Trajectory of Mesoscale Eddies in the Kuroshio Recirculation Region

Transcription

1 Journal of Oceanography, Vol. 57, pp. 471 to 480, 2001 Trajectory of Mesoscale Eddies in the Kuroshio Recirculation Region NAOTO EBUCHI 1 * and KIMIO HANAWA 2 1 Center for Atmospheric and Oceanic Studies, Graduate School of Science, Tohoku University, Aoba, Sendai , Japan 2 Department of Geophysics, Graduate School of Science, Tohoku University, Aoba, Sendai , Japan (Received 12 April 2000; in revised form 9 January 2001; accepted 10 January 2001) Trajectories of mesoscale eddies in the Kuroshio recirculation region were investigated by using sea surface height (SSH) anomaly observed by the TOPEX/POSEIDON and ERS altimeters. Cyclonic and anticyclonic eddies have been traced on maps of the filtered SSH anomaly fields composed from the altimeter observations every ten days. Both the cyclonic and anticyclonic eddies propagate westward in the Kuroshio recirculation region from a region south of the Kuroshio Extension. The propagation speed of these eddies has been estimated as about 7 cm s 1, which is much faster than the phase speed theoretically estimated for the baroclinic first-mode Rossby wave in the study area. It was also found that in the Izu-Ogasawara Ridge region, most of eddies pass through the gap between the Hachijojima Island and Ogasawara (Bonin) Islands, and some of the eddies decay around the Izu-Ogasawara Ridge. It seems that the trajectory of the eddies is crucially affected by the bottom topography. In the region south of Shikoku and east of Kyushu, some of the eddies coalesce with the Kuroshio. It is also suggested that this coalescence may trigger the path variation of the Kuroshio in the sea south of Japan. Keywords: Mesoscale eddy, Kuroshio recirculation, satellite altimetry, remote sensing. 1. Introduction Ebuchi and Hanawa (2000) investigated mesoscale eddies in the Kuroshio recirculation region south of Japan by using surface current data measured by an Acoustic Doppler Current Profiler installed on a ferry that shuttles regularly between Tokyo and Chichijima, Ogasawara (Bonin) Islands, and sea surface height (henceforth SSH) anomaly derived from the TOPEX/POSEIDON altimeter. They found many cyclonic and anticyclonic eddies in the Kuroshio recirculation region. It was confirmed by XBT cross sections that these eddies have a robust subsurface structure with displacement of isotherms comprising the main thermocline greater than 200 m. The spatial and temporal scales of the eddies were determined by lag-correlation analyses in space and time. The eddies are circular in shape with a diameter of 250 km and a temporal scale of 80 days. The typical maximum surface velocity and SSH anomaly associated with the eddies are * Corresponding author. ebuchi@ocean.caos.tohoku. ac.jp Copyright The Oceanographic Society of Japan. cm s 1 and 15 cm, respectively. The frequency of occurrence, temporal and spatial scales, and intensity are all nearly the same for the cyclonic and anticyclonic eddies. The westward propagation speed of the eddies is estimated to be 6.8 cm s 1, which is much faster than the phase speed theoretically estimated for the baroclinic first-mode Rossby wave with or without a mean current. Ebuchi and Hanawa (2000) also suggested that these eddies may be generated in the Kuroshio Extension region, propagating westward in the Kuroshio recirculation region, based on the spatial distribution of SSH variations. There have also been several studies of these eddies using hydrographic observations and drifting buoys (e.g., Kawai, 1972; Kitano, 1975; Cheney et al., 1980; Mizuno and White, 1983), infrared images from satellites (e.g., Yasuda et al., 1992), and SSH fields observed by satellite altimeters (e.g., Ichikawa and Imawaki, 1994; Aoki et al., 1995; Aoki and Imawaki, 1996; Kobashi and Kawamura, 2001). In most of these studies, however, attention has been paid only to cold (warm) eddies generated on the south (north) of the Kuroshio Extension. In addition, the processes of generation, propagation, and decay of the eddies have not been fully understood. 471

