On Structure and Temporal Variation of the Bifurcation Current off the Kii Peninsula

The Kii Bifurcation Current is often found along the southwest coast of the Kii Peninsula, and its frequency of occurrence reaches about 70% in the period from 1988 to 1996 (Takeuchi et al., 1998a). In order to clarify the structure and short-period variability of the Kii Bifurcation Current, detailed observations were made four times on board the R/V Seisui-maru of Mie University on October 29–31, 1996, on June 24–26, 1997, October 14–16, 1997, and December 3–4, 1997. The measured horizontal structure of the Kii Bifurcation Current indicates that the eastern portion of the Current (eastward flow near Cape Shionomisaki) consists of a part of the current zone of the Kuroshio. It is shown that the current structure, including the Kii Bifurcation Current in the vicinity of Cape Shionomisaki, is stable when the Kuroshio is flowing in a stationary straight path, but that the current structure is considerably changed when small-scale eddies pass by the cape. Such short-period variation can be monitored by using the daily variation of the sea level difference between Kushimoto and Uragami. In particular, in the case of October 29–31, 1996, when an eminent small-scale eddy passed by Cape Shionomisaki, and when the Kuroshio axis tentatively moved southwards about 50 km apart from the coast, the Kii Bifurcation Current seems to have disappeared.     

Why does the Sea Level Difference between Kushimoto and Uragami Show Periods of Large Meander and Non-Large Meander Paths of the Kuroshio South of Japan?

The sea level difference between Kushimoto and Uragami, located to the west and east of the southern tip of the Kii Peninsula, is relatively large in periods of non-large meander path (nLMP) of the Kuroshio south of Japan in comparison with periods of large meander path (LMP). Based on this clear relationship, the sea level difference between Kushimoto and Uragami has been used as an index showing the periods of nLMP and those of LMP of the Kuroshio south of Japan. It has been pointed out that warm and saline Kuroshio water, separated from the main path of the Kuroshio, has a tendency to approach the western area off Kii Peninsula to off Muroto Peninsula in periods of nLMP, while it approaches the eastern area off Kii Peninsula to Omae-zaki in periods of LMP. On the basis of this observational evidences, the dynamic background underlaying the well-known relationship between the Kuroshio path and the sea level difference between Kushimoto and Uragami is examined in the present study, using the temperature and salinity data observed by Wakayama Prefectural Fisheries Experimental Station and Fisheries Research Institute of Mie. It is shown that deviations in vertically integrated specific volume off Kushimoto and Uragami almost equal deviations in observed sea level at Kushimoto and Uragami, respectively. It is also shown that the difference in vertically integrated specific volume between off Kushimoto and off Uragami almost equals the difference in observed sea level between Kushimoto and Uragami. As for the Kuroshio water, the high-temperature contribution is predominant for its specific volume rather than that of high salinity, which yields thermal expansion in comparison with coastal water. Because the difference in vertically integrated specific volume between off Kushimoto and off Uragami almost equals the difference in observed sea level between Kushimoto and Uragami, it is concluded that the relationship between the Kuroshio path and sea level difference between Kushimoto and Uragami is caused by the different approaching of the warm Kuroshio water between in nLMP periods and in LMP periods.     

Sea level variations along the south coast of Japan and the large meander in the Kuroshio

Sea level variations from 1974 through 1976 at 9 stations on the south coast of Japan (from west to east, Aburatsu, Tosa-shimizu, Muroto-misaki, Kushimoto, Uragami, Owase, Toba, Maisaka and Omaezaki) were analysed in relation to the large meander in the Kuroshio. From May to July in 1975, a small maximum in sea level variation was observed at every station west of Cape Shionomisaki from Aburatsu to Kushimoto. It propagated eastward along with the eastward propagation of a small meander in the Kuroshio until it reached Kushimoto, when the sea levels at Uragami and Owase started to rise sharply. This remarkable rise appeared at all stations in August when a large meander in the Kuroshio was established. The mean sea level at the stations east of Cape Shionomisaki from Uragami to Omaezaki rose by about 10 cm. The difference in sea level variations between the regions east and west of Cape Shionomisaki, which had been present before the rise, disappeared. A similar characteristic of sea level variation was also found in the generation stage of the large meander in 1959. The sea level variations along the south coast of Japan indicate that, prior to the generation of the large meander, the small meander in the Kuroshio was generated southeast of Kyushu and propagated eastward and that, just when this meander reached off Cape Shionomisaki, a large scale oceanic event covering over the whole region of the south coast of Japan occurred. This large scale event seems to be one of the necessary conditions for the generation of the large meander in the Kuroshio off Enshû-nada.     

