• An, H. S., 1977: A numerical experiment of the M2 tide in the Yellow Sea. J. Oceanogr. Soc. Japan, 33 , 103110.

  • Banas, N. S., , P. MacCready, , and B. M. Hickey, 2009: The Columbia River plume as cross-shelf exporter and along-coast barrier. Cont. Shelf Res., 29 , 292301.

    • Search Google Scholar
    • Export Citation
  • Bang, I. K., , and H. J. Lie, 1999: A numerical experiment on the dispersion of the Changjiang River plume. J. Korean Soc. Oceanogr., 34 , 185199.

    • Search Google Scholar
    • Export Citation
  • Batchelder, H. P., , C. A. Edwards, , and T. M. Powell, 2002: Individual-based models of copepod populations in coastal upwelling regions: Implications of physiologically and environmentally influenced diel vertical migration on demographic success and nearshore retention. Prog. Oceanogr., 53 , 307333.

    • Search Google Scholar
    • Export Citation
  • Beardsley, R. C., , R. Limeburner, , H. Yu, , and G. A. Cannon, 1985: Discharge of the Changjiang (Yangtze River) into the East China Sea. Cont. Shelf Res., 4 , (1–2). 5776.

    • Search Google Scholar
    • Export Citation
  • Blanton, J. O., , L. Y. Oey, , J. Amft, , and T. N. Lee, 1989: Advection of momentum and buoyancy in a coastal frontal zone. J. Phys. Oceanogr., 19 , 98115.

    • Search Google Scholar
    • Export Citation
  • Chang, P.-H., , and A. Isobe, 2003: A numerical study on the Changjiang diluted water in the Yellow and East China Seas. J. Geophys. Res., 108 , 3299. doi:10.1029/2002JC001749.

    • Search Google Scholar
    • Export Citation
  • Chen, C., , and R. C. Beardsley, 1995a: A numerical study of stratified tidal rectification over finite-amplitude banks. Part I: Symmetric banks. J. Phys. Oceanogr., 25 , 20902110.

    • Search Google Scholar
    • Export Citation
  • Chen, C., , and R. C. Beardsley, 1995b: A numerical study of stratified tidal rectification over finite-amplitude banks. Part II: Georges Bank. J. Phys. Oceanogr., 25 , 21112128.

    • Search Google Scholar
    • Export Citation
  • Chen, C., and Coauthors, 2008: Physical mechanisms for the offshore detachment of Changjiang Diluted Water in the East China Sea. J. Geophys. Res., 113 , C02002. doi:10.1029/2006JC003994.

    • Search Google Scholar
    • Export Citation
  • China Ocean Press, 1992: Marine Atlas of Bohai Sea, Yellow Sea and East China Sea: Hydrology. China Ocean Press, 524 pp.

  • Choi, B. H., 1990: Development of fine-grid numerical tidal models of the Yellow Sea and the East China Sea. J. Korean Soc. Coastal Ocean Eng., 4 , 233244.

    • Search Google Scholar
    • Export Citation
  • Choi, B. H., , K. O. Kim, , and H. M. Eum, 2002: Digital bathymetric and topographic data for neighboring seas of Korea (in Korean with English abstract). J. Korean Soc. Coastal Ocean Eng, 14 , 4150.

    • Search Google Scholar
    • Export Citation
  • Choi, B.-J., , and J. L. Wilkin, 2007: The effect of wind on the dispersal of the Hudson River plume. J. Phys. Oceanogr., 37 , 18781897.

    • Search Google Scholar
    • Export Citation
  • Csanady, G. T., 1974: Barotropic current over the continental shelf. J. Phys. Oceanogr., 4 , 357371.

  • Egbert, G. D., , and S. Y. Erofeeva, 2002: Efficient inverse modeling of barotropic ocean tides. J. Atmos. Oceanic Technol., 19 , 183204.

    • Search Google Scholar
    • Export Citation
  • Egbert, G. D., , A. F. Bennett, , and M. G. G. Foreman, 1994: TOPEX/Poseidon tides estimated using a global inverse model. J. Geophys. Res., 99 , 2482124852.

    • Search Google Scholar
    • Export Citation
  • Galperin, B., , L. Kantha, , S. Hassid, , and A. Rosati, 1988: A quasi-equilibrium turbulent energy model for geophysical flows. J. Atmos. Sci., 45 , 5562.

    • Search Google Scholar
    • Export Citation
  • Guo, X., , and T. Yanagi, 1998: Three-dimensional structure of tidal current in the East China Sea and the Yellow Sea. J. Oceanogr., 54 , 651668.

    • Search Google Scholar
    • Export Citation
  • Ichikawa, H., , and R. C. Beardsley, 2002: The current system in the Yellow and East China Seas. J. Oceanogr., 58 , 7792.

  • Kang, S. K., , S. R. Lee, , and H. J. Lie, 1998: Fine grid tidal modeling of the Yellow and East China Seas. Cont. Shelf Res., 18 , 739772.

    • Search Google Scholar
    • Export Citation
  • Kang, S. K., , M. G. G. Foreman, , H.-J. Lie, , J.-H. Lee, , J. Cherniawsky, , and K.-D. Yum, 2002: Two-layer tidal modeling of the Yellow and East China Seas with application to seasonal variability of the M2 tide. J. Geophys. Res., 107 , 3020. doi:10.1029/2001JC000838.

    • Search Google Scholar
    • Export Citation
  • Kantha, L. H., , I. Bang, , J.-K. Choi, , and M.-S. Suk, 1996: Shallow water tides in the seas around Korea. J. Korean Soc. Oceanogr., 31 , 123133.

    • Search Google Scholar
    • Export Citation
  • Katoh, O., , K. Morinaga, , and N. Nakagawa, 2000: Current distributions in the southern East China Sea in summer. J. Geophys. Res., 105 , (C4). 85658573.

    • Search Google Scholar
    • Export Citation
  • Kim, I.-O., , and H.-K. Rho, 1994: A study on China coastal water appeared in the neighbouring seas of Cheju Island. J. Korean Fish. Soc., 27 , 515528.

    • Search Google Scholar
    • Export Citation
  • Kim, K., , H. K. Rho, , and S. H. Lee, 1991: Water masses and circulation around Cheju-Do in summer. J. Oceanogr. Soc. Korea, 26 , 262277.

    • Search Google Scholar
    • Export Citation
  • Lee, H.-C., , A. Rosati, , and M. J. Spelman, 2006: Barotropic tidal mixing effects in a coupled climate model: Oceanic conditions in the northern Atlantic. Ocean Modell., 11 , 464477.

    • Search Google Scholar
    • Export Citation
  • Lee, S.-H., , and R. C. Beardsley, 1999: Influence of stratification on residual tidal currents in the Yellow Sea. J. Geophys. Res., 104 , (C7). 1567915701.

