Extended Abstract (292K)8D.8
A Simulation of Hurricane Danny (1997) at Landfall using an artificial initial vortex
Sytske K. Kimball, University of South Alabama, Mobile, AL
Hurricane Danny (1997) was a small category 1 hurricane that originated near the coast of Louisiana on the 14th of July 1997. Before dawn on the 19th of July Danny reached the mouth of Mobile Bay in Southern Alabama and made landfall on the southeast side of the bay around midday on the 19th. The storm almost stalled for 24 hours in the Mobile Bay area while torrential rains devastated the region. Before entering the bay, Danny was a symmetric system. A strong asymmetry in the windfield and the convective activity evolved once the storm entered the bay. Intense convection occurred in the western eyewall with rainfall rates increasing dramatically at the same time. Precipitation nearly vanished in the eastern half of the storm. The change in rainfall distribution coincided with a change in the distribution of the low-level winds with maximum winds occurring at a lower altitude in the western eyewall than in the eastern eyewall. Danny reached its maximum intensity during its asymmetric phase in Mobile Bay.
Using the PSU/NCAR mesoscale model, MM5, a simulation of hurricane Danny is carried out to investigate Danny's unusual evolution of rainfall and wind distribution while the system was located in the Mobile Bay area. The use of 3 km horizontal resolution allows explicit modeling of convection, a process crucial to the accurate simulation of hurricane precipitation distribution and amount. MM5 is initialized with GFDL initial fields except for the SST fields which were obtained from the NAVY (OTIS). The original GFDL bogus vortex, which was much larger than the real Danny, is removed and replaced with a smaller bogus vortex of similar intensity to Danny. The simulation begins on 18 July at 12Z, when Danny is located over water to the south east of New Orleans, and ends on 20 July at 00 Z when the storm starts decaying over the Florida Panhandle.
The storm track and intensity are found to be highly sensitive to the initial structure of the initial bogus vortex. A broader and taller vortex (one where the 17 m s-1 isotach spans a larger diameter and reaches higher levels in the atmosphere) moves further to the west, past Mobile Bay, before recurving to the north. Shorter and/or more narrow vortices tend to recurve to the north earlier, making landfall to the east of Mobile Bay. A possible explanation is that steering is provided by upper level westerlies, hence, short or narrow vortices will not ‘feel' this steering flow. Motion of such vortices is dominated by beta gyre steering, giving the initial eastward track a more northward component. Ideally, the vertical and horizontal dimensions of the initial bogus vortex should equal those of the real Hurricane Danny. Unfortunately data giving a detailed enough three dimensional picture of Danny are not available.
Once a simulation with a track comparable to Danny's track is obtained, the influences of the warm waters of Mobile Bay and the surrounding topography on Danny's evolution and slow motion, will be explored.
Session 8D, Tropical Cyclone Simulation: Dynamics II (Parallel with Sessions 8A, 8B, and 8D)
Wednesday, 1 May 2002, 8:30 AM-10:30 AM
Previous paper