Interannual correlations between sea level difference at the south coast of Japan and wind stress over the north pacific

Relationships of the sea level differences between Naze and Nishinoomote and between Kushimoto and Uragami with wind stress over the North Pacific are examined for interannual variability. These sea level differences are considered to be indications of Kuroshio transport in Tokara Strait and Kuroshio path south of Enshu-nada, respectively. In the sea level difference between Kushimoto and Uragami, dominant variations are found to have periods of about seven years and 3–4 years. The variation of about 7-year period, which corresponds to that in the Kuroshio path between the large meander and non-large meander, is coherent with the variation of the wind stress curl in a region about 2,400 km east of the Kii Peninsula, where negative stress curl weakens about two years before the sea level difference drops (i.e. the large meander path in the Kuroshio generates). The variation of the 3–4 year period is coherent with that of the wind stress in a large area covering the eastern equatorial Pacific, which suggests that it links with global-scale atmospheric variations. Interannual variation in sea level difference between Naze and Nishinoomote is not coherent with that between Kushimoto and Uragami, which suggests that it is not related to the variation of the Kuroshio path south of Enshu-nada, but is coherent with that of the zonally-integrated Sverdrup transport in the latitudinal zone along 30°N. It is suggested that the interannual variation of the Kuroshio transport in Tokara Strait can be explained by the barotropic response to the wind stress.     

Intrusion of Kuroshio-Derived bottom cold water into Osaka Bay and its possible cause

The bottom layer in Osaka Bay was occupied in August from 1980 to 1995 by a water mass of relatively low temperature and rich in nutrients in comparison to previous and following decades. The relationship between Osaka Bay bottom temperature and Kuroshio axis location south of Kii Peninsula has been investigated using the dataset archived by Osaka Research Institute of Environment of Agriculture and Fisheries and axis-location data provided by Marine Information Research Center. The correlation between bottom temperature in the bay and Kuroshio-axis distance from Cape Shionomisaki indicates that the bottom temperature in the bay decreases when the Kuroshio axis is a long distance from the cape, and vice versa. To investigate why the temperature distribution depends on the axis location, composite temperature maps are depicted using summer temperature data from 1970 to 2005 archived in the Japan Oceanographic Data Center (JODC) after dividing all data into two groups with different axis locations. These temperature maps and temperature-salinity plots using the same JODC data suggest a scenario: cold water in the Kuroshio intermediate layer is first upwelled on the eastern side of the Kii Peninsula and thereafter moves westward as a coastal boundary current in the Kelvin wave sense of the Northern Hemisphere when the Kuroshio axis is located around 74-km distance from Cape Shionomisaki. This scenario is validated using internal Froude number maps depicted using the JODC-archived hydrographic data and geostrophic current fields. In addition, the reanalysis daily data provided by Japan Coastal Ocean Predictability Experiment are used for the validation.     

Bifurcation current along the southwest coast of the Kii Peninsula

Along the southwest coast of the Kii Peninsula, a bifurcation current is regularly observed. By using ADCP data taken on board the R/V Wakayama of the Wakayama Prefectural Fisheries Experimental Station, characteristics of this bifurcation current are analyzed. The occurrence frequency of the bifurcation current reaches about 70% in the period from 1988 to 1996. The bifurcation point appears to be changeable and occurs almost evenly between Cape Ichie and Cape Shionomisaki. The current divergence in the alongshore direction was also investigated. Positive divergence values dominated in the whole analyzed area, and an onshore current appears to be dominant along the southwest coast of the Kii Peninsula, except in 1990 when the Kuroshio flowed in a large meandering path.     