2 In the present study, time series of SSH anomaly maps composed from observations by the TOPEX/ POSEIDON and ERS altimeters have been utilized to trace the mesoscale eddies in the Kuroshio recirculation region which were investigated by Ebuchi and Hanawa (2000) along the Tokyo-Ogasawara line. As demonstrated by the previous studies, the SSHs observed by satellite altimeters with wide coverage in space and high resolution in time provide an essential tool to investigate the spatial and temporal evolutions of the mesoscale eddies. 2. Data In the present study we utilize the AVISO Maps of Sea Level Anomalies (MSLA) altimeter products for the first five years of the combined TOPEX/POSEIDON and ERS-1/ERS-2 missions. These products are provided by the Collecte Localisation Satellites (CLS) Space Oceanography Division, Toulouse, France. The maps were generated from the along-track geophysical data records of the two altimeters for a period from October 22, 1992 to June 3, 1998, and were produced, using the conventional repeat-track analysis method, from the altimeter SSHs corrected for instrumental errors, environment perturbations (wet tropospheric, dry tropospheric, and ionospheric effects), ocean wave influence (sea state bias), tide influence (ocean tide, earth tide, and pole tide) and inverse barometric effect (AVISO/Altimetry, 1996). Orbit error was also corrected for the ERS data (Le Traon et al., 1995; Le Traon and Ogor, 1998). The maps were obtained using an improved space-time objective analysis method which takes account of long-wavelength errors (Le Traon et al., 1998). The MSLA products are available every 10 days with a (latitude longitude) spatial resolution. Fig. 1. (a) Bathymetry of the study area. Contours are drawn in the area shallower than 4000 m at intervals of 2000 m (thick line) and 500 m (thin line). Labels a h represent: a, Kyushu; b, Shikoku; c, Honshu; d, Hachijojima Island; e, Ogasawara (Bonin) Islands; f, Iojima Island; g, Izu-Ogasawara Ridge; h, Shatsky Rise. (b) Ground tracks of the TOPEX/POSEIDON (thick line) and ERS (thin line) altimeters. Thick broken line shows domain where mesoscale eddies are traced in SSH anomaly maps (see Subsection 3.1). 472 N. Ebuchi and K. Hanawa

3 Merging data from the two different altimeter missions, TOPEX/POSEIDON and ERS, provides better space-time sampling of oceanic features, especially for the mesoscale eddies which are the target of the present study. However, there is a gap in the ERS-1 data for a period from December 26, 1993 to March 31, 1995, when ERS-1 adopted different orbits for ice monitoring and geodetic missions. The MSLA products generated from only the TOPEX/POSEIDON data are used for this period. Although the maps produced with or without the ERS data may have different spatial resolutions, this has no great effect on the results of tracing the eddies, as presented in the following section. Following Ebuchi and Hanawa (2000), a recursivetype band-pass filter with cut-off periods of and 40.7 days and a cut-off gain of 45 db per octave was applied to the time series of the SSH anomaly at each grid point of the map, to extract signals associated with the mesoscale eddy activities. The cut-off periods mentioned above were determined based on a result of spectral analysis of the SSH anomaly (see figure 10 in Ebuchi and Hanawa, 2000). The study area is from 20 to 40 N and from 125 to 175 E, as shown in Fig. 1, excluding the marginal seas of the Japan and East China Seas. Bathymetry is shown in Fig. 1(a). Ground tracks of the TOPEX/POSEIDON and ERS altimeters are shown in Fig. 1(b). 3. Results and Discussion An example of eddy tracing using maps of filtered SSH anomaly fields is shown in Subsection 3.1. Eddy trajectories in the Kuroshio recirculation region are described in Subsection 3.2. The eddy path over the Izu-Ogasawara Ridge and propagation speed of eddies are also described in this section. Eddy generation and decaying areas, and eddy coalescence with the Kuroshio and resulting the path variation of the Kuroshio are discussed in Subsections 3.3 and 3.4, respectively. 