As discussed above, since the current speed of the Kuroshio is relatively weak to the west of the cape and increases from west to east in the vicinity of the cape, it is not feasible that the inertia force due to the strong Kuroshio current sends the Kuroshio water into the shallow shelf region against the law of the vorticity conservation. It would be reasonable to think that the counter-clockwise eddy in the Kii Channel is strengthened and the Kuroshio current zone approaches to Cape Shionomisaki at the time of the straight path, and that the resulted Kii Bifurcation Current draws the offshore Kuroshio water into the coastal region. Or, if the Kuroshio water intrudes into the coastal region to the west of Cape Shionomisaki due to some reason, the Bifurcation Current would be generated and the eddy in the Kii Channel might be strengthened.
The tide gauge station at Kushimoto is located to the west of the land passage which connects Cape Shionomisaki to the main part of the Kii Peninsula. At the time of the straight path, the region just off the Kushimoto tide gauge station is covered with the Kuroshio water, while that off the Uragami tide gauge station is covered with the relatively cold and heavy coastal water (existence of the current-shadow zone of the Kuroshio to the east of Cape Shionomisaki would enhance this tendency). Thus, the large sea level difference is created between two tide gauge stations. There is no reason that such difference occurs at the time of the large meander path, as the coastal water to the west of the cape would have the similar nature to that to the east of the cape.
DISCUSSION
The detailed oceanic structure in the vicinity of Cape Shionomisaki, the tip of the Kii Peninsula, was discussed mainly for the time of the straight path. Similar detailed observations in this area would be needed also at the time of the large meander path. However, the oceanographic condition in the coastal region at the time of the large meander path would be simpler, by judging from the variation nature of the sea level difference between Kushimoto and Uragami: the difference is very changeable at the time of the straight path in comparison with that at the time of the large meander path (see for example figure 7 of Kawabe, 1980). As discussed before, Fujita ( 1997) investigated the correlation between the position of the Kuroshio axis measured due southward from Cape Shionomisaki (in km) and the sea level difference between the Kushimoto and Uragami (in cm) (Fig.9). Although considerable scattering of the data points is seen, it is clear that the sea level difference at the time of the large meander path is smaller (generally smaller than 25 cm) regardless to the large variability in the distance of the Kuroshio axis from the coast (ranging from 40 to 320 km). On the other hand, at the time of the straight path, the distance is nearly constant and is from 30 to 50 km, but the sea level difference is very variable and ranges from about 25 to 55 cm. This appears to support our results that the strong current zone of the Kuroshio touches the coast and bottom slope just off Cape Shionomisaki at the time of the straight path, and that the resulted separation of the coastal water to the west of the cape from that to the east causes the relatively large sea level difference between the two tidal stations, which are separated by only 14 km.
