Principal Investigator: Dr. Mark D. Powell
Collaborating scientist(s):
Samuel H. Houston
Ignacio Ares
Objective: To develop a real-time hurricane surface wind analysis system for forecast, damage assessment and recovery planning support.
Rationale: Monitoring tropical cyclone intensity is required to produce accurate forecasts and warnings. Timely evacuations and preparations before the storm help to save lives and property but losses are inevitable even with perfect forecasts. In 1992, despite accurate warnings, Hurricane Andrew left 200,000 Americans homeless a nd demolished property in excess of $25 billion; had landfall occurred only 30 km further north the loss would have approached $65 billion. Mitigating a portion of such losses may be possible by effective use of monitoring information during and after the event. As the hurricane approaches, real-time analyses assist forecasters in determining the location of wind maxima and the extent of hurricane and gale force winds. During landfall, real-time analyses help identify communities experiencing the most severe winds and storm su rge. After landfall, correlations of damage with meteorological predictors will allow construction of models to assist in damage assessment before visual inspections are possible.
Method: As the hurricane approaches, measurements are gathered from reconnaissance aircraft, ships, buoys, and land stations in real-time (Figure 1). These winds are then standardized to represent a common framework for exposure, height and averaging time based on documented methods; modeling aspects of the analysis are limited to boundary layer adjustments of flight level reconnaissance measurements to the surface. Adjustment methods are subject to a continuous verification process involving independent comparisons with ground truth. The observations undergo visual interactive quality control (Figure 2) before being sent to an objective analysis platform. Once analyzed the wind field is represented by a streamline and contour plot which could then be combined with radar imagery and high resolution geography (Figure 3). This information readily conveys locations expected to experience the highest wind speeds. The information would then be used as input to damage assessment models (Figure 4), and geographic information systems, providing emergency infrastructure managers valuable decision-making tools and guidance for planning response and recovery operations (Figure 5).
Accomplishment: A model for damage assessment to utility structures (poles, transformers, and conductors) was constructed for Florida Power and Light Company (major sponsors of the research) based on geographically referenced damage statistics from Hurricane Andrew of 19 92, and insurance industry based damage loss information from both Andrew and Hurricane Hugo of 1989. According to FPL, even a 10% improvement in their ability to estimate losses after a hurricane of the magnitude of Andrew, would save on the order of $1 million. A paper describing this work, "Real-time Damage Assessment in Hurricanes" was presented at the 21st Conference on Hurricanes and Tropical Meteorology. This hurricane season, the surface wind analysis system was evaluated in the operational environment of the National Hurricane Center. Over 80 real-time analyses were conducted including the landfalls of Hurricanes Allison, Erin, and Opal in Florida, Luis and Marilyn in the Caribbean, and the approach of Hurricane Felix to the U.S.east coast. We spent considerable time interacting with the hurricane forecasters, gaining valuable input for improvements to the analysis system.
Key reference:
Powell, M. D., S. H. Houston, and I. Ares, 1995: Real-time damage assessment in hurricanes. 21st Conference on Hurricanes and Tropical Meteorology, April 24-28, Miami, FL. American Meteorological Society.
Click here to return to the AOML project overview page.
Click here to return to the Advance Short-Term Warning & Forecast Services page.