Warm Water Intrusion from the Kuroshio into the Coastal Areas South of Japan

Transcription

1 Journal of Oceanography Vol. 49, pp. 607 to Warm Water Intrusion from the Kuroshio into the Coastal Areas South of Japan AKIHIDE KASAI, SHINGO KIMURA and TAKASHIGE SUGIMOTO Ocean Research Institute, University of Tokyo, Minamidai, Nakano-ku, Tokyo 164, Japan (Received 22 September 1992; in revised form 25 January 1993; accepted 22 April 1993) 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 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. 1. Introduction It is well known that the coastal oceanographic conditions in the Pacific side of Japanese Islands from the south of Kyushu to Boso Peninsula are significantly affected by the fluctuating of the Kuroshio path and its branched current (Fujimoto, 1987; Imasato and Qiu, 1987; Awaji et al., 1991). In particular, the Kuroshio usually takes a very flexible path in the south of the Kumano-nada Sea (KNS) and Enshu-nada Sea (ENS) (Kawabe, 1985). Therefore, warm water from the Kuroshio intrudes into these coastal areas with high frequency. Sometimes these intrusions cause changes in fish migration and behavior, and continuously affect fishing conditions in the coastal areas (Nishimura, 1987; Takeuchi, 1987; Sugimoto and Tameishi, 1992). Kimura and Sugimoto (1988, 1990) found that warm water masses associated with short time variations of the Kuroshio path intrude into the coastal areas with a period of about 20 days. They also pointed out that the intrusions have considerable effects on the coastal oceanographic and fishing conditions. The intrusion would also affect heat budget in the coastal region according to the frequency of the intrusions. Toba et al. (1991) observed the southwestward outbreaks from the Kuroshio and showed that a large heat was transported to the outer region. The warm water could transport heat budget to the coastal area in the same manner. Accordingly, detailed description of the intrusion is necessary to make clear the relationship between these phenomena and the intrusion. Furthermore, it is important to explain the mechanism of the intrusion from the Kuroshio into the coastal areas such as the KNS or the ENS. In recent years, infrared satellite images have been used to provide information for horizontal oceanic changes in the coastal areas (Shibata, 1983; Matsumura, 1986; Takeuchi,

2 608 A. Kasai et al. 1987; Qiu et al., 1990). However, the analysis of the interaction between the KNS or the ENS and the Kuroshio based on satellite images is difficult because consecutive cloud-free images are rare. Another continuous information of the oceanic conditions is required to discuss the interaction between the Kuroshio and the coastal areas which fluctuate with a period of a few weeks. Sea surface temperature (SST) and velocity observed by using ferry boats are useful for the analyses. Nagata and Takeshita (1985) analyzed the SST data in the Tokara Strait obtained by a ferry boat, and found that the continuous observation of the SST across the strait yields a good tool for monitoring fluctuations of the Kuroshio path. Sugimoto and Kobayashi (1987) developed the observational system for ferry boats to obtain the sea surface temperature and current velocity with ship drift simultaneously. In this paper, we analyze SST observed by a ferry boat in the south of Japan, satellite thermal images and CTD data to study the characteristics of short time fluctuation of the temperature in the coastal area. In particular, the periodicity and horizontal and vertical structure of the warm water intruding into ENS or KNS from the Kuroshio are studied according to the familiar classification of non-large meander type of the Kuroshio path. 2. Observational Methods SST in the coastal waters, inshore side of the Kuroshio have been observed by a monitoring Fig. 1. Locations of ship cruising course (solid line) and stations at which sea surface temperature (SST) was analyzed ( ). Numerals attached to the stations indicate station numbers. KNS and ENS denote the Kumano-nada and the Enshu-nada Seas respectively. Open and solid triangles indicate CTD stations observed by Shizuoka Prefectural Experimental Fishery Station and by Japan Meteorological Agency respectively.