3.1 An example of eddy tracing Figure 2 shows an example of a map of the filtered SSH anomaly fields. In the Kuroshio Extension region around 35 N, there is a train of high amplitude anomalies suggesting the meander path of the Kuroshio Extension. In the region along 30 N south of the Kuroshio and Kuroshio Extension, where the Kuroshio recirculation exists (Ebuchi and Hanawa, 1995, 2000), a train of cyclonic and anticyclonic eddies can be discerned. This train of cyclonic and anticyclonic eddies, being located by turns, extends further from the Kuroshio Extension region (around 150 E) to the Kuroshio region south of Kyushu (130 E). These eddies, which Ebuchi and Hanawa (2000) investigated along the Tokyo-Ogasawara line, are the target of the present study. Although there are several cyclonic and anticyclonic eddies along 20 N, they are considered as fluctuations of the subtropical front (Kawamura et al., 1995; Kobashi and Kawamura, 2001). Therefore, we eliminated these eddies travelling along the 20 N latitude from the present analysis. As reported by Ebuchi and Hanawa (2000), the diameter of the eddies is a few hundred kilometers, typically 250 km. An example of time series of the maps is shown in Fig. 3. For example, we can trace three eddies (labeled a, b, and c ), which are considered to be pinched off from the Kuroshio Extension. As time went by, these eddies moved westward along N, and coalesced with the Kuroshio in a region east of Kyushu. Consequently, these eddies have traveled over approximately Fig. 2. Example of the filtered map of the sea surface height (SSH) anomaly fields (May 9, 1997). Contour interval is 5 cm. Thick and thin lines represent positive and negative SSH anomalies, respectively. Thick broken line shows study domain where mesoscale eddies are traced in SSH anomaly maps. Trajectory of Mesoscale Eddies in the Kuroshio Recirculation Region 473

4 Fig. 3. Time series of filtered maps of SSH anomalies every 20 days from February 23, 1994 to May 29, Contour interval is 5 cm. Thick and thin lines represent positive and negative SSH anomalies, respectively km during a period of about 300 days from March 1994 through January The shape, size, and intensity of the eddies apparently seem to vary largely while they were travelling. However, it should be noted that the change represented not only variations of the eddies themselves, but was affected by the coarse spatial resolution (approximately 250 km) of the TOPEX/POSEIDON altimeter ground tracks. That is, when the eddy is situated between the ground tracks by chance, then anomaly amplitudes and shape tend to be reduced and modified. The maximum error in the determination of locations of the eddies can be estimated about 125 km from the spacing of ground tracks of the TOPEX/POSEIDON altimeter (approximately 250 km). A more detailed discussion of the errors in the interpolated SSH fields from observations of the TOPEX/POSEIDON altimeter was published by Kuragano and Shibata (1997). In the present paper we trace the trajectory of the eddies in the time series of the filtered SSH anomaly maps using following method. We traced only eddies that appeared in the region shown by a thick broken line in Figs. 1 and 2. In the Kuroshio and Kuroshio Extension regions it is sometimes difficult to recognize the eddies continuously in time, since the altimeters can provide only the SSH anomaly, and mean SSH field is not used to compose absolute SSH in the present study. However, the propagation of the eddies can be easily traced in regions away from the Kuroshio and Kuroshio Extension, as shown in Fig. 3. Therefore, we eliminated the regions of the Kuroshio and Kuroshio Extension. The northern boundary of the study domain is shifted southward in the eastern part in order to maintain sufficient distance from the mean Kuroshio Extension path which shifts southward in the region (Mizuno and White, 1983). We recognized only eddies which satisfy the following conditions at least every 40 days: (1) diameter of the contour line of 10 cm in the filtered SSH anomaly is greater than 100 km, and (2) approximately circular in shape. These parameters are determined from the results of the previous study (Ebuchi and Hanawa, 2000). As 474 N. Ebuchi and K. Hanawa