    • Search Google Scholar
    • Export Citation
  • Lefevre, F., , C. Le Provost, , and F. H. Lyard, 2000: How can we improve a global ocean tide model at a regional scale? A test on the Yellow Sea and the East China Sea. J. Geophys. Res., 105 , 87078725.

    • Search Google Scholar
    • Export Citation
  • Lie, H. J., , C. H. Cho, , J. H. Lee, , and S. Lee, 2003: Structure and eastward extension of the Changjiang River plume in the East China Sea. J. Geophys. Res., 108 , 3077. doi:10.1029/2001JC001194.

    • Search Google Scholar
    • Export Citation
  • Lien, R. C., , and M. C. Gregg, 2001: Observations of turbulence in a tidal beam and across a coastal ridge. J. Geophys. Res., 106 , 45754591.

    • Search Google Scholar
    • Export Citation
  • , X., , F. Qiao, , C. Xia, , J. Zhu, , and Y. Yuan, 2006: Upwelling off Yangtze River estuary in summer. J. Geophys. Res., 111 , C11S08. doi:10.1029/2005JC003250.

    • Search Google Scholar
    • Export Citation
  • Marchesiello, O., , J. C. McWilliams, , and A. Shchepetkin, 2001: Open boundary condition for long-term integration of regional oceanic models. Ocean Modell., 3 , 121.

    • Search Google Scholar
    • Export Citation
  • Mellor, G. L., , and T. Yamada, 1982: Development of a turbulence closure model for geophysical fluid problems. Rev. Geophys. Space Phys., 20 , 851875.

    • Search Google Scholar
    • Export Citation
  • Moon, J.-H., , N. Hirose, , J.-H. Yoon, , and I.-C. Pang, 2009: Effect of the along-strait wind on the volume transport through the Tsushima/Korea Strait in September. J. Oceanogr., 65 , 1729.

    • Search Google Scholar
    • Export Citation
  • Mourn, J. N., , D. R. Caldwell, , J. D. Nash, , and G. D. Gunderson, 2002: Observations of boundary mixing over the continental slope. J. Phys. Oceanogr., 32 , 21132130.

    • Search Google Scholar
    • Export Citation
  • Oey, L. Y., 1986: The formation and maintenance of density fronts on the U.S. southeastern continental shelf during winter. J. Phys. Oceanogr., 16 , 11211135.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., , and N. Imasato, 1990: A numerical study on the formation of the Kuroshio countercurrent and the Kuroshio branch current in the East China Sea. Cont. Shelf Res., 10 , 165184.

    • Search Google Scholar
    • Export Citation
  • Shchepetkin, A., , and J. C. McWilliams, 1998: Quasi-monotone advection schemes based on explicit locally adaptive dissipation. Mon. Wea. Rev., 126 , 15411580.

    • Search Google Scholar
    • Export Citation
  • Shchepetkin, A., , and J. C. McWilliams, 2005: The Regional Oceanic Modeling System: A split-explicit, free-surface, topography-following-coordinate ocean model. Ocean Modell., 9 , 347404.

    • Search Google Scholar
    • Export Citation
  • Simpson, J. H., , and J. R. Hunter, 1974: Fronts in the Irish Sea. Nature, 250 , 404406.

  • Song, Y., , and D. B. Haidvogel, 1994: A semi-implicit ocean circulation model using a generalized topography-following coordinate. J. Comput. Phys., 115 , 228244.

    • Search Google Scholar
    • Export Citation
  • Visser, A. W., 1997: Using random walk models to simulate the vertical distribution of particles in a turbulent water column. Mar. Ecol. Prog. Ser., 158 , 275281.

    • Search Google Scholar
    • Export Citation
  • Wensu, W., 1988: Yangtze brackish water plume—Circulation and diffusion. Prog. Oceanogr., 21 , 373385.

  • Wright, D. G., , and J. W. Loder, 1985: A depth-dependent study of the topographic rectification of tidal currents. Geophys. Astrophys. Fluid Dyn., 31 , 169220.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Horizontal distributions of surface salinity (a) in July 1986 obtained by Chen et al. (2008), (b) in August 1996 observed by National Fisheries Research and Development Institute (NFRDI, in South Korea), and (c) in July 1997 obtained by Lie et al. (2003).

  • View in gallery

    Model domain bottom topography (m). Rotated square box indicates the model domain. The solid line and dashed box are the location of the vertical section (CJ line) in section 3 and the area for the freshwater volume calculation (WJ area) in section 4, respectively.

  • View in gallery

    Distribution of daily mean surface current and salinity in the East China Sea for the initial condition (t = 0) after 30 days of spinup. The model was spun up with constant Changjiang river discharge of 60 000 m3 s−1, harmonic tidal forcing, and climatological fields for 30 model days.

  • View in gallery

    Horizontal distributions of daily mean surface salinity every 4 days from 12 to 32 model days after spinup. Contour interval is 1.0.

  • View in gallery

    Vertical distributions of daily mean salinity every 4 days from 12 to 32 model days after spinup. Contour interval is 1.0.

  • View in gallery

    Vertical sections of salinity distribution between the Changjiang mouth and Jeju Island obtained in August 1997 (data from Cheju National University). These sections are close to our model section CJ line of Fig. 2.

  • View in gallery

    (a) Time–space plot of the surface salinity along the S line of Fig. 4a and (b) the sea level variation at the center of the S line.

  • View in gallery

    (a) Trajectory of a particle released at the depth of 20 m and surface salinity distribution on day 16 after the spinup. Time series of (b) depth and (c) salinity for the particle trajectory. The particle was released at day 13 during spring tide. The dashed line indicates 30-, 40-, and 50-m isobaths.

  • View in gallery

    (a) Trajectory of a particle released at the depth of 20 m and surface salinity distribution on day 25 after spinup. Time series of (b) depth and (c) salinity for the particle trajectory. The particle was released at day 23 in neap tide. The dashed line indicates 30-, 40-, and 50-m isobaths.

  • View in gallery

    Vertical structure of mean (a),(d) salinity, (b),(e) vertical velocity w, and (c),(f) vertical eddy viscosity Km along the CJ line at (left) day 16 in spring and (right) day 24 during neap tide. Contour interval is 1.0 in salinity, 0.2 × 10−4 m s−1 in vertical velocity w, and 1.0 × 10−2 m2 s−1 in vertical eddy viscosity Km.

  • View in gallery

    (a) Sea surface temperature derived from a satellite image taken on 17 Jun 1992 (Lee and Beardsley 1999). (b) Model-predicted sea surface temperature at day 16 during spring tide: sea surface salinity (white contour line) is superimposed for comparison.

  • View in gallery

    Time–space plot of sea surface salinity along the S line of Fig. 4a simulated by the M2 tide.