3 Warm Water Intrusion into Coastal Areas 609 Fig. 2. Typical paths of the Kuroshio. Lines B, Bs, C, Cs, and N denote the B-, Bs-, C-, Cs-, and N-type paths respectively. system set up in a ferry named Sunflower Tosa which shuttles between Tokyo and Kouchi via Shingu for two days. The ferry boat cruises in a straight line between 136 E and 139 E as shown in Fig. 1. A sensor attached to an intaking pipe for cooling engines measures SST every five minutes (Sugimoto and Kobayashi, 1987; Kimura and Sugimoto, 1990). Temperatures for desired geographic locations were determined on the basis of the ship s location and time lapse monitored by a Loran-C or a GPS (Global Positioning System) receiver. The data observed from April 1987 to September 1989 were analyzed in this paper. For detailed understanding of horizontal structure of the warm water from the Kuroshio, infrared thermal images from the satellite NOAA-11 were used. CTD hydrographic data on a monthly basis along E observed by Shizuoka Prefectural Experimental Fishery Station and data along 138 E by Japan Meteorological Agency were also used to describe the vertical structure of the warm water. The classification of the type of the Kuroshio path is dependent on the definition by Nitani (1969) which is based on the Kuroshio axis (Fig. 2). The Kuroshio axis is determined by a contour of 15 C at a depth of 200 m (Kawai, 1969) on the basis of Prompt Report of Oceanographic Conditions published by the Japan Maritime Safety Agency. The B- and C-type paths are defined as small meander ones which flow through the north and south of Hachijojima Island respectively. The N-type path is defined as a straight one which flows close to the south coast of Japan. In addition, we define S shape meandering B- and C-type paths off Izu Pen. or Boso Pen. as Bsand Cs-type paths respectively. 3. Results 3.1 Satellite thermal images of the intrusion The satellite images of warm water intrusion from the Kuroshio taken on 25 April 1988, 30 July 1988, and 11 March 1989 are shown in Fig. 3. The darker and the brighter tones in the images indicate warmer and colder water respectively. Figure 3(a) shows a typical intrusion pattern when the Kuroshio takes a B-type path. The warm water separated from the main Kuroshio current off Izu Peninsula intrudes into the ENS and moves westward with a filament protruding from the top of the warm water. The warm water shown in Figs. 3(a) and (c) spreads into the

4 610 A. Kasai et al. (a) Fig. 3. Satellite AVHRR thermal images of warm water intrusion; (a) observed 25 April 1988, (b) observed 30 July 1988, and (c) observed 11 March coastal areas and raised the coastal SST. Filaments generated from the Kuroshio front are shown in Fig. 3(b). According to satellite images, since the filaments do not grow to cover the ENS or the KNS, they would have little effect on the coastal areas compared with the large warm water intrusions as shown in Figs. 3(a) and (c). Figure 4 shows a schematic view of the sequence of the warm water intrusion deduced from the satellite NOAA-11 from 9 to 28 March. Figure 4(b) corresponds to Fig. 3(c). The evolution of the warm tongue and the filament in the area of interest is presented in this figure. Initially, a tongue of warm water (T1) intrudes into the Kumano-nada Sea from the south of Cape Shionomisaki and then it extends northeastward on 9 March. At the same time, another tongue stretches out westward off the ENS (T2). It seems that the tongue T1 was generated when the Kuroshio brushed against Cape Shionomisaki. The tongue T2 may have generated by the local S shape meandering path of the Kuroshio off Boso Peninsula which is shown by Bs- or Cs-type paths in Fig. 2. The generation mechanisms are discussed in detail in Subsection 3.2. Figure 4(b) suggests the northeastward movement of T2 and the warm water intrusion into the ENS on 11 March. T1 shows complicated deformation but little development occurred. On 14 March, the tongue T2 merged with tongue T1 and filaments spread from both tongues. After that, the tongue T1 and the west part of T2 mixed with the coastal waters and the both tongues became obscure on 20 March. It is difficult to distinguish between the coastal and the warm water because of the mixture, however, the remaining warm water of T2 is shown in the eastern side of the ENS. A new tongue intrudes into the KNS on 28 March similar to that on 9 March. The sequence of the warm water shows that large parts of the coastal areas were covered with the warm water from the Kuroshio. Moreover, it only took one or two weeks for the warm water to mix with the coastal waters. It reflects that the warm water must have significant effect on the coastal areas.

