Analyzed in this paper are the 16-yr （1988-2003） tropical cyclone （TC） intensity data from three major forecast centers of the western North Pacific, i.e., China Meteorological Administration （CMA）, Regional Spe...Analyzed in this paper are the 16-yr （1988-2003） tropical cyclone （TC） intensity data from three major forecast centers of the western North Pacific, i.e., China Meteorological Administration （CMA）, Regional Specialized Meteorological Center Tokyo （RSMC Tokyo）, and Joint Typhoon Warning Center （JTWC） of the United States. Results show that there are significant discrepancies （at 1% significance level） in the intensity of TCs among the three centers, with a maximum difference for the same TC over 30 m s^-1. The flight reconnaissance over TC can minish the discrepancy to some extent. A climatic and persistent prediction model is set up to study the impact of initial data from different forecast centers on the prediction of TC intensity. It is obtained that the root mean square error （RMSE） of a 4-yr independent test is the largest using data from JTWC, while the smallest using data from RSMC Tokyo. Average absolute deviation in 24-h intensity prediction is 2.5 m s^-1 between CMA and RSMC Tokyo data, and 4.0 m s^-1 between CMA and JTWC data, with a maximum deviation reaching 21 m s^-1. Such a problem in the initial value increases the difficulty in intensity prediction of TCs over the western North Pacific.展开更多
基金Supported jointly by funds from the Shanghai Typhoon Institute, the National Natural Science Foundation of China (40333025)the Ministry of Science and Technology of China (2005DIB3J104), and the Forecasting system Laboratory of NMC CMA.
文摘Analyzed in this paper are the 16-yr （1988-2003） tropical cyclone （TC） intensity data from three major forecast centers of the western North Pacific, i.e., China Meteorological Administration （CMA）, Regional Specialized Meteorological Center Tokyo （RSMC Tokyo）, and Joint Typhoon Warning Center （JTWC） of the United States. Results show that there are significant discrepancies （at 1% significance level） in the intensity of TCs among the three centers, with a maximum difference for the same TC over 30 m s^-1. The flight reconnaissance over TC can minish the discrepancy to some extent. A climatic and persistent prediction model is set up to study the impact of initial data from different forecast centers on the prediction of TC intensity. It is obtained that the root mean square error （RMSE） of a 4-yr independent test is the largest using data from JTWC, while the smallest using data from RSMC Tokyo. Average absolute deviation in 24-h intensity prediction is 2.5 m s^-1 between CMA and RSMC Tokyo data, and 4.0 m s^-1 between CMA and JTWC data, with a maximum deviation reaching 21 m s^-1. Such a problem in the initial value increases the difficulty in intensity prediction of TCs over the western North Pacific.