5 Fig. 4. Trajectories of (a) cyclonic and (b) anticyclonic eddies. The five-year time series of maps from October 22, 1992 to June 3, 1998, yielded tracings of 21 cyclonic and 22 anticyclonic eddies. mentioned above, the coarse spatial resolution of the SSH fields may reduce the amplitude and size of the eddies. Therefore, we adopted the time window of 40 days by considering the persistency of mesoscale eddies. Given the mean phase speed of eddies (6.8 cm s 1 ) determined by Ebuchi and Hanawa (2000), the time for the eddies to travel across the spacing of ground tracks of the TOPEX/ POSEIDON altimeter (250 km) is estimated as approximately 42 days. The centers of the eddies are determined by the local maximum or minimum of the SSH anomaly, and are traced remembering that the phase speed does not exceed 15 cm s 1. By using these criteria, the eddy labeled a in Fig. 3 could also be recognized and traced, even though it looks small and weak in some of the SSH anomaly maps (e.g., 14 May 1994). Note that the temporal interval of the SSH maps used in the present study is 10 days, while Fig. 3 shows the maps every 20 days. 3.2 Trajectories of eddies Figures 4(a) and (b) separately show the trajectories of cyclonic and anticyclonic eddies which are traced by using the method described in the previous section. We were able to trace 43 eddies (21 cyclonic and 22 anticyclonic eddies) from the five-year time series. It can be seen that both the cyclonic and anticyclonic eddies depart from a region south of the Kuroshio Extension, propagate westward across the Izu-Ogasawara Ridge, and reach the western edge of the basin southeast of Kyushu. No significant differences in the trajectory are discernible between the cyclonic and anticyclonic eddies. The trajectory of mesoscale eddies coincides with the typical path of the Kuroshio recirculation (Suga and Hanawa, 1995). Histograms of eddies which passed across the meridian of 135, 140, 145, and 150 E are calculated in order to examine effects of the bottom topography around the Izu-Ogasawara Ridge on the movement path of the mesoscale eddies. The result is shown in Fig. 5. To the east of the Izu-Ogasawara Ridge (around 150 E), the distribution of mesoscale eddies is almost flat over the meridian from 24 to 32 N. However, the eddies concentrate from 29 to 31 N around the Izu-Ogasawara Ridge (145 and 140 E). This latitude from 29 to 31 N just corresponds to a gap in the bottom topography between the Hachijojima Island and Ogasawara (Bonin) Islands (see Fig. 1). This result suggests that the trajectory of the mesoscale eddies is largely affected by the bottom topography and the eddies tend to pass through the gap. Trajectory of Mesoscale Eddies in the Kuroshio Recirculation Region 475

6 Fig. 5. Histograms of eddies which passed across the meridians of 135, 140, 145, and 150 E. Histograms of cyclonic (upper panel), anti-cyclonic (middle) and both types of eddies (lower). Fig. 6. Propagation velocity vectors of mesoscale eddies averaged in box of 1 1. Both cyclonic and anticyclonic eddies are treated together in the calculation. In addition, we tried to examine the effects of the bottom topography on the other properties of the eddies, such as their intensity and size, but could not obtain significant results. One possible reason is that the spatial resolution of the altimeter maps used in the present study is too coarse to capture the detailed features of the eddies. Since the zonal spacing of the TOPEX/POSEIDON altimeter ground tracks (approximately 250 km) is comparable to the horizontal scale of the eddies (Ebuchi and Hanawa, 2000), the observed intensity and size of an eddy may vary according as the altimeter observation was made across the center of the eddy or not, as mentioned before. Figure 6 shows the velocity vectors of the eddies averaged in grids of 1 1. Both cyclonic and anticyclonic eddies are treated together. Only vectors in grids where more than three eddies passed are shown in the figure. The typical standard deviation of the velocity components in these grids is 2 3 cm s 1. The mean westward 476 N. Ebuchi and K. Hanawa

