A high-resolution sea surface temperature (SST) data derived from several satellites is used to investigate the variability of the thermal front off northeastern Taiwan. Hidden by a dominant annual cycle, the SST data cannot reveal the thermal front fluctuation in the form of Hovmller diagram. An innovative methodology has been applied to the SST satellite imagery to derive the SST Standardized Index (SSTSI), capable of revealing the frontal variability with multiple time scales. Principal component analysis shows that the SSTSI variation consists mainly of two modes. Mode 1 represents a strong annual cycle related to the seasonal reversal of the monsoonal winds. The temperature gradient is enhanced in winter and a cold dome is observed off northern Taiwan in summer. Mode 2 is highly correlated with the upstream Kuroshio variability. The shoreward (seaward) migration of the thermal front takes place when the Kuroshio transport weakens (strengthens). The results are consistent with transports estimated by tidal gauge measurements, satellite altimeter-based sea level anomaly, and surface flow patterns derived from high-frequency radars. Mode 2 is coherent with the Kuroshio transport through the East Taiwan Channel at periods of 120 and 45 d with a time lag of 40 and 11 d, respectively. This 120 d fluctuation is due to the interaction between westward-propagating eddies and the Kuroshio east of Taiwan, while the 45 d signal arises from the Kuroshio's self-instability. The interannual variations of the SST pattern in winter and summer are also discussed.
ASJC Scopus subject areas