抄録
A methodology to describe the variability of the Kuroshio based on satellite thermal observation is presented. A sharp front is found at the coastal-side edge of the warmest area along the Kuroshio in each thermal image (hereafter, the "Kuroshio surface front"). The front usually is not a continuous line but consists of two or more parallel frontal lines. Because this structure is derived by a frontal eddy, we may use the farthest one from the coast and its downstream extension as an indicator of the flow of the Kuroshio. The relationship between the Kuroshio axis and the surface Kuroshio front is as follows: The surface front always matches with the axis in the area of the Kuroshio Extension east of Japan whereas it tends to match with the northern edge of the strong flow of the Kuroshio in the East China Sea and the area south of Japan. The surface Kuroshio front therefore is a good measure of the location of the Kuroshio flow if we note only the difference in the relationship in each region. By applying this methodology to the thermal imagery observed from February 1990 to March 1991, we successfully recognized the transition of the pattern of the paths of the Kuroshio in the summer of 1990. We also found that the path of in the regions south of Shikoku Island and off Bo^-so^- Peninsula is more stable than those in the other regions.