7 propagation speed is estimated as about 7 cm s 1, which is much faster than the phase speed theoretically estimated for the baroclinic first-mode Rossby wave in this region, as discussed by Aoki et al. (1995) and Ebuchi and Hanawa (2000). The value of phase speed also agrees with the results of Kuragano and Kamachi (2000). This figure also clearly shows that eddies tend to move toward the gap in the Izu-Ogasawara Ridge in the east of the ridge. Ebuchi and Hanawa (2000) reported that the eddies have a robust subsurface structure, where displacement of the isotherms consisting the main thermocline exceeds 200 m (their figures 6 and 7). It is considered that the train of positive and negative SSH anomalies in this region represents not the simple propagation of SSH signals related to the Rossby waves, which can be observed elsewhere in mid latitudes with medium spatial scales (e.g., Aoki et al., 1995), but rather SSH signals associated with the nonlinear mesoscale eddies. Actually, the local Rossby number, which is defined as the ratio of orbital velocity of eddy to the phase speed, can be estimated to be greater than unity in the region near the Tokyo- Ogasawara line, when we use an orbital velocity of 15 cm s 1 and a phase speed of 7 cm s 1 (Ebuchi and Hanawa, 2000). The eddies traced in the present study over the Kuroshio recirculation region are also a few hundred kilometers in diameter and are greater than 15 cm in amplitude. A rough estimation of the mean surface slope from these values gives an orbital velocity of about 10 cm s 1, suggesting the local Rossby number might be larger than unity over the region. 3.3 Generation and disappearance areas of eddies Figure 7 shows locations where the eddies were first recognized in the study domain (a), and where the eddies disappeared from the maps of the SSH anomaly or moved out the domain (b). In the figure, cyclonic and anticyclonic eddies are indicated by different symbols. Some of the eddies shown in Fig. 4 do not appear in this figure, since these eddies were already in existence when the time series of SSH started. Figure 7(a) shows that eddies can be traced back to a region south of the Kuroshio Extension, suggesting that generation of the eddies is closely related to the Kuroshio Extension. There is no significant difference in the location of the first recognition between the cyclonic and anticyclonic eddies. As seen in Figs. 2 and 3, the adjacent area of the Kuroshio Extension is the most energetic region of variations in the North Pacific (Tai and White, 1990; Qiu et al., 1991; Aoki and Imawaki, 1996). It is easy to expect that the Kuroshio Extension may play a role in the generation of the eddies. However, it is difficult to determine precisely where eddies are detached from the Kuroshio Extension in the present analysis. The band-pass filtered Fig. 7. Locations where the eddies were first recognized in the study domain (a), and where eddies disappeared from maps of SSH anomaly or departed from the domain (b). Cyclonic and anticyclonic eddies are indicated by circles and triangles, respectively. SSH anomalies utilized in the present study are inadequate to investigate the detachment process of the eddies, which is the reason why we eliminated the Kuroshio Extension region from the present analysis. Absolute SSH fields may help us to observe meander of the Kuroshio Extension and generation of the eddies, as demonstrated by Ichikawa and Imawaki (1994). However, it is difficult to compose a precise mean SSH field with high spatial resolution to express the detailed structure of the meander of the Kuroshio Extension. Several studies have been conducted on the generation of cold and warm eddies in the Kuroshio Extension region (e.g., Kawai, 1972; Cheney, 1977; Mizuno and White, 1983; Yasuda et al., 