5 Warm Water Intrusion into Coastal Areas 611 (b) (c) Fig. 3. (continued). 3.2 The warm water intrusions observed by the ferry boat The annual and semiannual components from the SST records are eliminated by subtracting two months running mean value to see the nature of short period variations. Time-space diagrams of SST deviation between 136 E and 139 E from April 1987 to September 1989 are shown in

6 612 A. Kasai et al. Fig. 4. Schematic view of warm water intrusions. The dashed lines indicate the course of the ferry boat. T1 and T2 denote the warm water tongues from the Kuroshio. Fig. 5. In the figure, the white lines indicate the warm water intrusions which are determined by the increase of the SST by more than 2.5 C in 10 days and by the spread more than 100 km along the course of the ferry boat. Since the warm water mass which intrudes into the coastal areas is changing its shape at all times and does not always move in one direction, the lines do not represent the motion of the warm water exactly. Two kinds of warm water intrusions from the Kuroshio into the coastal areas are easily recognized in Fig. 5; one intrudes into the ENS and spreads westward (solid line), and the other intrudes into the KNS with a smaller spreading (dashed line). The large warm water intrusion occurred 14, 10, and 9 times during the period in which the Kuroshio took B-, C-, and N-type paths for 12, 9, and 6 months respectively. Although large intrusions appeared with periods of 30 to 60 days while the Kuroshio took a small meander path, they appeared with a shorter period while the Kuroshio took a straight path. Warm water intruded more often into the ENS than into the KNS when the Kuroshio took a small meander path. The warm water intrusion into the ENS was larger in time and space than

7 Warm Water Intrusion into Coastal Areas 613 Fig. 5. Time-space diagram of the SST deviation; (a) from April 1987 to March 1988, (b) from April 1988 to March 1989, and (c) from April to September White solid and dashed lines indicate the warm water intrusion into the ENS and the KNS respectively. Arrows with B, C, and N denote the periods of B-, C-, and N-type paths of the Kuroshio, respectively. into the KNS except for the intrusion Q (Fig. 5(a)). This huge warm water intrusion at the end of August 1988 was caused by the temporal approach of the main Kuroshio current; the Kuroshio took an N-type path temporally. The intrusion Q, accordingly, was not considered a typical warm water intrusion that often appeared during the small meander path. Besides the large water intrusions which are indicated with white lines, warm water often intrudes into the coastal areas on a smaller scale which appears as complex striped patterns in Fig. 5. For example, a typical pattern is displayed in Fig. 5(b) from January to February 1989 between 136 E and 138 E. Such intrusions occurred with a period of less than one month. The effect of filaments or eddies which derived from the Kuroshio front shown in Fig. 3(b) may appear as the pattern. The relationship between the intrusion into the ENS and the path of the Kuroshio is

8 614 A. Kasai et al. Fig. 5. (continued). investigated to clarify the timing of the warm water intrusion. It was expected that the intrusion into the ENS is related to local S shape meandering pattern off Izu or Boso Peninsula in the previous section. Prompt Report of Oceanographic Conditions shows that the offshore distances from the Kuroshio axis to Izuoshima Island and Cape Inubozaki become short when the Kuroshio takes S shape meandering pattern during B- and C-type paths respectively. Accordingly, the distances are used as indexes of the local S shape meandering in this study. Figure 6 shows the distances and the times of the warm water intrusion into ENS. It is well known that the Kuroshio changes its path in a large scale with a long period. However, Fig. 6 shows that the Kuroshio changed its path with a period of less than a few month and preferred to take S shape meandering path. It is clear from Fig. 6 that when the Kuroshio takes S shaped Bs- or Cs-type path, the warm water tends to intrude into the ENS. On the other hand, the warm water did not intrude and the SST in the ENS decreased while the distance was long, for example October 1987 (Figs. 5 and 6).

9 Warm Water Intrusion into Coastal Areas 615 Fig. 5. (continued). Fig. 6. Off-shore distances of the Kuroshio axis (a) from Izuoshima Island during B-type path, and (b) from Cape Inubozaki during C-type path. Triangles indicate the time of the warm water intrusion into the ENS.

10 616 A. Kasai et al. Fig. 7. Power spectra of SST at (a) St. 4, (b) St. 10, and (c) St. 16 during the small meander type (B- and C-type) of the Kuroshio path (from April 1987 to March 1989).

11 Warm Water Intrusion into Coastal Areas 617 Fig. 8. Power spectra of SST at (a) St. 4, (b) St. 10, and (c) St. 16 during the straight type (N-type) of the Kuroshio path (from April to September 1989).

