A numerical and observational study of the genesis of concentric eyewall hurricanes

Abstract

This work attempts to understand the dynamics of the genesis of concentric eyewall hurricanes. More specifically, we focus on the effects of external eddy forcing associated with upper-level wave asymmetries in the environment of tropical cyclones, and through what processes these effects can be achieved. Our approach is a combination of numerical modeling and observational case studies. We have made use of two numerical models, namely a simple two-layer model and a two-dimensional cloud resolving non-hydrostatic model. The latter is called the full physics model for short. Owing to the lack of direct measurements of upper-level atmospheric conditions, we choose to use reanalysis data from National Centers for Environmental Prediction (NCEP) and National Center for Atmospheric Research (NCAR) and European Center for Medium-Range Weather Forecasts (ECMWF). A somewhat lengthy evaluation suggests that both datasets are marginally suitable for case studies of tropical cyclones. Our major conclusions are as follows: ** Numerical simulations from the simple and full physics models suggest that the genesis of concentric eyewall hurricane results from finite-amplitude wind induced surface heat exchange (WISHE) instability of the tropical atmosphere. ** The results from the full physics model suggest that the role of the eddy forcing is like the catalyst in a chemical reaction. The forcing helps manifest internal finite-amplitude instabilities which themselves are driven by surface enthalpy fluxes. ** The results from our extensive case studies suggest that a causal relationship does not always exist between environmental forcing and genesis of a secondary eyewall. Some cases, for example, Hurricane Allen of 1980, Hurricane Elena of 1985 and Hurricane Opal of 1995, show a good and clear relationship between their eyewall replacement cycles and their external forcings. Some cases, for example, Hurricane Gilbert of 1998 and Hurricane Andrew of 1992, show some degree of causal relationship. Some cases, for example, Hurricane Emily of 1993 and Hurricane Gabrielle of 1989, show a weak or close to no causal relationship. ** With the results of Hurricane Frederic of 1979 and results from concentric eyewall hurricanes, we can conclude that the interaction between a tropical cyclone and its upper-level synoptic environment is neither sufficient nor necessary for the genesis and development of concentric eyewall cycles in reality. ** The maps of isentropic potential vorticity (PV) only provide qualitative information on the occurrence of the interaction. The strength of the interaction should be determined quantitatively by the eddy PV fluxes which should be calculated in a storm-moving coordinate system. The discrepancy between the numerical results and the case studies' results leads us to hypothesis two mechanisms of the genesis. One is the interaction between a hurricane and the ocean underneath. The other is the tilting of high PV inner core with the storm and followed up projection of cyclonic vorticity down to the ocean surface.