1992; Ichikawa and Imawaki, 1994). However, only the detachment of cyclonic eddies (or cold rings) has been discussed with reference to the south of the Kuroshio Extension. The cyclonic eddies are considered to be generated by pinch-off from the meandering Kuroshio Extension. However, this process of pinch-off detachment cannot explain the generation of cyclonic eddies in the south of the Kuroshio Extension, which are observed as frequently and significantly as the anticyclonic eddies in the Kuroshio recirculation region (Ebuchi and Hanawa, 2000). Warm outbreaks from the Trajectory of Mesoscale Eddies in the Kuroshio Recirculation Region 477

8 Fig. 8. Locations of the Kuroshio axis, reproduced from Prompt Report on Oceanographic Condition (Hydrographic Department of the Japanese Maritime Safety Agency, 1995). western boundary current described by Cornillon et al. (1986) and Toba et al. (1991) might be one of possible mechanisms. Further studies of the physical mechanisms of generation of these eddies are needed to solve this problem. Figure 7(b) shows that most of eddies disappeared at the western edge of the basin south of Shikoku and east or south of Kyushu. As shown in Fig. 3, these eddies are considered to coalesce with the Kuroshio. There is also a cluster of points around the Izu-Ogasawara Ridge, suggesting some of the eddies decayed due to the effects of the bottom topography. 3.4 Coalescence of eddy with the Kuroshio and resulting path variation of the Kuroshio As mentioned in Subsection 3.1, around the end of January 1995, a large cyclonic eddy b coalesced with the Kuroshio in the east of Kyushu (see Fig. 3, the panels after 18 Feb. 1995). After the coalescence, a small meander in the Kuroshio started to propagate eastward from the south of Kyushu. Figure 8 is reproduced from Prompt Report on Oceanographic Condition (Hydrographic Department of the Japanese Maritime Safety Agency, 1995) to show locations of the Kuroshio axis in the sea south of Japan from January to May The small meander departed from the east of Kyushu at almost the same time as the coalescence, and grew in amplitude during the downstream propagation. This propagation of the meander is also discernible in the altimeter data (Fig. 3). In the period from October 1992 to June 1998, similar phenomena were observed eight times. This kind of eddy coalescence and resulting variations of the Kuroshio path and transport have been reported in previous studies. Yoshikawa et al. (1998) described the coalescence of mesoscale eddies with the Kuroshio at the Tokara Strait. Ichikawa (2001) found a correlation between SSH variation associated with mesoscale eddies and the surface transport of the Kuroshio at the Tokara Strait, and suggested that the coalescence of the eddies to the east of Taiwan affects the Kuroshio transport. Kobashi and Kawamura (2001) also reported that SSH variations with a period of 3 6 months related to the mesoscale eddies propagate along the Subtropical Front region in the western North Pacific (20 25 N) to the Kuroshio region along the southern coast of Japan. Similar to these studies, the result of the present study also suggests that the coalescence of mesoscale eddies may have a large effect on the Kuroshio path, although the physical mechanisms are not clear at present. A subsequent study on the relationship between the coalescence of the eddies and variations of the Kuroshio path south of Japan is now underway and results will soon be presented elsewhere. 478 N. Ebuchi and K. Hanawa

