Hurricane Andrew's Landfall in South Florida. Part II: Surface Wind Fields and Potential Real-Time Applications

Mark D. Powell Hurricane Research Division, NOAA/AOML, Miami, Florida

Search for other papers by Mark D. Powell in
Current site
Google Scholar
PubMed
Close
and
Samuel H. Houston Hurricane Research Division, NOAA/AOML, Miami, Florida

Search for other papers by Samuel H. Houston in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

All available wind data associated with Hurricane Andrew's passage were analysed for periods corresponding to landfall south of Miami and emergence from southwest Florida. At landfall in southeast Florida, maximum sustained 1-min surface wind speeds VM1 reached just over 60 m s−1 in the northern eyewall over land; by the time Andrew exited the Florida peninsula, the peak value of VM1 over land decreased to 40–45 m s−1. Radar reflectivity observations from Tampa and Melbourne could not support an obvious correlation of convective cell development with coastal convergence during landfall on the southeast coast. On the southwest coast, however, convective cell development in the southern eyewall was supported by a coastal convergence maximum. Comparison of the wind swath with two independent Fujita-scale damage maps indicated that peak swath speeds compared well with damage-derived speed equivalents in the worst damaged areas but were higher than equivalents in moderately damaged areas. Comparison of the analysis maximum wind swath with an engineering survey of damaged homes suggests that homes exposed to a wide range of wind directions while subjected to high wind speeds suffered the most damage. Potential real-time applications of wind field products include warning dissemination, emergency management, storm surge and wave forecasting, and wind engineering. Development of damage assessment models for disaster mitigation is addressed from the viewpoint of an electrical utility.

Abstract

All available wind data associated with Hurricane Andrew's passage were analysed for periods corresponding to landfall south of Miami and emergence from southwest Florida. At landfall in southeast Florida, maximum sustained 1-min surface wind speeds VM1 reached just over 60 m s−1 in the northern eyewall over land; by the time Andrew exited the Florida peninsula, the peak value of VM1 over land decreased to 40–45 m s−1. Radar reflectivity observations from Tampa and Melbourne could not support an obvious correlation of convective cell development with coastal convergence during landfall on the southeast coast. On the southwest coast, however, convective cell development in the southern eyewall was supported by a coastal convergence maximum. Comparison of the wind swath with two independent Fujita-scale damage maps indicated that peak swath speeds compared well with damage-derived speed equivalents in the worst damaged areas but were higher than equivalents in moderately damaged areas. Comparison of the analysis maximum wind swath with an engineering survey of damaged homes suggests that homes exposed to a wide range of wind directions while subjected to high wind speeds suffered the most damage. Potential real-time applications of wind field products include warning dissemination, emergency management, storm surge and wave forecasting, and wind engineering. Development of damage assessment models for disaster mitigation is addressed from the viewpoint of an electrical utility.

Save