12 618 A. Kasai et al. Although the warm water often intrudes into the coastal areas during N-type path which indicated by white line in Fig. 5(c), it is doubtful whether the warm water is similar to the one during B- or C-type path. This is related to locations of the observation line and the Kuroshio path. Since the Kuroshio flows near the cruise line during N-type path, eddies or filaments derived from the frontal disturbances would be cross the line and rise the SST easier than during B- and C-type path. Satellite images (which are not shown here) show that large warm water rarely intrudes into the ENS when the Kuroshio takes N-type path. A wake is formed behind a peninsula and a ridge when the Kuroshio brushes them as shown by Fig. 4. The intrusion into the KNS could be caused by the wake associated with the approach of the Kuroshio current to Cape Shionomisaki. Accordingly, the offshore distance from the Kuroshio axis to Cape Shionomisaki would be related to the intrusion of the warm Kuroshio water into the KNS. Since the warm water infrequently intruded into the KNS on a large scale, there is a less correlation between the intrusion and the distance unfortunately. However, Fig. 5 shows that the warm water in the period of C- or N-type path was more likely to intrude into the Fig. 9. (a) Coherence squared and (b) phase lags of 50-day fluctuation between the SST at St. 17 and all stations during small meander type (B- and C-type) of the Kuroshio path. Lags are presented in days.

13 Warm Water Intrusion into Coastal Areas 619 KNS than during B-type path. The distance from Cape Shionomisaki to the Kuroshio during C- or N-type path becomes shorter than that during B-type path. Therefore, the short-termed change of the Kuroshio path is one of the main causes of the warm water intrusion into the ENS and KNS. 3.3 Periodicity of the warm water intrusion The spectra of the time series of raw SST were calculated using Fast Fourier Transform. The 19 stations with 10 minute longitude intervals between 136 E and 139 E shown in Fig. 1 were selected for time series analysis. The spectra of SST at St. 4, 10, and 16 during small meander path (B- and C-type paths) and straight path (N-type path) are presented in Figs. 7 and 8 respectively. St. 4, 10, and 16 are located on the western, middle, and eastern parts of the observation line respectively (Fig. 1). Figure 7 shows that the SST fluctuations have prominent spectral peaks at an about 50-day period at the all stations during small meander type of the Kuroshio path. They also have less prominent but distinct peaks at an about 20-day period at the western stations on the cruise line Fig. 10. (a) Coherence squared and (b) phase lags of 20-day fluctuation between the SST at St. 1 and all stations during straight type (N-type) of the Kuroshio path. Lags are presented in days.

14 620 A. Kasai et al. regardless of the path (Figs. 7 and 8). The peak with 50-day periodic fluctuation reflects the large size warm water intrusions into the ENS which are indicated by white solid lines in Figs. 5(a) and (b). Since the warm water often intruded into the ENS and spread to the KNS, the peak at an about 50-day period in the eastern part of the observation line is recognized clearer than that in the western part. Figures 9 and 10 show the coherences squared and phase lags of 50- and 20-day periods phenomena at each station relative to St. 17 and St. 1 respectively. Adequate correlations over the all stations reflect that the whole coastal areas are covered with the large warm water intruding with 50-day period (Fig. 9(a)). The phase lags behind as one moves eastward from St. 17. The phase lag is about 60 degree between St. 17 and St. 7; it indicates that the warm water spreads to the west at about 20 cm s 1 (Fig. 9(b)). The warm water intruding with a 20-day period extends in small area, because the SST fluctuations are highly coherent only between St. 1 and stations in the KNS (Fig. 10(a)). This is consistent with the energy peak at 20 days becomes obscure in the middle part of the observation line (Figs. 7 and 8). The warm water which intrudes with a period of 20 days is small and it mixes with coastal waters or diffuse in a short time. It is indicated from Fig. 10(b) that the warm water propagates to the west at 25 cm s 1 in the KNS. 3.4 Vertical structure and heat release of the warm water A large warm water intrusion into the KNS occurred from late January to early February 1989 indicated as intrusion R in Fig. 5(b). Shizuoka Prefectural Experimental Fishery Station carried out salinity temperature-depth observations on a monthly basis along E. A schematic image of the warm water intrusion R on February 7 and the observation line are shown in Fig. 11 which represents the line across the intrusion near its tip. The vertical structure of the warm water is investigated by comparing the three vertical distributions of the temperature and the salinity observed on early January, February, and March. On early January and March, the warm water did not intrude. Figure 12 shows the vertical sections along E (Fig. 2) on January 11, February 7, and March 2, Sections for the warm water intrusion indicated as R in Fig. 5 correspond to Fig. 12(b). Comparing Fig. 12(b) with Fig. 12(a) which shows distributions when no intrusion occurred, a cyclonic eddy accompanied by an uplifted cold and less saline water, and warm and saline Kuroshio water lay to the north of the eddy were recognized in February. However, both the warm water and cold eddy disappeared in March as shown in Fig. 12(c). The figure shows that the temperature and salinity of the water shallower Fig. 11. Schematic image of the warm water intrusion deduced from the satellite image taken on February 7, 1989.