9 4. Summary and Concluding Remarks In the present study the trajectory of mesoscale eddies in the Kuroshio recirculation region was investigated by using the sea surface height anomaly observed by the TOPEX/POSEIDON and ERS altimeters. Cyclonic and anticyclonic eddies were traced on the time series of the maps of the SSH anomaly fields composed from the altimeter observations every ten days. It was found that both the cyclonic and anticyclonic eddies propagate westward in the Kuroshio recirculation region from a region south of the Kuroshio Extension. The west-east propagation speed of the eddies can be estimated as approximately 7 cm s 1, which is much faster than the phase speed theoretically estimated for the baroclinic first-mode Rossby wave. In the Izu-Ogasawara Ridge region, most of the eddies pass through the gap between Hachijojima Island and Ogasawara (Bonin) Islands, while some of the eddies decay around the Izu-Ogasawara Ridge. It is suggested that the trajectory of the eddies is crucially affected by the bottom topography. Some of the eddies coalesce into the Kuroshio in the region south of Shikoku and east of Kyushu. It was suggested that both the cyclonic and anticyclonic eddies are generated in the Kuroshio Extension region. However, the physical mechanism of the generation of the eddies still needs further investigations. In particular the generation of anticyclonic eddies in the south of the Kuroshio Extension is difficult to explain in terms of the pinch-off detachment from meanders. Dynamic models might help us to explore instabilities of the Kuroshio Extension and generation of the mesoscale eddies. The trajectory of the mesoscale eddies investigated in the present study coincides with the typical path of the Kuroshio recirculation. Suga and Hanawa (1995) showed that the subtropical mode water in the North Pacific (NPSTMW), which is characterized by a thermostad between the seasonal and main pycnoclines widely distributed in the northwestern subtropical gyre, is advected by the Kuroshio recirculation current. It might be hypothesized that the anticyclonic eddies, which has minimum vertical temperature gradient in the subsurface layer (figure 7 of Ebuchi and Hanawa, 2000), play a role in carrying NPSTMW. On the other hand, some cyclonic eddies might carry the North Pacific Intermediate Water (NPIW) and this process under the thermocline might explain the existence of NPIW with salinity less than that of surrounding NPIW (e.g., Kawai, 1980). Dedicated analyses of hydrographic observations together with altimetric data are necessary in order to verify these hypotheses. It was also suggested that the coalescence of the mesoscale eddies with the Kuroshio may affect the path variations of the Kuroshio in the sea south of Japan. A small meander departed from the south of Kyushu when a large cyclonic eddy coalesced, and propagated downstream. This observation is consistent with results of Ichikawa (2001) and Kobashi and Kawamura (2001), who suggested that the coalescence of mesoscale eddies affects the path variation and transport of the Kuroshio. Further studies are also needed to clarify the physical processes how the eddies influence the Kuroshio. Acknowledgements The Maps of Sea Level Anomalies (MSLA) altimeter products utilized in the present study have been produced and distributed by the Collecte Localisation Satellites (CLS), Space Oceanography Division, Toulouse, France, as a part of the European Union Environment and Climate Project AGORA (ENV4-CT ) and DUACS (ENV4-CT ) with financial support from the CEO (Centre for Earth Observation) programme and Midi-Pyrenees regional council. The CD-ROMs are produced by the AVISO/Altimetry operations centre. The ERS products were generated as a part of the proposal Joint analysis of ERS-1, ERS-2 and TOPEX/POSEIDON altimeter data for oceanic circulation studies selected in response to the Announcement of Opportunity for ERS- 1/2 by the European Space Agency (Proposal code: A02.F105). The present study was partially supported by Grants-in-Aid for an International Cooperative Research Project (project No ), and the NEAR-GOOS (North East Asian Regional-Global Ocean Observing System) Project (project No ), from the Ministry of Education, Science, Sports and Culture, Japan. References Aoki, S. and S. Imawaki (1996): Eddy activities of the surface layer in the western North Pacific detected by satellite altimeter and radiometer. J. Oceanogr., 52, Aoki, S., S. Imawaki and K. Ichikawa (1995): Baroclinic disturbances propagating westward in the Kuroshio Extension region as seen by a satellite altimeter and radiometers. J. Geophys. Res., 