Influence of Mesoscale Eddies on Variations of the Kuroshio Path South of Japan

The influences of mesoscale eddies on variations of the Kuroshio path south of Japan have been investigated using time series of the Kuroshio axis location and altimeter-derived sea surface height maps for a period of seven years from 1993 to 1999, when the Kuroshio followed its non-large meander path. It was found that both the cyclonic and anticyclonic eddies may interact with the Kuroshio and trigger short-term meanders of the Kuroshio path, although not all eddies that approached or collided with the Kuroshio formed meanders. An anticyclonic eddy that revolves clockwise in a region south of Shikoku and Cape Shionomisaki with a period of about 5–6 months was found to propagate westward along about 30°N and collide with the Kuroshio in the east of Kyushu or south of Shikoku. This collision sometimes triggers meanders which propagate over the whole region south of Japan. The eddy was advected downstream, generating a meander on the downstream side to the east of Cape Shionomisaki. After the eddy passed Cape Shionomisaki, it detached from the Kuroshio and started to move westward again. Sometimes the eddy merges with other anticyclonic eddies traveling from the east. Coalescence of cyclonic eddies, which are also generated in the Kuroshio Extension region and propagate westward in the Kuroshio recirculation region south of Japan, into the Kuroshio in the east of Kyushu, also triggers meanders which mainly propagate only in a region west of Cape Shionomisaki. This revised version was published online in July 2006 with corrections to the Cover Date.     

Transition processes between the three typical paths of the Kuroshio

Transitions between the three typical paths of the Kuroshio south of Japan (the nearshore and offshore non-large-meander paths and the large-meander path) are described using sea level data at Miyake-jima and HachijÔ-jima in the Izu Islands and temperature data at a depth of 200 m observed from 1964 to 1975 and in 1980.In transitions between the nearshore and offshore non-large-meander paths the variation of the Kuroshio path occurs first in the region off Enshû-nada between the Kii Peninsula and the Izu Ridge and subsequently over the ridge. In the nearshore to offshore transition the offshore displacement of the path occurs first off Enshû-nada and then develops southeastwardly in the direction of HachijÔ-jima. In the reverse transition shoreward displacement occurs first off Enshû-nada and then throughout the region west and east of the Izu Ridge. The position of the Kuroshio south of Cape Shiono-misaki (the southernmost tip of the Kii Peninsula) is almost fixed near the coast throughout these transition periods, and significant variations of the Kuroshio path only occur east of the cape. The nearshore to offshore and offshore to nearshore transitions can be estimated to take about 25 and 35 days, respectively, during which the variation of the Kuroshio path over the Izu Ridge occurs for the last 11 and 25 days.The transitions between the non-large-meander and large-meander paths show that the large-meander path is mostly formed from the nearshore non-large-meander path and always changes to the offshore non-large-meander path.     

Spectral properties of sea level and time scales of Kuroshio path variations

Spectral properties of sea levels at Naze, Nishinoomote, Kushimoto, Uragami, Miyake-jima and HachijÔ-jima are examined for the non-large-meander (February 1964 – May 1975) and large-meander (October 1975 – December 1979) periods, and the periodicity of variation of the Kuroshio path is clarified.The large meander of the Kuroshio occurs with a primary period of about 20 years and secondary period of 7 to 8. 5 years. During the non-large-meander period, the Kuroshio alternately takes the nearshore and offshore non-large-meander paths with a primary period of 1. 6–1. 8 years. This variation is moreover composed of 110-day, around 195-day and annual periods. The 110-day variation of the Kuroshio path appears to have influence on the coastal sea levels between the Kii Peninsula and the Izu Ridge;i. e., the coastal sea levels rise and fall with one-month time lag after the Kuroshio has begun to approach and leave the Japanese coast. During the large-meander period, the 70 and 110-day variations are remarkable in sea levels south of Japan except Miyake-jima and HachijÔ-jima. The 70-day variation is highly coherent throughout the south coast of Japan; the coherent area of the 110-day variation seems to be smaller.The sea-level variations at Naze and Nishinoomote are not significantly coherent for any of the periods except for annual and semiannual cycles during both the non-large-meander and large-meander periods. That is, the sea-level variations are incoherent between the onshore and offshore sides of the Kuroshio, except for seasonal variation.     