15 Warm Water Intrusion into Coastal Areas 621 Fig. 12. Vertical temperature ( C) and salinity (anomaly from 34.00, ) sections observed by R/V Surugamaru along N on (a) January 11, (b) February 7, and (c) March 2, 1989.

16 622 A. Kasai et al. than 150 m clearly increased in the warm water and the isograms were close at the depth of 150 m in February (Fig. 12(b)). Japan Meteorological Agency observed CTD hydrographic data along 138 E on May 4 and 5 in 1987 which was across the warm water intrusion P (Fig. 5(a)). The vertical distributions of the temperature and salinity (which are not shown here) show a similar structure to Fig. 12(b) and the depth of the warm water is about 150 m. It seems that the warm water which intrudes from the Kuroshio have a depth of 150 m at least, although it attain deeper because both observation lines crossed near the tip of the warm water. 4. Discussion Based on the SST data observed by using a ferry boat, satellite thermal images, and CTD data, characteristics of the warm water from the Kuroshio into coastal areas in the south of Japan were analyzed in the period of the Kuroshio took non-large meander path. The results show that the warm water intrusion from the Kuroshio into the ENS and the KNS is characterized by two remarkable types. The schematic images of the two typical warm water intrusions are illustrated in Fig. 13. One is the large intrusion into the ENS with a period of about 50 days. The intrusion is associated with the local S shape meandering of the Kuroshio path and spreads into the whole coastal region. Taira and Teramoto (1981) measured the current velocity in the large meandering Kuroshio and they found fluctuations in two periodic bands; one is near 33 days and the other is near 100 days. Their study agreed with our result that the fluctuations of the Kuroshio have two characteristic periods but disagreed with respect to the periods. The 50-day periodic fluctuation is recognized in this study when the Kuroshio takes a small meander path but not a straight path. Accordingly, it is considered that the warm water intrusions with an about 50-day period is typical phenomena in the period of small meander path but are uncommon large meander or straight path of the Kuroshio. The warm water mixed with the coastal waters and the structure disappeared within one Fig. 13. Schematic images of the warm water intrusion; (a) caused by the local S shape meandering Kuroshio path and (b) associated with the Kuroshio frontal disturbance.

