Large-scale climate condition for the interannual variation of typhoon activity near Taiwan with an application to seasonal climate prediction

Typhoon is the most threatening weather system that often causes enormous property damage and loss of life in Taiwan. Because of its catastrophic socio-economic consequences, it is important to understand large-scale climate factors that are instrumental in year-to-year typhoon variability. Through the correlation and stepwise regression analysis, several large-scale parameters in key locations are indentified in this study as representations of the climate features which condition the variations of the statistics of tropical cyclone (TC) activity near Taiwan. Three large-scale climate variables, namely, relative vorticity at 850 hPa (Vor850), vertical wind shear and sea level pressure over the western and central North Pacific from the antecedent May are selected as predictors. Among the three predictors, Vor850 is found to be the most important because it reflects variations of the ridge position of the westward extension of the western Pacific subtropical high. The forecast verification and skill evaluation of the seasonal prediction models of TC frequency and accumulated cyclone energy (ACE) are discussed. In addition to the pure statistical forecast method, a dynamical-statistical approach is presented on the basis of the simultaneous correlation between large-scale parameters and TC activities. A large-scale climate index (LSI) defined as the difference of the 850-hPa vorticity averaged over two critical areas is used as a conditioning factor that captures the variations of the statistics of TC activity during the typhoon peak season July-August-September. Finally, a dynamical-statistical seasonal climate procedure built on the statistical relationship between LSI and typhoons and the products of the operational seasonal climate forecast systems at the Central Weather Bureau of Taiwan is proposed.