| The motivation of this thesis is to obtain greater understanding of the controls on hurricane intensity. The results from three different hurricane models are presented: (1) a Geophysical Fluid Dynamics Laboratory Hurricane Prediction System simulation of Hurricane Opal (1995) using a reality-based, three-dimensional representation of the atmosphere; (2) the Emanuel (1995a) highly-simplified, axisymmetric hurricane model; and (3) the Rotunno and Emanuel (1987) axisymmetric, cloud-resolving, nonhydrostatic, grid model.; From the Opal simulation, we conclude that the popularly identified features of the environment of the storm (ocean eddies, trough-interaction, jet entrance regions) are not able to explain the modulations of intensification found in the simulation; although vertical shear is a strong candidate for the eventual weakening of the storm. This points to a thermodynamic interpretation of intensification. The Emanuel (1995a) model was designed to test a hypothesis for largely thermodynamic development of a hurricane. While confirming steady state intensity predictions of the Emanuel (1995b) maximum potential intensity theory (E-MPI) to within 5 meters per second, sensitivities of the model to purely numerical parameters argue against use the of this model as an operational forecast model.; With the Rotunno and Emanuel (1987) model, E-MPI can be greatly exceeded. Since this model should be faithful to many of the approximations of E-MPI theory, the remaining assumptions of E-MPI theory can be considered in turn. The model produces a significant flux of heat from the eye to the eyewall, allowing the modeled eyewall to ascend against a slight static stability, and this behavior is the key violation of E-MPI theory. The stabilization is a measure of the transfer of heat to the eyewall, and by accounting for this second source of heat, an ad hoc modification of E-MPI can explain most of the “superintensity” of the modeled storm.; We provide evidence that this active interaction between the eye and eyewall may operate in real hurricanes from published observations and in more-realistic, three-dimensional simulations of hurricanes. We propose that a new MPI should be developed to include this influence on hurricane intensity. |
