; (3) isolate the role surface sensible heating plays in the overall development of the simulated mesocyclone. Results from the SST gradient experiment indicate that a moderate enhancement of the SST distribution significantly affects the timing of the initial cyclogenesis and the maximum intensity of the simulated frontal circulation. For the “no turbulent heat flux” experiment, it appears that the elimination of surface sensible heating does not radically alter the overall structure of the simulated mesocyclone. However, the rate of development during the early stage of cyclogenesis, the absolute peak intensity of the system as well as the vertical depth of the simulated mesoscale frontal feature were all noticeably reduced when compared with the control simulation. The initial development of a closed low level circulation was delayed by nearly 18 h in the absence surface latent heat fluxes. Once formed, the system intensified throughout the 48-h period of integration, but unlike the control experiment, a mesoscale frontal feature south of the main circulation center was not simulated. Results from the “no surface moisture flux/no moist convection” simulation illustrate that moist convective processes play a dominant role in the overall development of the mesoscale cyclone. For this particular case, a weak and extremely shallow circulation was simulated after 24 h. This circulation quickly eroded however, and was virtually non-existent for integration times greater than 39 h.