  • View in gallery

    Vertical structure of mean (a),(d) salinity, (b),(e) vertical velocity w, and (c),(f) vertical eddy viscosity Km along the CJ line at day 16 during (left) spring tide and (right) the M2 tide. Contour interval is 1.0 in salinity, 0.2 × 10−4 m s−1 in vertical velocity w, and 1.0 × 10−2 m2 s−1 in vertical eddy viscosity Km.

  • View in gallery

    (a) Time–space plot of the surface salinity along the S line of Fig. 4a simulated for half-period of the spring–neap tide and (b) the sea level variation at the center of the S line.

  • View in gallery

    Response of the low-salinity water to wind magnitudes at model day 20, 28, and 36. Winds blow northwestward with magnitude (a)–(c) 0.02, (d)–(f) 0.035, and (g)–(i) 0.05 N m−2. Contour interval for salinity is 1.0.

  • View in gallery

    Time series of the freshwater volume across the WJ line in Fig. 2 with a northwestward wind of magnitude 0.02 (dotted line), 0.035 (solid line), and 0.05 N m−2 (dashed line).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 96 96 15
PDF Downloads 46 46 14

Offshore Detachment Process of the Low-Salinity Water around Changjiang Bank in the East China Sea

View More View Less
  • 1 Department of Earth System Science and Technology, Interdisciplinary Graduate School of Engineering Science, Kyushu University, Kasuga-Koen, Kasuga, Japan
  • | 2 Research Institute for Applied Mechanics, Kyushu University, Kasuga-Koen, Kasuga, Japan
  • | 3 Department of Oceanography, Cheju National University, Jeju, South Korea
© Get Permissions
Full access

Abstract

A patchlike structure of low-salinity water detached from the Chanjiang “Diluted Water” (CDW) is frequently observed in the East China Sea (ECS). In this study, the offshore detachment process of CDW into the ECS is examined using a three-dimensional numerical model. The model results show that low-salinity water is detached from the CDW plume by the intense tide-induced vertical mixing during the spring tide period when the tidal current becomes stronger. During the spring tide, thickness of the bottom mixed layer in the sloping bottom around Changjiang Bank reaches the mean water depth, implying that the stratification is completely destroyed in the entire water column. As a result, the offshore detachment of CDW occurs in the sloping side of the bank where the tidal energy dissipation is strong enough to overcome the buoyancy effect during this period. On the other hand, the surface stratification is retrieved during the neap tide period, because the tidal current becomes substantially weaker than that in the spring tide. Wind forcing over the ECS as well as tidal mixing is a critical factor for the detachment process because the surface wind primarily induces a northeastward CDW transport across the shelf region where tide-induced vertical mixing is strong. Moreover, the wind-enhanced cross-isobath transport of CDW causes a larger offshore low-salinity patch, indicating that the freshwater volume of the low-salinity patch closely depends on the wind magnitude.

* Current affiliation: Research Institute for Applied Mechanics, Kyushu University, Fukuoka, Japan

Corresponding author address: Jae-Hong Moon, Research Institute for Applied Mechanics, Kyushu University, Kasuga-Koen 6-1, Kasuga 816-8580, Japan. Email: jhmoon@riam.kyushu-u.ac.jp

Abstract

A patchlike structure of low-salinity water detached from the Chanjiang “Diluted Water” (CDW) is frequently observed in the East China Sea (ECS). In this study, the offshore detachment process of CDW into the ECS is examined using a three-dimensional numerical model. The model results show that low-salinity water is detached from the CDW plume by the intense tide-induced vertical mixing during the spring tide period when the tidal current becomes stronger. During the spring tide, thickness of the bottom mixed layer in the sloping bottom around Changjiang Bank reaches the mean water depth, implying that the stratification is completely destroyed in the entire water column. As a result, the offshore detachment of CDW occurs in the sloping side of the bank where the tidal energy dissipation is strong enough to overcome the buoyancy effect during this period. On the other hand, the surface stratification is retrieved during the neap tide period, because the tidal current becomes substantially weaker than that in the spring tide. Wind forcing over the ECS as well as tidal mixing is a critical factor for the detachment process because the surface wind primarily induces a northeastward CDW transport across the shelf region where tide-induced vertical mixing is strong. Moreover, the wind-enhanced cross-isobath transport of CDW causes a larger offshore low-salinity patch, indicating that the freshwater volume of the low-salinity patch closely depends on the wind magnitude.

* Current affiliation: Research Institute for Applied Mechanics, Kyushu University, Fukuoka, Japan

Corresponding author address: Jae-Hong Moon, Research Institute for Applied Mechanics, Kyushu University, Kasuga-Koen 6-1, Kasuga 816-8580, Japan. Email: jhmoon@riam.kyushu-u.ac.jp

1. Introduction

The water masses in the East China Sea (ECS) are characterized by various factors such as the Kuroshio water, river discharge, tidal mixing and atmospheric forcing (Chang and Isobe 2003). Among these factors, the effect of freshwater dominates the distributions of surface salinity in summer when the rivers runoff are high, especially associated with the Changjiang River contributing about 90% of the whole river discharge into the interior of the ECS (Beardsley et al. 1985; Kim et al. 1991; Chang and Isobe 2003).

Observations and numerical simulations have shown that the low-salinity water around northern ECS during summer originates from the Changjiang “Diluted Water” (CDW; Wensu 1988; Kim et al. 1991; Kim and Rho 1994) and that wind forcing significantly affects the northeastward extension of CDW as it is transported by wind-induced Ekman flow (Bang and Lie 1999; Chang and Isobe 2003). On the other hand, the observations frequently show a patchlike structure of the low-salinity water detached from the CDW plume as shown in Fig. 1. An isolated low-salinity patch appears between the Changjiang river mouth and Jeju Island and mainly moves northeastward to Jeju Island. Lie et al. (2003) suggested that under upwelling-favorable southerly winds, the CDW patches separate from the shallow shelf area and then are advected offshore toward Jeju Island as a series of low-salinity patches, based on summer observations.

Oey (1986) and Blanton et al. (1989) found that an isolated low-salinity patch in the inner-shelf region of the South Atlantic Bight could form at the outer edge of the front as a result of wind-induced upwelling. Their results are fairly consistent with simple wind-induced Ekman theory proposed by Csanady (1974). He suggested that an upwelling-favorable wind tends to produce an offshore water transport near the surface and an onshore water transport near the bottom.

In the ECS, a few modeling studies explored the northeastward extension of CDW (Bang and Lie 1999; Chang and Isobe 2003) showed the wind-induced northeastward transport; however, the patch structure of low-salinity water between the Changjiang River mouth and Jeju Island was not reproduced in their models. Recently, Chen et al. (2008) emphasized that the horizontal resolution used in these models was too coarse; thus, the models were unable to simulate the mesoscale structure of CDW. They suggested that the low-salinity patch can be formed by the detachment of anticyclonic eddies as a result of baroclinic instability in the CDW front, based on a high-resolution, unstructured-grid coastal ocean model.

