Development of objective forecast guidance on tropical cyclone rapid intensity change

Meteorological Applications

Hiu‐fai Tam¹  Chun‐wing Choy²  Wai‐kin Wong² 

[1] Department of Physics Chinese University of Hong Kong Shatin Hong Kong;Hong Kong Observatory Kowloon Hong Kong;

关键词: Hong Kong Observatory;    intensity forecast;    operational forecasting;    rapid intensification;    tropical cyclone;   

DOI  :  10.1002/met.1981

来源: DOAJ

【 摘 要 】

Abstract Forecasting tropical cyclone (TC) intensity and rapid intensification (RI) are major challenges for numerical weather prediction models. In the present study, a model post‐processing technique, namely the TC intensity guidance on rapid intensification (TINT‐RI), has been developed for the western North Pacific basin (South China Sea excluded). It combines logistic regression and the naïve Bayes classifier to provide RI forecasts up to the next 48 hr. Predictors describing physical and environmental conditions are taken from the outputs of the European Centre for Medium‐Range Weather Forecasts model and the National Oceanic and Atmospheric Administration tropical cyclone heat potential field. Several observed TC characteristics such as persistence (change of past 12 hr intensity) are also included in the algorithm. To further reduce the false alarm ratio, the TINT‐RI makes reference to the intensity forecasts from a previously developed statistical‐dynamical TC intensity forecast model and the change of model vertical wind shear. Verification for TCs with RI reveals that the TINT‐RI is significantly more skilful than the direct model outputs from major global numerical weather prediction models and climatology. A case study on extreme RI of Super Typhoon Hato (1713) during its substantial intensification over the Luzon Strait before entering the northern part of the South China Sea is presented to discuss the performance of the TINT‐RI. Potential ways to improve the TINT‐RI such as using sea surface temperature and its anomaly as the oceanic predictor are discussed.

【 授权许可】

Unknown