Sea level changes south of Japan associated with the non-large-meander path of the Kuroshio

Sea levels south of Japan from 1964 to 1975 are examined in terms of the nearshore and offshore non-large-meander (NLM) paths of the Kuroshio and the transitions between them.The sea-level anomalies from the annual variations on the south coast of Japan are much larger during the transition from the nearshore to offshore NLM paths than during the reverse transition by 9 cm on average. This characteristic can be seen only in the coastal region of the Kuroshio-flowing area, so that the sea-level difference of Naze minus Nishinoomote (indicator of Kuroshio velocity) during the offshore to nearshore transition is larger by 15 cm than during the reverse transition.The transition from the offshore to nearshore NLM paths occurs when the velocity of the Kuroshio is large or increasing, while the nearshore to offshore transition occurs when it is small or decreasing. The former transition occurs whenever the velocity increases greatly, whereas the latter one does not always occur even though the velocity decreases.The sea-level difference between Kushimoto and Uragami is highly coherent with the alternate appearance of the nearshore and offshore NLM paths. Offshore NLM paths longer than 2.5 months appear during large falls of the sea-level difference of Kushimoto minus Uragami, while large rises of the sea-level difference correspond to long-lasting nearshore NLM paths. The mean sea-level difference during the nearshore NLM path is larger by 4 cm than that during the offshore NLM path.     

Influence of the Kuroshio Fluctuations on Sea Level Variations along the South Coast of Japan

The correlation between the Kuroshio and coastal sea level south of Japan has been examined using the altimetry and tide gauge data during the period 1992–2000. The sea level varies uniformly in a region bounded by the coast and the mean Kuroshio axis, which stretches for several hundred kilometers along the coast. These variations are related with the Kuroshio velocity, as coastal sea level decreases (or increases) when the Kuroshio is faster (or slower). To the east of the Kii Peninsula, where sea level variations are different from these to the west, movement of the Kuroshio axis additionally affects coastal sea level variations.     

Sea level variations at the Izu Islands and typical stable paths of the Kuroshio

Differences in daily mean sea level between Kushimoto and Uragami and daily mean sea levels at Miyake-jima and HachijÔ-jima in the Izu Islands are examined during the 1964–1980 period, and characteristics of the typical paths of the Kuroshio corresponding to the dominant sea level states are described.Sea levels at the islands show three dominant states: high and low sea level states (45 % and 31 %) in the non-large-meander period (October 1963 –7 August 1975) and high sea level state (64 %) in the large-meander period (8 August 1975–15 March 1980). This indicates the existence of three typical paths of the Kuroshio, and the states correspond to the nearshore and offshore non-large-meander paths and the typical large-meander path, respectively. The first path is located near the coast throughout the whole southern area off Japan, the second path leaves the coast around the Izu Ridge and passes south of HachijÔ-jima, and the third path is located near the coast over the ridge after meandering far to the south of Enshû-nada.The positions of the three typical paths are almost the same in the farthest upstream and downstream regions south of Japan between 131E and 142E. The nearshore and offshore non-large-meander paths overlap between Kyûshû and the Kii Peninsula, being invariably close to the coast, while the typical large-meander path south of Shikoku is located offshore and changes its position meridionally.At the mid-depth of 400 m the nearshore non-large-meander and typical large-meander paths pass the Izu Ridge through the deep channel between Miyake-jima and HachijÔ-jima, while the offshore non-large-meander path passes through the deep region south of HachijÔjima. The path of the Kuroshio at mid-depth is well constrained by the bottom topography of the Izu Ridge.     

Favorable conditions for cold-water intrusion from the Kuroshio intermediate layer into Osaka Bay

From 1980 to 1995, in August, the bottom layer of Osaka Bay was occupied by cold, nutrient-rich water compared with that observed during both previous and subsequent decades. To investigate the mechanisms for the intrusion of bottom-layer cold water into Osaka Bay, the intrusion into Osaka Bay via the Kii Channel is simulated using a finite-volume coastal ocean model with unstructured triangular cell grids. The initial conditions, boundary conditions, and surface temperature given to the model are obtained from daily reanalysis data provided by the Japan Coastal Ocean Predictability Experiment. The model shows that cold water uplifted on the eastern side of the Kii Peninsula is propagated westward at 1.0 m/s as a coastal boundary current; it reaches the Kii Channel mouth when the Kuroshio axis is located around 74 km south of Cape Shionomisaki. However, the modeled cold water mass at the Kii Channel mouth does not intrude further to the north of the Kii Channel; therefore, another mechanism is required to explain the cold-water intrusion into the bottom layer of Osaka Bay. A plausible explanation is the estuarine circulation established by the freshwater supply at the bay head. When the river runoff is included in the model without forced vertical mixing, the temperature in Osaka Bay decreases 6.6 days later than the temperature decreases at the Kii Channel mouth. Furthermore, the shoreward current speed in the bottom layer of the modeled estuarine circulation is 15 cm/s, which provides the mechanism required for the cold water mass to pass the Kii Channel.     