However, another possibility for the detachment of CDW might be related with tidal effects. The tides in the Yellow Sea (YS) and ECS are well known to have large tidal ranges and produce strong tidal currents (Choi 1990; Lee and Beardsley 1999). Tidal mixing is one of the factors forming the characteristic of water masses in the ECS due to vertical mixing, especially at the shallow shelf regions. Generally, barotropic tidal energy is concentrated in coastal regions with energy greater by several orders of magnitude than in deep regions, and the dissipation of tidal energy induces vertical mixing in the coastal region and continental slope region (Lien and Gregg 2001; Mourn et al. 2002; Lee et al. 2006). This tidal mixing is important in stirring river plumes, and some simulations without tidal forcing frequently show that the river runoff flow forms an unrealistically thin and fresh surface plume (Lee et al. 2006; Choi and Wilkin 2007).

Various numerical calculations have been computed to investigate tides in the YS and ECS using both two- or three-dimensional finite difference (An 1977; Choi 1990; Kang et al. 1998; Kantha et al. 1996) and finite element (Lefevre et al. 2000) models. However, most studies focused on barotropic features in this area. Even the three-dimensional models were generally computed with homogeneous density.

Chen and Beardsley (1995a) studied tidal rectification processes in a stratified fluid over an idealized geometry of Georges Bank. When initial stratification is included, tidal currents become modified over the sloping sides of the bank by nonlinear advection, the baroclinic pressure gradient, and vertical friction. Internal waves at tidal and higher frequencies are generated over the sloping sides of the bank, and tidal mixing occurs in the bottom boundary layer, which leads to horizontal tidal mixing fronts near the shelf break. In the ECS, Kang et al. (2002) examined the barotropic and baroclinic responses of tides and tidal currents using a two-layer numerical model, and the model results showed that relatively large current shears appeared around the shelf slope off the Changjiang River. For summertime stratification, Lee and Beardsley (1999) found that the vertical mixing by the M2 tide occurs in the bottom boundary layer over the sloping bottom around Changjiang Bank. They also suggested that the bottom mixed layer thickness varies with the spring–neap cycle in the tidal currents, and the water column over Changjiang Bank could be well mixed during spring tide when tidal currents becomes stronger. Moreover, the shelf slope region off the Changjiang is consistent with a strong upwelling area, as observed by Lü et al. (2006), and they suggested that tidal mixing plays a predominant role in inducing the upwelling.

In view of these effects of tidal forcing, we expect that the intense tidal mixing over the sloping bottom around Changjiang Bank can affect the offshore detachment of the CDW as the CDW is well mixed with deeper saline water in the slope region. In this paper, we use a three-dimensional ocean circulation model to develop insight on how the offshore detachment of CDW responds to some external forces. The numerical model setup and conditions are described in the next section. The result of simulating the offshore detachment process of the low-salinity water is represented in section 3. Responses of the detachment of CDW to tidal and wind forcing are discussed in section 4. Finally, interpretation of the numerical simulations is summarized in section 5.

2. Numerical model and setup

The numerical model used in this study is the Regional Ocean Modeling System (ROMS). ROMS is a three-dimensional, free surface, hydrostatic, primitive equation numerical ocean model based on the nonlinear terrain-following coordinate (s coordinate) of Song and Haidvogel (1994). In ROMS the advection operators are third order and upstream biased, designed to reduce dispersive errors and the excessive dissipation rates needed to maintain smoothness, thereby effectively enhancing resolution on a given grid. Details of the ROMS computational algorithms are suggested by Shchepetkin and McWilliams (1998, 2005). The level 2.5 scheme of vertical turbulent mixing closure was chosen for this study to provide a realistic parameterization of vertical mixing and a free surface to simulate long gravity wave propagation (Mellor and Yamada 1982; Galperin et al. 1988). This turbulent closure considers the energetics of the mixing explicitly by solving prognostic equations for turbulent kinetic energy and length scale. The horizontal advection scheme in this study is third order with implicit diffusion, which has a velocity-dependent, hyperdiffusion dissipation as the dominant truncation error (Shchepetkin and McWilliams 1998); therefore, no explicit diffusivities or viscosities are needed to maintain stable solutions.

The model domain is a rotated rectangle with a wall on the northern side and three open boundaries including the ECS, as shown in Fig. 2. The model horizontal resolution is about with 20 levels in the vertical stretched terrain-following coordinate. The bottom topography is extracted from a combination of two topographic datasets, the 5-minute gridded elevations/bathymetry for the world (ETOPO5; National Geophysical Data Center) and the Sung Kyun Kwan University (SKKU) 1-min digital bathymetric and topographic data (Choi et al. 2002). Bottom stress is parameterized with a quadratic drag law using a drag coefficient of 2.5 × 10−3 (Lee and Beardsley 1999).

For the lateral boundary conditions, the monthly mean temperature, salinity, and velocity of the northwestern Pacific Ocean model (Moon et al. 2009) are used in this model. This climatology is used at open boundaries for all prognostic variables, following the method described in Marchesiello et al. (2001), and tidal harmonic forcing (10 constituents) from the Ocean Topography Experiment (TOPEX)/Poseidon global tidal model (TPXO6) analysis (Egbert et al. 1994; Egbert and Erofeeva 2002) are applied as surface height and depth-averaged velocity boundary conditions. From the northwestern Pacific model, the seasonal volume transport through Taiwan Strait is set to 0.8 Sv (1 Sv ≡ 106 m3 s−1) in winter and 2.7 Sv in summer, and the outflow transport through Korea/Tsushima Strait is set to 2.3 Sv in winter and 3.2 Sv in summer. The model includes only one river source, the Changjiang, which is the largest river in Asia (Chang and Isobe 2003), and the volume flux (at zero salinity) enters through the model cells of river outflow.

The model initialization is with the temperature and salinity for the month of June, obtained from a northwestern Pacific model with ⅙° horizontal resolution (Moon et al. 2009). To reproduce a realistic freshwater bulge near the river mouth in June, the model was spun up with constant Changjiang River discharge of 60 000 m3 s−1, tidal forcing, and the June climatology wind fields [Comprehensive Ocean–Atmosphere Data Set (COADS)] for 30 model days. The discharge is a typical summertime value (Chen et al. 2008). After 30 days of spinup, the model was run for an additional 60 days with a constant uniform northwestward wind stress (magnitude of 0.035 N m−2) as an experimental study.

