Evaluating The Performance Of Tropical Cyclone Intensity Guidance Over Western North Pacific

The overall statistics of the accuracy of eight statistical and dynamical tropical cyclone intensity guidance techniques used in the East China Regional Meteorological Center during the 2008 and 2009 western North Pacific typhoon seasons were evaluated herein. Skill scores which were calculated based on contingency tables are computed to verity the overall performance of the eight TC intensity guidance techniques, including the forecasts if those forecasts call for a typhoon to dissipate or if the real typhoon dissipates. The probabilities of the forecast intensity inaccuracy being within 6 m s-1 and greater than 15 m s-1 were examined in the study, also the forecast biases and forecast errors were calculated. The overintensity forecast errors at typhoon dissipation were verified as well.The results show that (1):12 h forecast skill of the six TC intensity guidance techniques was highest, while the lowest skill of all the six models did not occur in 72 h forecasting. The forecast errors of TCSP and BABJ are smallest, which probabilities of the forecast intensity inaccuracy are within 6 m s-1 and the highest errors produced by JAPN has the probabilities are greater than 15 m s-1.The forecasts from 36 to 72 h of TCSP and XSQD gave more false alarms reporting extant typhoons when they actually dissipated. For events of incorrect early dissipation forecasts, PGTW indicated relatively high probabilities and BABJ exhibited lowest probabilities. The PLSC technique manifested good performance in typhoon dissipation forecasts during the 2008 and 2009 typhoon seasons. (2):In the intensification stage, the majority of the techniques indicated > 60% probabilities of the errors of the forecast 12 h intensity change within±5 m s-1 at the forecast intervals, while none had capability to predict the rapid intensification and most of them had have a bias toward smaller 48 h intensity changes at the beginning of the stage. (3):The majority of the guidance techniques showed > 70% probabilities of the errors of the forecast 12 h intensity change within±5 m s-1 at the forecast times during the decay phase, and several techniques have certain capability of predicting the rapid decay cases. It is found that the evaluated statistical models had difficulty in predicting the distribution of intensity change in the 48 h intensity changes at the beginning of the intensification stage and the decay 36 h after peak intensity of the decay phase.