Prediction of slope water intrusion into the Kii channel in summer

Intrusions of the warm, oligotrophic surface slope water (SSW) and the cold, nutrient-rich bottom slope water (BSW) from the continental slope influence the annual variations in water temperature and nutrient concentrations in the Kii Channel in August. In order to evaluate the relationships between both these intrusions and the distance of the Kuroshio axis from Cape Shionomisaki (Kuroshio distance), a Distance-Intrusion-Diagram (DID) for temperature, which can reproduce the vertical temperature profile of the channel, was constructed by analyzing the temperature and Kuroshio distance records in August for 1967–2001. DIDs for nutrients (nitrate and phosphate) are also constructed by using the relationship between the nutrient concentration and water temperature. The only explanatory variable in the DIDs is the Kuroshio distance. The DID for temperature predicts that the SSW occupies almost the entire water column when the Kuroshio approaches Cape Shionomisaki (Kuroshio distance = 18.5 km). When the Kuroshio distance lies in the range 18.5–74 km, the BSW thickness increases proportionally to the Kuroshio distance increment while the SSW thickness decreases. The BSW occupies the largest portion of the channel when the Kuroshio distance is 74 km. Further, beyond 74 km, the BSW thickness reduces gradually. Yearly variations in the temperature and concentrations of nitrate and phosphate were hindcast with the DIDs. The results revealed that the Kuroshio distance contributes 70%, 35%, and 30% of the variances in temperature, nitrate concentration, and phosphate concentration, respectively.     

A steady model of typical non-large-meander paths of the Kuroshio Current

Conditions south of Cape Shiono-misaki for the nearshore and offshore non-large-meander (NLM) paths of the Kuroshio Current are studied using a two-layer reduced gravity model. A steady and non-diffusive state is assumed, and the conservation laws of Bernoulli's function and potential vorticity along the current axis are used. Spatial changes of velocity and depth of the current are imposed as boundary conditions south of Cape Shiono-misaki. These conditions are based on the facts that are ohserved south of the cape: abrupt acceleration of the Kuroshio and the spatial change of sea levels. The current paths east of the cape are computed.This model reproduces well the actual nearshore NLM paths. It also produces offshore NLM paths west of and over the Izu Ridge, but not east of the Izu Ridge. Diffusion of vorticity may be important for the eastern part of the offshore NLM path.An increase of velocity south of the cape is necessary for producing realistic NLM paths. The velocity increase accompanies a decrease of current depth owing to Bernoulli's conservation, and the depth decrease in turn diminishes the absolute vorticity owing to the potential vorticity conservation. The velocity increase, on the other hand, strengthens the negative velocity shear and diminishes the relative vorticity. If the decreases of the relative and absolute vorticities compensate each other, the path goes excessively southward owing to the negative curvature south of the cape. Dominance of the relative vorticity change over the absolute vorticity change prevents, the path from shifting southward and causes the realistic NLM paths.The NLM paths need different amplitudes of the changes south of the cape depending on the velocity and transport of the current, but in any case, the nearshore NLM path needs larger changes than the offshore NLM path. This property and the amplitude of the changes are consistent with observations.     

Path Transition of the Kuroshio Due to Mesoscale Eddies: A Two-Layer, Wind-Driven Experiment

We have executed numerical experiments using a two-layer, wind-driven ocean model with simplified coastal geometry and bottom topography to investigate the possibility of the Kuroshio path transition due to mesoscale eddies. A straight path easily changes into a meandering path due to the eddy action. For this transition, an anticyclonic eddy is preferable to a cyclonic one when imposed in the beginning region of the Kuroshio (east of Taiwan). When imposed southeast of Kyushu, on the other hand, a cyclonic eddy is more effective than an anticyclonic one. The reverse transition, from a meandering to a straight path, did not occur at all in this experiment. This revised version was published online in August 2006 with corrections to the Cover Date.     