To validate the model-generated tides, we performed harmonic analysis for some major components and then compared it with the observed tidal elevations and tidal currents taken from Choi (1990), Guo and Yanagi (1998), and Lee and Beardsley (1999). The comparison showed good agreement (not shown). For example, in the M2 tide current the mean differences and standard deviations for the u component are only (−2.3 ± 6.1) × 10−2 m s−1, and for υ component are (−2.7 ± 6.5) × 10−2 m s−1 at 16 stations. In the S2 tide current, the values for u and υ components are (−0.21 ± 2.8) × 10−2 and (−0.26 ± 3.3) × 10−2 m s−1, respectively.

3. Results

a. Adjustment of the ECS circulation

For the initial condition (t = 0) after spinup, the ECS circulation, including the intrusion of the Kuroshio east of Taiwan, the Taiwan Warm Current (TWC), and the Tsushima Warm Current, reaches a quasi steady state (Fig. 3). The Kuroshio, entering the ECS east of Taiwan with a mean speed of ∼1 m s−1, flows northeastward along the continental shelf break of the ECS. The TWC flows northeastward parallel to the 50-m isobath, reaches the Kuroshio frontal area (around 30°N), and then deflects northward to the southern sea of Korea and Korea/Tsushima Strait. As the TWC enters a submerged river valley off the Changjiang and then turns anticyclonically along the 50-m isobath, the current width (speed) increases (decreases) in the central part of the ECS. The simulated current pattern of ECS coincides well with the observational current distribution (Qiu and Imasato 1990; Katoh et al. 2000; Ichikawa and Beardsley 2002). The CDW flows out from the Changjiang, turning anticyclonically in the river mouth area shallower than 50 m before flowing southward along the Chinese coast. A tongue-shaped CDW plume extending from the river is quite similar to that of the observed distribution in June (China Ocean Press 1992).

b. Detachment of CDW around Changjiang Bank

Figure 4 shows the horizontal distributions of daily mean salinity at the surface from 12 to 32 model days after spinup. By model day 12 (Fig. 4a), low-salinity water less than 30 psu extends northeastward from the Changjinag mouth as a tongue driven by the spatially uniform northwestward wind. The extended low-salinity water detaches from the main body of the river plume from day 16 to day 20 (Figs. 4b, 4c), and the detached low-salinity patch with spatial size of 80–150 km moves northeastward to Jeju Strait by wind-induced Ekman flow. The distribution of low-salinity patch approaching Jeju Strait is similar to the observed distribution in Fig. 1b (result of the hydrographic survey in early August 1996).

The offshore detachment of CDW occurs in the region around 32.5°N, 123.5°E as relatively high salinity water intrudes into the shoreward side of the outer edge of the CDW plume. During days 24–28 the surface saline water is displaced by low-salinity water advected from main body of the river plume (Figs. 4d, 4e). Thereafter, the low-salinity water starts to detach again in the same region at day 32 (Fig. 4f).

Figure 5 represents the vertical profiles of salinity along the CJ line in Fig. 2 during the same period in Fig. 4. During 12 days the CDW is mainly distributed from the river mouth to Changjiang Bank as a 10–15-m thick plume in the surface layer; thereafter, it spreads farther offshore from the river mouth as a wind-induced Ekman transport. When the surface current transports freshwater from the Changjiang mouth, the low-salinity water detaches from the main body of the CDW plume along the sloping side of Changjiang Bank (near 123.5°E and depth of 30–50 m in Fig. 5) from day 16 to 20. The water column in the bottom slope region shows weak stratification, indicating that the CDW is well mixed with deeper saline water by strong vertical mixing. The well-mixed water becomes stratified from day 24 to 28 because the surface water is displaced by low-salinity water advected from the river plume (Figs. 5d, 5e). Thereafter, the low-salinity water separates again from the CDW plume in the bottom slope region at day 32 (Fig. 5f).

Vertical sections of the salinity distribution between the Changjiang mouth and Jeju Island were obtained in August 1997 (data from Cheju National University), which is close to our model section CJ line (Fig. 6). A strong halocline is developed 10 m below the surface along the sections. The halocline is weakened in the sloping bottom region where isohaline spreads vertically toward the surface and the bottom, indicating formation of a surface-to-bottom front over the sloping bottom (Lee and Beardsley 1999). Lie et al. (2003) also showed such a well-mixed water structure between the river mouth and Jeju Island. As the water column over the slope region is nearly homogeneous, the surface low-salinity water is separated from the main CDW plume, suggesting that strong tidal mixing occurs there. This structure of salinity is fairly consistent with the model results.

In our experiment, the offshore detachment of CDW occurs periodically over a two–week cycle. A time–space plot of the surface salinity along the S line in Fig. 4a was shown in Fig. 7a to identify the variability of its offshore detachment. The sea level variation at the center of the S line was also shown in Fig. 7b to compare with the period of tides. The low-salinity patches less than 26 psu show a variation with about a two-week period. Variation of its offshore detachment is consistent with the spring–neap tide cycle in this area. During spring tide relatively high salinity water intrudes into the CDW plume and then the low-salinity water is separated from the main CDW plume, while the surface saline water is displaced again by the low-salinity water advected from the river plume during neap tide. The spatial scale of low-salinity patches tends to be large as the amplitude of the neap tide becomes smaller. Modeled response suggests that the offshore detachment of CDW is significantly related to the tide-induced vertical mixing over the sloping bottom around Changjiang Bank, which varies with the tidal current during the spring–neap cycle.

We conducted an additional experiment with a northward wind as a comparative study. However, the result with northward wind was not shown in this paper because the spatial distribution and variations of CDW in the northward wind case were very similar to that found in the case with the northwestward wind, although the low-salinity water in the ECS shifted a little southeastward. Chen et al. (2008) also showed a similarity between the results with northwestward and northward winds.

c. Particle-tracking experiments

To better visualize the vertical mixing process for formation of the low-salinity patch, a particle tracking experiment is conducted in this study; particles are tracked in the process that integrates the 3D velocity fields from ROMS. A fourth-order Runge–Kutta scheme is used for the integration with a model time step. Particles in ROMS are subject to vertical diffusion using the “random displacement” or “random walk” scheme in addition to advection in all three dimensions (Banas et al. 2009). This scheme, suggested by Visser (1997) and Batchelder et al. (2002), adds a random vertical velocity scaled by the local vertical diffusivity from ROMS, to advective velocity at every time step. For an individual particle, a change in vertical position from zn to zn+1 over a time step Δt is given by (Banas et al. 2009)
i1520-0485-40-5-1035-e1
where wadv is the advective velocity, R a normally distributed random function with mean 0 and variance 1, and k the vertical diffusivity. Details of a particle in the ROMS computation are described by Banas et al. (2009). Two particles here are launched near 32.25°N, 123.5°E in water depth shallower than 40 m, where the low-salinity water detaches from the CDW plume. One particle (P1 in Fig. 8a) is released at model day 13 during spring tide, while the other (P2 in Fig. 9a) is released at model day 23 during neap tide. Both particles are released at the depth of 20 m.

