Hurricane Track Forecasting

Strategic Element: Advance short-term forecasts and warnings.

Principal Investigator: Hugh E. Willoughby

Objective: Basic physical understanding and forecasts of tropical cyclone motion. Observational studies with aircraft, dropsondes, and satellites. Numerical models, theory, and data assimilation.

Narrative: Historically, errors in track forecast have improved about 1% a year. By the late 1980s, a 115 mile (185 km) error in a 24-hour forecast reflected the state of the forecaster's art. During the 1995 and 1996 seasons, TPC/NHCs errors were < 90 miles (148 km). This reduction may have been due in part to good fortune, but many incremental improvements in procedure and the availability of superb numerical forecast models, such as the model developed at NOAA's Geophysical Fluid Dynamics Laboratory and HRD's own VICBAR barotropic model, are primarily responsible. One measure of the forecast system's present effectiveness is the decline in the human toll. Hurricane Andrew, although it destroyed $27B in property, killed only 15 people in South Florida through the direct force of wind and water. Acceleration of the downward trend in track forecast errors promises large economic and social benefits.

Since the early 1980s the Hurricane Research Division has conducted "Synoptic Flow Missions" in which the two WP-3D research aircraft deployed dropsondes to define the synoptic-scale flow around hurricanes. The added observations specified the "initial condition" more accurately for the computer models so that the errors in the predicted tracks were 16-30% smaller. During the 1997 hurricane season, NOAA's newly commissioned Gulfstream IV SP jet surveillance aircraft flew similar missions near several hurricanes. If the resulting warning areas can be reduced by 25%, the reduction in "overwarning" costs could be $50M per occurrence or $100M per year, assuming two threatened landfalls. Similar annual savings in damage are possible if more accurate and timely warnings can reduce the storms effect on property by as little as 1%. Finally, some parts of the coastline--the Florida Keys, New Orleans, and many barrier islands from South Texas through the Carolinas--require a longer time to evacuate than usual forecast lead times. Every improvement in track forecasts reduces the likelihood of a hurricane disaster in which thousands might die.

Duration: 1981 through the present.

Reference:

Burpee, R. W., J. L. Franklin, S. J. Lord, R. E. Tuleya, and S. D. Aberson, 1996: The impact of omega dropwindsondes on operational hurricane track forecast models. Bull. Amer. Meteor Soc, 77, 925-933.

Investigations

OBJECTIVE ANALYSIS OF THE HURRICANE AND ITS ENVIRONMENT

THE IMPACT OF DROPWINDSONDES ON OPERATIONAL HURRICANE TRACK FORECAST MODELS

PREDICTION OF MODEL PERFORMANCE

ENSEMBLE FORECASTING OF TROPICAL CYCLONES

ADAPTIVE OBSERVATIONS TO IMPROVE HURRICANE TRACK FORECASTS

EVALUATION OF GPS DROPWINDSONDES

ARE THE BETA GYRES NORMAL MODES?

SEMISPECTRAL MODELS OF HURRICANE MOTION AND INTENSIFICATION.