17 Warm Water Intrusion into Coastal Areas 623 month (Fig. 4). Assuming that the warm water which intrudes into coastal areas with 50-days period has a width of 100 km, a length of 200 km, and 2.5 degrees in temperature higher than the coastal waters, the volume of the warm water is m 3 and heat inflow into the coastal area is estimated to be about J by the one event. Since the warm water intrudes into coastal areas with a period of 50 days, the rate of heat transport to the ENS and the KNS is calculated to be 130 W/m 2. The warm water intrusion supplies enough heat to make up the loss due to the evaporation because the rate of heat loss by the evaporation is estimated to be about 150 W/m 2 in this area (Pickard and Emery, 1982). The other type of the intrusion is in a small scale with a period of 20 days and the warm water propagates at 25 cm s 1 to the west. The 20-day periodic fluctuation of the Kuroshio is well recognized in temperature and velocity fields. Shibata (1983) found temperature fluctuations with a period of 20 days in the East China Sea and attributed the fluctuations to the Kuroshio front meanders. Qiu et al. (1990) also revealed the small meanders of the Kuroshio front with periods of 14 to 20 days. Kimura and Sugimoto (1993) measured current velocity of the Kuroshio off Cape Shionomisaki and found that the Kuroshio has a fluctuation with a period of days. These studies and our analysis suggest that the peak with 20-days fluctuation was generated by the small intrusion caused by the disturbance of the Kuroshio front propagating from upstream side. Characteristics such as wave length, phase speed, and periodicity of the intrusion from the Kuroshio have been well described in advanced observations. However, it still remains to find out the cause of those characteristics. Studies of the frontal meander and disturbance based on the numerical models have been done in the Gulf Stream region or California Current System. For example, Ikeda et al. (1984) used analytical and numerical models to represent the California Current System and showed that the topographic features with a spatial periodicity act as a trigger to initiate the short wavelength meanders. It is necessary to apply the models to the Kuroshio Current to ensure completion of the cause of the intrusion. Acknowledgements The authors wish to express their thanks to the captains and crews of Sunflower Tosa for their support and help in making the observations and also to Shizuoka Prefectural Experimental Fishery Station for providing the monthly observation data. References Awaji, T., K. Akitomo and N. Imasato (1991): Numerical study of shelf water motion driven by the Kuroshio: Barotropic model. J. Phys. Oceanogr., 21, Fujimoto, M. (1987): On the flow types and current stability in Tosa Bay and adjacent seas. Umi to Sora, 62, (in Japanese). Ikeda, M., W. J. Emery and L. A. Mysak (1984): Seasonal variability in meanders of the California Current System off Vancouver Island. J. Geophys. Res., 89, Imasato, N. and B. Qiu (1987): An event in water exchange between continental shelf and the Kuroshio off southern Japan: Lagrangian tracking of a low-salinity water mass on the Kuroshio. J. Phys. Oceanogr., 17, Kawabe, M. (1985): Sea level variations at the Izu Islands and typical stable paths of the Kuroshio. J. Oceanogr. Soc. Japan, 41, Kawai, H. (1969): Statistical estimation of isotherms indicative of the Kuroshio axis. Deep-Sea Res., 16, Kimura, S. and T. Sugimoto (1988): Characteristics of short period fluctuations in oceanographic and fishing conditions in the coastal area of Enshu-nada Sea. Bull. Japan. Soc. Fish. Oceanogr., 52, (in Japanese). Kimura, S. and T. Sugimoto (1990): Intrusion processes of warm water mass from the Kuroshio into the coastal area of Kumano-nada and Enshu-nada Seas. Bull. Japan. Soc. Fish. Oceanogr., 54, (in Japanese).

18 624 A. Kasai et al. Kimura, S. and T. Sugimoto (1993): Short-period fluctuations in meander of the Kuroshio path off Cape Shionomisaki. J. Geophys. Res., 98, Luther, M. and J. M. Bane (1985): Mixed instabilities in the Gulf Stream over the continental slope. J. Phys. Oceanogr., 15, Matsumura, S. (1986): Fluctuation of sea condition at the Kuroshio coastal side counter current area observed by NOAA-APT images. Sora to Umi, 8, (in Japanese). Nagata, Y. and K. Takeshita (1985): Variation of the sea surface temperature distribution across the Kuroshio in the Tokara Strait. J. Oceanogr. Soc. Japan, 41, Nishimura, A. (1987): The variability of hydrographic condition in Kumano-nada Sea from the satellite thermal images. Mar. Sci. Mon., 19, 8, (in Japanese). Nitani, H. (1969): Variation of the Kuroshio path in a few years. Bull. Japan. Soc. Fish. Oceanogr., 14, (in Japanese). Pickard, G. and W. J. Emery (1982): Descriptive Physical Oceanography. Pergamon Press, Oxford, 249 pp. Qiu, B., T. Toda and N. Imasato (1990): On Kuroshio front fluctuations in the East China Sea using satellite and in situ observational data. J. Geophys. Res., 95, C10, Shibata, A. (1983): Meander of the Kuroshio along the edge of continental shelf in the East China Sea. Umi to Sora, 58, (in Japanese). Sugimoto, T. and M. Kobayashi (1987): Current observation system using ship drift and its applications in Kumanonada and Enshu-nada Seas. J. Oceanogr. Soc. Japan, 43, Sugimoto, T. and H. Tameishi (1992): Warm-core rings, streamers and their role on the fishing ground formation around Japan. Deep-Sea Res., 39, Taira, K. and T. Teramoto (1981): Velocity fluctuations of the Kuroshio near the Izu Ridge and their relationship to current path. Deep-Sea Res., 10, Takeuchi, J. (1987): Warm water tongue and coastal upwelling in the south of Kumano-nada Sea. Mar. Sci. Mon., 19, 8, (in Japanese). Toba, Y., H. Kawamura, K. Hanawa, H. Otobe and K. Taira (1991): Outbreak of warm water from the Kuroshio south of Japan A combined analysis of satellite and OMLET oceanographic data. J. Oceanogr. Soc. Japan, 47,