The particles follow the tidal excursion ellipse with a mean cross-isobath (northeastward) movement as shown in Figs. 8a and 9a. Depth and salinity values of the two particles are also represented in time, Figs. 8b, 8c and 9b, 9c, respectively. Two days after particle P1 was released during spring tide, the particle tends to oscillate up and down between the surface and near bottom. After 2 days when the low-salinity water starts to detach from the main freshwater stream, the particle oscillates within the surface layer from 0 to 5 m as the particle is trapped within the surface low-salinity patch. Thereafter, the particle quickly moves northeastward to Jeju Island. During the same periods, the salinity of particle P1 varies with depth of the particle. When the particle upwells toward (downwells from) the surface, the salinity tends to become lower (higher). On the other hand, the particle released during neap tide exhibits a small oscillating trajectory below 10-m depth during the 4 days from when the particle was released and then upwells toward the surface with a tidally filtered path toward the northeastward. This result shows that vertical motion of the particles was substantially increased along the sloping bottom during spring tide compared with that during neap tide.

The particle-tracking experiment suggests that the low-salinity water can be detached from the CDW plume as a result of vigorous vertical mixing over the sloping bottom when tidal energy becomes greater during spring tide. This process of vertical mixing has been missing in previous modeling studies owing to the neglect of tidal forcing.

d. Vertical mixing process of CDW

The vertical structure and magnitude of a tidally rectified flow critically depends on the magnitude of tidal current, friction, and stratification. The magnitude of residual flow tends to increase as the bottom slope becomes steeper and the tidal current becomes stronger (Chen and Beardsley 1995a,b). The friction associated with turbulent mixing in the bottom boundary layer can cause vertical shear in the tidal current, which in turn leads to vertically nonuniform momentum advection for a tidal flow (Wright and Loder 1985). Moreover, Chen and Beardsley (1995a,b) emphasized that, when stratification is added, the tidal mixing occurs in the bottom boundary layer, leading to horizontal tidal mixing fronts. The thickness of a tidally mixed bottom layer is proportional to the magnitude of the tidal current and inversely to stratification, using a simple energy argument. Lee and Beardsley (1999) applied this interpretation of Chen and Beardsley (1995a) to the YS and ECS and showed that tidal mixing with stratification occurs in the bottom boundary layer over a sloping bottom, which, here, leads to surface tidal fronts around Changjiang Bank.

Figure 10 shows the simulated vertical structure of mean salinity, vertical velocity w, and vertical eddy viscosity Km along the CJ line during the spring (Figs. 10a–c) and neap tides (Figs. 10d–f). The salinity section for the spring tide shows that the halocline is completely destroyed in the slope region around Changjiang Bank, resulting in a locally homogeneous water column (Fig. 10a). As a result, the low-salinity water at the surface is detached from the main river plume at the tidal mixing front. The vertical velocity shows a strong cross-isobath double-cell circulation pattern on the sloping side of Changjiang Bank where the water tends to be upwelled to the surface along the bottom slope and then downwelled on both sides of the front (Fig. 10b). The maximum vertical velocity is about 1.0 × 10−4 m s−1 in the upwelling region. During the neap tide the halocline becomes relatively stronger than that of the spring tide, and the offshore detachment of CDW is not found (Fig. 10d). Vertical motions are relatively weak on the bottom slope (Fig. 10e).

The model-predicted Km for the spring tide has a parabolic structure in the vertical, which is confined within the bottom mixed boundary layer over the sloping bottom and quickly decreases in magnitude upward from there. In particular, the parabolic vertical shape appears within the entire water column above the sloping bottom, implying that the water salinity could be well mixed (Fig. 10c). However, the magnitude of Km and the thickness of the bottom mixed boundary layer for the neap tide are significantly reduced over the sloping bottom around Changjiang Bank (Fig. 10f). The bottom boundary layer for the spring tide reaches the surface above the bottom slope around Changjiang Bank. This indicates that the entire water column is well mixed and, thus, the CDW is detached from the main freshwater plume. In contrast, the halocline becomes stronger for the neap tide because the bottom boundary layer cannot reach the surface. It is considered that the tide-induced vertical mixing is significantly weaker than for the spring tide.

Simpson and Hunter (1974) found that vertical tidal mixing is predominantly controlled by surface buoyancy flux and tidal energy dissipation, ignoring the effects of wind mixing, horizontal friction and advection, and freshwater input from the rainfall and river discharge. When tidal energy dissipation is strong enough to overcome the buoyancy input, the water will be vertically well mixed. Otherwise, the water will remain stratified. With initial linear stratification, Chen and Beardsley (1995a) showed that the vertically mixed bottom layer thickness hm is written
i1520-0485-40-5-1035-e2
where γ is the bottom friction coefficient, usually taken as 2.5 × 10−3; δ is the efficiency of tidal kinetic energy dissipation used over the period ΔT to produce the potential energy needed for vertical mixing, the value of which is 3.7 × 10−3, suggested by Simpson and Hunter (1974); U is a typical tidal velocity; and N the Brunt–Väisälä frequency. This estimation, in brief, implies that the vertical thickness of a tidally mixed bottom layer is proportional to the magnitude of the tidal current and inversely proportional to stratification.

Lee and Beardsley (1999) checked the model result of YS and ECS for M2 tidal mixing using the above simple energy argument. They estimated that the thickness of the bottom mixed boundary layer can reach ∼30 m above the bottom slope around Changjiang Bank, ignoring freshwater input from the Changjiang River discharge. In our experiment, Eq. (2) with U = 0.8 m s−1 as a tidal current for the spring tide–predicted mixed layer with a thickness of 36 m above the bottom slope around Changjiang Bank. Although the surface buoyancy flux due to freshwater discharge from the river is included in our model, the predicted mixed layer is larger than that of Lee and Beardsley (1999), because the tidal current becomes stronger for the spring tide than that of the M2 tide. The thickness of mixed layer is close to the mean water depth over Changjiang Bank, implying that the entire water column above the sloping bottom is well mixed for the spring tide. On the other hand, the thickness of mixed layer is significantly reduced for the neap tide because the tidal current becomes substantially weaker.

The satellite image of sea surface temperature taken on 17 June 1992 (Fig. 11a, obtained from Lee and Beardsley 1999) shows that relatively cold water forms a front over the shelf slope around Changjiang Bank. The satellite image, taken during a spring tide, indicates that the lunar age on 17 June 1992 is 17 days. Figure 11b shows the model-predicted sea surface temperature for the spring tide: sea surface salinity (white contour line) is superimposed for comparison. The modeled sea surface temperature shows relatively cold surface water around Changjiang Bank: a sharp surface front with this cold water formed along the bottom slope around Changjiang Bank. The occurrence of the surface cold water region coincides with the SST image in Fig. 11a. Note that the surface front position over the Changjiang Bank is quite consistent with the region where the CDW detaches from the main body of the river plume (Fig. 11b). The result suggests that intense tidal mixing during spring tide can produce relatively high salinity surface water with increased thickness of the bottom mixed boundary layer. As a result, the offshore detachment of CDW occurs at the surface tidal front around Changjiang Bank; therefore, the low-salinity patch is formed.

