Vortical hot towers, their aggregate effects and their resolution dependence in the formation of Hurricane Diana (1984)

In this study we provide new insights into the vortical hot tower route to tropical cyclogenesis as simulated in a full-physics numerical weather prediction model (MM5) at near-cloud resolving scales. We investigate the genesis of Hurricane Diana with two main goals: 1) determine if the primary conclusions of Hendricks et al. (2004) and Montgomery et al. (2006) change as horizontal grid-spacing decreases from 3 km to 1 km and 2) examine more closely the influence of vortical hot towers on the tropical cyclogenesis process from the perspective of a very-high resolution mesoscale simulation. We employ traditional weather analysis techniques and new innovative means of displaying large and complex datasets to investigate the interaction between the cloud-scale features and the larger system-scale environment. The results are compared to prior studies to assess if simulated vortical hot tower dynamics exhibit a significant dependence on model resolution.

We find the basic physics of the vortical hot tower pathway is largely unchanged as grid-spacing decreases from 3 km to 1 km for simulations of Hurricane Diana. The differences between our high resolution simulation and coarser resolution simulations are mainly associated with fine-scale variability. Our 1 km simulation represents nearly an order of magnitude more convective towers with smaller spatial scales than what was observed in previous simulations. We find maximum updraft velocities in our 1 km simulation typically between 15 m s^-1 and 20 m s^-1 with instantaneous maximum values as high as 35 m s^-1, though these values typically decrease during the simulation. We also find that, while the cores in the vortical hot towers are significantly moistened by the vertical transport of moisture in the updraft, the larger-scale environment actually dries significantly due to horizontal advection. We examine a series of vortex merger events and find that merger activity is a ubiquitous and important aspect of the genesis of Hurricane Diana. Lastly, we perform a set of sensitivity experiments to determine if the vortical hot tower route to tropical cyclogenesis is a pre-WISHE process. We artificially cap the wind-dependent surface fluxes of latent and sensible heat and find that the genesis of Hurricane Diana proceeds despite this limitation. Our results broadly confirm previous work using coarser numerical resolution and provide new insights into the hypothesized upscale growth process in incipient hurricanes.