Oceanic front, especially Kuroshio front, is an important phenomenon that is of great significance for scientific research, national economy and military uses. However, Kuroshio front to the east of Taiwan （KFETW in brief） was rare investigated. In this study, reanalysis method is used to study the KFETW＇s temporal and spatial variability and frontogenesis mechanism. It is found that although surface thermal front to the east of Taiwan is not obvious, there is an all-year strong Kuroshio thermal front called KFETW under the surface. The KFETW is connected to the south section of Kuroshio front in the East China Sea （KFECS in brief） and distributes along the east coastline of Taiwan. The KFETW has multi-scale variation feature. It has significant seasonal signal, and its intensity and width reach their maximum in summer. By using the reanalysis results obtained from this study, frontogenesis and changing mechanisms of the KFETW are discussed. It is found that both the Kuroshio and up-welling to the east of Taiwan can affect this front, and the up-welling may be the predominant factor in KFETW＇s frontogenesis and maintenance mechanism.
By using the reanalysis data, the impact of oceanic eddies and frontal wave on Kuroshio front to the east of Taiwan （KFETW） is studied. The result indicates that cold eddies （warm eddies） corresponding to the first baroclinic mode of Rossby wave can weaken （strengthen） the strength of the KFETW and narrow （widen） the width of this front. A frontal wave of the KFETW during January to February in 1991 is detected from the reanalysis data. And the trough （crest） of the frontal wave may weaken （strengthen） the strength of the KFETW and narrow （widen） the width and thickness of this front. It is found through the diagnostic analysis of the energy source of the frontal wave that the contribution of barotropic instability or that of baroclinic instability is more than that of Ketvin-Helmholtz （K-H） instability by 1 - 2 order of magnitude, and the contribution of the baroclinic instability is 5 times than that of the barotropic instability, thereby the frontal wave is basically driven by the baroclinic instability.