The model horizontal resolution used in this study may not be sufficient to resolve eddy formation via baroclinic instability associated with detailed channel bathymetry near the river mouth, as pointed out by Chen et al. (2008). However, here we focused on the detachment of cross-isobath CDW plume between the river mouth and Jeju Island in tide-induced vertical mixing as a possible mechanism, which is different from Chen et al. (2008). Lee and Beardsley (1999) emphasized that a well-mixed water structure above the shelf slope off the river mouth observed in summer could be formed by tide-induced vertical mixing, based on three-dimensional tidal model with its horizontal resolution of ⅛°. In view of this effect of tidal forcing, our model’s horizontal resolution is sufficient to resolve the sloping bottom between the river mouth and Jeju Island; therefore, it is likely adequate for describing the formation of detached low-salinity feature of the tidal mixing.

4. Discussion

a. Response to tides

To examine the response of the detachment of CDW to the tides, numerical experiments were conducted for two cases of tide: 1) the M2 semidiurnal tide and 2) a half-period of the fortnightly spring–neap tide. In the second case, the period of each tidal constituent is shortened to half of its original period while maintaining the original amplitude, implying that the spring–neap modulation has a one-week cycle. Figure 12 shows the space–time plot of the surface salinity predicted by the M2 tide along the S line (Fig. 4a). The low-salinity water patch less than 26 psu is not shown in contrast to the spring–neap tide case, although the M2 tide is known to have a large amplitude and produce a strong tidal current in the shelf region of the ECS.

In comparison with the vertical profiles at spring tide (Fig. 13), strong upper-layer stratification associated with the river discharge is not destroyed by bottom mixing by the M2 tide on the bottom slope around Changjiang Bank. When we take U = 0.6 m s−1 as the M2 tidal current, hm in Eq. (2) becomes ∼27 m, which is shallower than the mean water depth over the sloping bottom, indicating that the bottom boundary mixed layer cannot reach the surface. This predicted bottom mixed layer thickness agrees well with the model mixed water column of ∼23 m for the M2 tide case. This estimation implies that the water remains stratified on the slope and low-salinity water at the upper layer cannot detach from the main body of CDW plume.

Figure 14 shows the time–space plot of the surface salinity predicted by the half-period of the spring–neap tide along the S line (Fig. 4a). The distribution of time–space surface salinity shows that the variation of the detachment of CDW is consistent with the half-cycle of the spring–neap tide. In this case, the spring–neap cycle shows a week period with a maximum tidal current of 0.8 m s−1 for the spring tide. This tidal current magnitude is adequate to support the tidal mixing mechanism as estimated above; therefore, the detached low-salinity feature varies with a half-period of the spring–neap tide. The results indicate that strong vertical mixing during spring tide, when the strong tidal current flows over the sloping bottom, causes the offshore detachment of CDW at the tidal front region and then forms a low-salinity patch between Changjiang Bank and Jeju Isalnd. It is, therefore, reasonable to consider that the offshore detachment of CDW can occur during spring tide when the tidal current is strongest.

b. Response to wind magnitudes

To examine how the response of the low-salinity patches differs with wind magnitudes, a northwestward wind stress is applied with magnitude 0.02 and 0.05 N m−2 for 60 days after spinup time. These wind stresses correspond to a speed of 3 and 7 m s−1, respectively. The river discharge is kept at 60 000 m3 s−1. The result with annual mean discharge of 28 000 m3 s−1 showed that the formation process and spatial pattern of CDW patches were not different from the high-river discharge case, although it shows a linear response to the amount of freshwater. This is one indication that the cross-isobath movement of CDW is primarily driven by the summer wind rather than the river discharge (Bang and Lie 1999; Chang and Isobe 2003).

For the 0.02 N m−2 weak northwestward wind, freshwater accumulates over Changjiang Bank, then begins to extend eastward along the TWC, which flows northeastward parallel to the 50-m isobath, enters a submerged river valley off Changjiang, and then moves to the interior of the ECS continental shelf along the 50-m isobath (Figs. 15a–c). Although small-scale low salinity patches are formed at the tidal-induced frontal region, the main freshwater stream moves eastward along the TWC. For 0.035 N m−2 the surface water is pushed northeastward off of Changjiang Bank and flows to Jeju Island across isobaths (Figs. 15d–f). The cross-isobath plume is clearly detached from the main body of CDW and forms a low-salinity water patch offshore. When the wind is increased to 0.05 N m−2, the CDW is quickly driven farther northeastward off of Changjiang Bank and clearly detached at the surface tidal front (Figs. 15g–i). The size of the low-salinity patches detached from the CDW plume is larger than that of weak wind cases because the wind-induced Ekman transport causes a larger offshore transport of CDW off Changjiang Bank.

To quantify and more directly compare the effect of wind magnitudes on the offshore transport, freshwater volume Vf in the WJ area of Fig. 2 is estimated by
i1520-0485-40-5-1035-e3
where s0 is a reference salinity and s is the salinity of the water column. For s0 we used the maximum value of modeled salinity in the region, which ensures that freshwater thickness is always positive (Choi and Wilkin 2007).

Time series of the freshwater volume in the WJ area with various wind magnitudes are plotted in Fig. 16. When the northwestward wind becomes stronger (weaker), the freshwater volume increases (decreases), indicating that the amount of freshwater varies with wind strength. This result shows that a stronger northwestward wind tends to enhance northeastward transport of CDW across the isobath as suggested in previous model studies (Bang and Lie 1999; Chang and Isobe 2003; Chen et al. 2008), and the wind-induced cross-isobath transport causes a larger offshore low-salinity patch of CDW.

Our experiments suggest that the offshore detachment of CDW can be caused by intense vertical mixing during spring tide, when tidal energy dissipation over the sloping bottom is strong enough to overcome the buoyancy effect. As mentioned above, however, the CDW around Changjiang Bank mostly flows eastward along the TWC, parallel to the 50-m isobath, when a weak northwestward wind blows (Figs. 15a–c). It indicates that the wind forcing plays a critical role in determining the cross-isobath movement of CDW. The tide-induced vertical mixing over the shelf slope can be important when the CDW is transported northeastward across the isobath by the northwestward wind. As a result, this process of its offshore detachment can be significantly intensified by the northwestward wind, with transport increasing with wind speed.

