Extended Abstract (172K)JP1.4
Improvements in Real-Time Statistical Tropical Cyclone Intensity Forecasts Using Satellite Data
Mark DeMaria, NOAA/NESDIS, Ft. Collins, CO; and M. Mainelli, L. K. Shay, J. A. Knaff, and J. P. Kossin
Forecasting intensity changes of tropical cyclones remains a challenging task. Over the past five years, the Statistical Hurricane Intensity Prediction Scheme (SHIPS) for the Atlantic basin has shown the greatest average forecast skill out to 48 hours. For the 2002 Atlantic hurricane season, an experimental version of SHIPS that includes new predictors derived from satellite data will be evaluated in real-time. The operational version of SHIPS includes predictors from climatology, persistence, the storm environment (vertical shear, etc) and sea surface temperature. The experimental version of SHIPS will also include predictors from GOES infrared (Channel 4) imagery, and ocean heat content determined from TOPEX/Poseidon and ERS-2 satellite altimetry data. Preliminary results with dependent data suggest that these new data sources can reduce the average forecast errors by 4-8%. Because the level of skill of SHIPS as determined by comparison of errors with those from a simple climatology and persistence model is only about 10-15%, a reduction in the average errors of up to 8% implies a significant increase in model skill. The impact of these new data sources will be evaluated by comparison of the experimental and operational SHIPS intensity forecasts.
Joint Poster Session 1, Operational Applications and Artificial Intelligence (Joint between 12th Conference on Satellite Meteorology and Oceanography and Third Conference on Artificial Intelligence Applications to Environmental Science)
Monday, 10 February 2003, 2:30 PM-4:00 PM
Previous paper Next paper