100, AVISO/Altimetry (1996): AVISO User handbook for Sea Level Anomaly altimeter products (edition 2.0), AVI-NT CN, CLS Space Oceanography Division, Toulouse, France. Cheney, R. E. (1977): Synoptic observations of the oceanic frontal system east of Japan. J. Geophys. Res., 82, Cheney, R. E., P. L. Richardson and K. Nagasaka (1980): Tracking a Kuroshio cold ring with a free-drifting surface buoy. Deep-Sea Res., 27, Cornillon, P., D. Evans and W. Large (1986): Warm outbreaks of the Gulf Stream into Sargasso Sea. J. Geophys. Res., 91, Ebuchi, N. and K. Hanawa (1995): Comparison of surface current variations observed by TOPEX altimeter with TOLEX- ADCP data. J. Oceanogr., 51, Ebuchi, N. and K. Hanawa (2000): Mesoscale eddies observed by TOLEX-ADCP and TOPEX/POSEIDON altimeter in the Kuroshio Recirculation region south of Japan. J. Oceanogr., Trajectory of Mesoscale Eddies in the Kuroshio Recirculation Region 479

10 56, Hydrographic Department of the Japanese Maritime Safety Agency (1995): Prompt Report on Oceanographic Condition, Nos Ichikawa, K. (2001): Variation of the Kuroshio in the Tokara Strait induced by meso-scale eddies. J. Oceanogr., 57, Ichikawa, K. and S. Imawaki (1994): Life history of a cyclonic ring detached from the Kuroshio Extension as seen by the Geosat altimeter. J. Geophys. Res., 99, 15,953 15,966. Kawai, H. (1972): Hydrography of the Kuroshio Extension. p In Kuroshio; Its Physical Aspects, ed. by H. Stommel and K. Yoshida, Univ. Tokyo. Press, Tokyo, Japan. Kawai, H. (1980): Rings south of the Kuroshio and their possible roles in transport of the intermediate salinity minimum and in formation of the skipjack and albacore fishing grounds. p In The Kuroshio IV: Proc. 4th CSK Symp., Tokyo, Japan, Feb Kawamura, H., Y. Sawa and F. Sakaida (1995): Satellite observations of 3 6 months variation in the Kuroshio and the Subtropical front. Umi to Sora, 71, 9 15 (in Japanese with English abstract). Kitano, K. (1975): Some properties of the warm eddies generated in the confluence zone of the Kuroshio and Oyashio Currents. J. Phys. Oceanogr., 5, Kobashi, F. and H. Kawamura (2001): Variations of sea surface heights with a period of 3 7 months in the subtropical gyre of the North Pacific. J. Geophys. Res. (submitted). Kuragano, T. and M. Kamachi (2000): Global statistical spacetime scales of oceanic variability estimated from the TOPEX/POSEIDON altimeter data. J. Geophys. Res., 105, Kuragano, T. and A. Shibata (1997): Sea surface dynamic height of the Pacific Ocean derived from TOPEX/POSEIDON altimeter data: Calculation method and accuracy. J. Oceanogr., 53, Le Traon, P. Y. and F. Ogor (1998): ERS-1/2 orbit improvement using TOPEX/POSEIDON: The 2 cm challenge. J. Geophys. Res., 103, Le Traon, P. Y., P. Gaspar, F. Bouyssel and H. Makhmara (1995): Using TOPEX/POSEIDON data to enhance ERS-1 data. J. Atmos. Ocean. Tech., 12, Le Traon, P. Y., F. Nadal and N. Ducet (1998): An improved mapping method of multi-satellite altimeter data. J. Atmos. Ocean. Tech., 25, Mizuno, K. and W. B. White (1983): Annual and interannual variability in the Kuroshio current system. J. Phys. Oceanogr., 13, Qiu, B., K. A. Kelly and T. M. Joyce (1991): Mean flow and variability in the Kuroshio Extension from Geosat altimetry data. J. Geophys. Res., 96, 18,491 18,507. Suga, T. and K. Hanawa (1995): The Subtropical Mode Water circulation in the North Pacific. J. Phys. Oceanogr., 25, Tai, C. K. and W. B. White (1990): Eddy variability in the Kuroshio Extension as revealed by Geosat altimetry; Energy propagation away form the jet, Reynolds stress, and seasonal cycle. J. Phys. Oceanogr., 20, 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, Yasuda, I., K. Okuda and M. Hirai (1992): Evolution of a Kuroshio warm-core ring: Variability of the hydrographic structure. Deep-Sea Res., 39, Suppl. 1, S131 S161. Yoshikawa, Y., M.-S. Kim and H. Mitsudera (1998): Mesoscale eddies and the Kuroshio current observed by TOPEX/ POSEIDON altimetric data. JAMSTECR, 33, JAMSTEC, Yokosuka, Japan, (in Japanese with English abstract). 480 N. Ebuchi and K. Hanawa