期刊論文

學年 108
學期 1
出版(發表)日期 2019-12-10
作品名稱 Combined Three-Stage 7-Day Weighted Analog Intensity Prediction Technique for Western North Pacific Tropical Cyclones: Demonstration of Optimum Performance
作品名稱(其他語言)
著者 Hsiao-Chung Tsai; Russell L. Elsberry
單位
出版者
著錄名稱、卷期、頁數 Weather and Forecasting 34(6), p.1979-1998
摘要 The original 7-day weighted analog intensity Pacific (WAIP) prediction technique is improved by developing a new version for the preformation stage and combining it with the bifurcation version that is especially for the intensification stage and the ending-storm stage WAIP. This combined three-stage WAIP includes a calibrated intensity spread designed to include 68% of the verifying intensities. In this demonstration of the optimum performance of the combined WAIP, the Joint Typhoon Warning Center (JTWC) best tracks, the time to formation (T2F), and ending-storm time are utilized as inputs along with the initial intensity [i.e., either 15 or 20 kt (1 kt ≈ 0.51 m s−1)]. In the new “Before Formation” (defined as 25 kt) stage of the WAIP, the intensity evolution is represented by a square function between the initial intensity and the T2F (25 kt), and very small mean absolute errors (MAEs) and intensity spreads are achieved. In the bifurcation version, correct selections between two cluster WAIP intensity evolutions lead to MAEs that slowly increase to 17 kt at 144 h, and the intensity spreads are relatively small as well. Because the ending-storm time constrains the analog selection in the WAIP, the MAEs begin to decrease after 72 h and are only 10 kt at 156 h. A case study is presented as to how an ensemble storm-track forecast along with the T2F could provide WAIP 7-day intensity predictions beginning in the preformation stage, which indicates the potential for earlier guidance for the JTWC intensity forecasts of western North Pacific tropical cyclones.
關鍵字 Tropical cyclones
語言 en_US
ISSN
期刊性質 國外
收錄於 SCI
產學合作
通訊作者
審稿制度
國別 USA
公開徵稿
出版型式 ,電子版,紙本
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/118018 )