Monitoring of generation and propagation of the Kuroshio small meander using sea level data along the southern coast of Japan

The occurrence of the small meander of the Kuroshio, generated south of Kyushu and propagating eastward, was examined using sea level data collected during 1961–1995 along the south coast of Japan. Intra-annual variation of the sea level was expanded by the frequency domain empirical orthogonal function (FDEOF) modes, and it was found that the second and third modes are useful for monitoring the generation and propagation of the small meander. The third FDEOF for periods of 10–100 days has a phase reversal between Hosojima and Tosa-shimizu with significant amplitude west of Kushimoto, and the amplitude of its time coefficient is large during the non-large-meander (NLM) period and has a significant peak when the small meander exists southeast of Kyushu. The second FDEOF for periods of 20–80 days has a phase reversal between Kushimoto and Uragami, and the amplitude of its time coefficient is large when the small meander propagates to the south of Shikoku. The third FDEOF mode allowed us to conclude that the small meander occurred 42 times from July 1961 to May 1995, most of them (38) occurring during the NLM periods. The second FDEOF mode permits the conclusion that half of the 38 small meanders reached south of Shikoku. Of these, five small meanders influenced transitions of the Kuroshio path from the nearshore NLM path; one caused the offshore NLM path and four brought about the large meander. About one-tenth of the total number of small meanders are related to the formation of the large meander.     

Chaotic Advection of the Shallow Kuroshio Coastal Waters

The shallow coastal water of the Enshu-Nada Sea (ENSW) recirculates regardless of whether the Kuroshio path is straight or has meanders. The recirculation is formed as a result of flow separation at the sharp coastline. The outputs of a recent numerical simulation of the Kuroshio current, including case of a short-term meander caused by an anticyclonic eddy, were analysed to track the motion of the ENSW. In contrast to the steady-flow cases, the unsteady cases showed that the ENSW discharges into the Kuroshio Extension region and intrudes further north into the Kuroshio-Oyashio confluence region due to chaotic advection. Two hyperbolic stagnation points of the velocity field characterise the transport paths; one south of the Izu peninsula and the other at the Kuroshio Extension. This mechanism exists even without the Ekman drift and may play an important role in the transportation of the fish eggs and larvae from the southern Japan spawning ground to the food abundant Kuroshio-Oyashio transition zone. This revised version was published online in August 2006 with corrections to the Cover Date.     

Numerical Study of a Kyucho and a Bottom Intrusion in the Bungo Channel,Japan: Disturbances Generated by the Kuroshio Small Meanders

The generation and propagation mechanisms of a Kyucho and a bottom intrusion in the Bungo Channel, Japan, have been studied numerically using the hydrostatic primitive equations by assuming density stratification during summer. The experiments are designed to generate a Kuroshio small meander in Hyuga-Nada, which acts as a trigger for these disturbances. After the current speed of the Kuroshio is changed, a small meander is generated. At the head of the small meander, warm Kuroshio water is engulfed, and encounters the southwest coast of Shikoku. However, convergence of heat flux on the bump off Cape Ashizuri suppresses the generation of a warm disturbance, if the current speed is large. As the cold eddy associated with the small meander approaches Cape Ashizuri, the heat flux diverges on the bump. This heat source forces a warm disturbance, which intrudes along the east coast of the Bungo Channel as a baroclinic Kelvin wave (a Kyucho). After the cold eddy passes off Cape Ashizuri, the Kuroshio approaches the bump again. Strong convergence of heat flux then occurs on the bump, which forces a cold disturbance. This disturbance propagates as a topographic Rossby wave along the shelf break at the mouth of the channel. After the topographic wave reaches the west end of the shelf break, it intrudes along the bottom layer of the channel as a density current (a bottom intrusion). These results suggest that a Kyucho and a bottom intrusion are successive events associated with the propagation of the small meander.     

Warm water intrusion from the Kuroshio into the coastal areas south of Japan

Sea surface temperature (SST) has been measured in the south of Japan using a thermometer set up in the ferry boat to investigate the characteristics of the warm water intrudes into the coastal areas from the Kuroshio. Time series analysis was applied to the SST data with satellite images and hydrographic observation data from April 1987 to September 1989. The results indicate that the warm Kuroshio water intruded into the coastal areas on the Enshu-nada and the Kumano-nada Seas intermittently with periods of about 50 and 20 days associated with the fluctuation of the Kuroshio path and the Kuroshio frontal disturbance respectively. The intrusion with a 50-day period was dominant when the Kuroshio took a stationary small meander path (B- and C-types). The warm water spread to the west at 20 cm s^–1, and was estimated to have a depth of 150 m at least and supply enough heat to make up the loss due to the evaporation in the coastal area. During the straight path of the Kuroshio, it was detected that the warm water intruded into coastal areas only with a 20-day period. The warm water that intrudes with a period of 20 days spreads to the west at 25 cm s^–1 in a small scale.