5. Summary

In this study, the mechanism for the offshore detachment of CDW into the ECS is examined using a three-dimensional numerical model. The model was initialized with temperature and salinity for the month of June obtained from the northwestern Pacific model (Moon et al. 2009), and then spun up with the climatological wind field (COADS), tidal harmonic forcing, and steady river discharge close to typical summer value (60 000 m3 s−1). After the spinup time, the model was run with a constant uniform northwestward wind stress (magnitude 0.035 N m−2).

The simulated CDW clearly shows a patch structure of low-salinity water detached from the CDW plume between the Changjiang and Jeju Island. The offshore detachment of CDW into the ECS occurs above the sloping bottom around Changjiang Bank as relatively high-salinity water intrudes into the shoreward side of the outer edge of the CDW plume. Thereafter, the surface saline water is displaced again by low-salinity water advected from the river plume. Such a variation of the detachment is consistent with the spring–neap cycle in this area. During spring tide, strong tidal current flows over the sloping bottom and the thickness of bottom mixed layer is close to the mean water depth over Changjiang Bank, implying that the entire water column above the sloping bottom is well mixed. On the other hand, the mixed layer thickness is significantly reduced during neap tide because the tidal current becomes substantially weaker. This result indicates that the intense tidal mixing during spring tide can produce relatively high salinity surface water with increased thickness of the bottom mixed boundary layer. The particle-tracking experiment also shows that vertical motion of the particles is substantially increased above the sloping bottom during spring tide compared with that of neap tide.

As a result, the offshore detachment of CDW occurs on the sloping side of the bank during spring tide when the tidal energy dissipation is strong enough to overcome the buoyancy effect. This detachment process can be generated by the interaction with tide-induced vertical mixing and horizontal wind-driven movement of CDW because the surface wind primarily induces a northeastward CDW transport across the shelf region where tide-induced vertical mixing is strong.

Moreover, the responses to wind magnitudes suggest that a stronger northwestward wind tends to enhance northeastward transport of CDW across the isobath as suggested in previous model studies, and the wind-induced cross-isobath transport causes a larger offshore low-salinity patch. It means that the freshwater volume of the low-salinity patch closely depends on the magnitude of wind forcing. Therefore, the process of offshore detachment of CDW can be significantly intensified by the northwestward wind with transport increasing with wind strength. The detachment position predicted by the model is fairly consistent with the surface frontal region in the satellite image of sea surface temperature taken during spring tide in summer.

This experimental study focused on the interaction between the tide-induced vertical mixing and the wind-driven cross-isobath transport of CDW toward Jeju Island under the condition of a constant northward/northwestward wind. According to our preliminary result, the detachment of CDW was also found in realistic daily wind cases. However, it was difficult to find the exact two-week variation of CDW for a realistic wind condition. This is probably because the frequency of low-salinity patch is modulated by the random wind at the time scale of a few days, which directly influences the spatial behavior of CDW in the ECS. The study on the frequency modulation owing to time-varying winds may be important to predict the salinity field in the ECS. The effect of time-varying winds will be investigated in a future study.

Acknowledgments

The authors express their sincere thanks to the journal editor and two anonymous reviewers for useful suggestions on improving the manuscript and the support partly by PM51500 from the Ministry of Land Transport and Maritime Affairs in Korea.

REFERENCES

  • An, H. S., 1977: A numerical experiment of the M2 tide in the Yellow Sea. J. Oceanogr. Soc. Japan, 33 , 103110.

  • Banas, N. S., , P. MacCready, , and B. M. Hickey, 2009: The Columbia River plume as cross-shelf exporter and along-coast barrier. Cont. Shelf Res., 29 , 292301.

    • Search Google Scholar
    • Export Citation
  • Bang, I. K., , and H. J. Lie, 1999: A numerical experiment on the dispersion of the Changjiang River plume. J. Korean Soc. Oceanogr., 34 , 185199.

    • Search Google Scholar
    • Export Citation
  • Batchelder, H. P., , C. A. Edwards, , and T. M. Powell, 2002: Individual-based models of copepod populations in coastal upwelling regions: Implications of physiologically and environmentally influenced diel vertical migration on demographic success and nearshore retention. Prog. Oceanogr., 53 , 307333.

    • Search Google Scholar
    • Export Citation
  • Beardsley, R. C., , R. Limeburner, , H. Yu, , and G. A. Cannon, 1985: Discharge of the Changjiang (Yangtze River) into the East China Sea. Cont. Shelf Res., 4 , (1–2). 5776.

    • Search Google Scholar
    • Export Citation
  • Blanton, J. O., , L. Y. Oey, , J. Amft, , and T. N. Lee, 1989: Advection of momentum and buoyancy in a coastal frontal zone. J. Phys. Oceanogr., 19 , 98115.

    • Search Google Scholar
    • Export Citation
  • Chang, P.-H., , and A. Isobe, 2003: A numerical study on the Changjiang diluted water in the Yellow and East China Seas. J. Geophys. Res., 108 , 3299. doi:10.1029/2002JC001749.

    • Search Google Scholar
    • Export Citation
  • Chen, C., , and R. C. Beardsley, 1995a: A numerical study of stratified tidal rectification over finite-amplitude banks. Part I: Symmetric banks. J. Phys. Oceanogr., 25 , 20902110.

    • Search Google Scholar
    • Export Citation
  • Chen, C., , and R. C. Beardsley, 1995b: A numerical study of stratified tidal rectification over finite-amplitude banks. Part II: Georges Bank. J. Phys. Oceanogr., 25 , 21112128.

    • Search Google Scholar
    • Export Citation
  • Chen, C., and Coauthors, 2008: Physical mechanisms for the offshore detachment of Changjiang Diluted Water in the East China Sea. J. Geophys. Res., 113 , C02002. doi:10.1029/2006JC003994.

    • Search Google Scholar
    • Export Citation
  • China Ocean Press, 1992: Marine Atlas of Bohai Sea, Yellow Sea and East China Sea: Hydrology. China Ocean Press, 524 pp.

  • Choi, B. H., 1990: Development of fine-grid numerical tidal models of the Yellow Sea and the East China Sea. J. Korean Soc. Coastal Ocean Eng., 4 , 233244.

    • Search Google Scholar
    • Export Citation
  • Choi, B. H., , K. O. Kim, , and H. M. Eum, 2002: Digital bathymetric and topographic data for neighboring seas of Korea (in Korean with English abstract). J. Korean Soc. Coastal Ocean Eng, 14 , 4150.

    • Search Google Scholar
    • Export Citation
  • Choi, B.-J., , and J. L. Wilkin, 2007: The effect of wind on the dispersal of the Hudson River plume. J. Phys. Oceanogr., 37 , 18781897.

    • Search Google Scholar
    • Export Citation
  • Csanady, G. T., 1974: Barotropic current over the continental shelf. J. Phys. Oceanogr., 4 , 357371.

  • Egbert, G. D., , and S. Y. Erofeeva, 2002: Efficient inverse modeling of barotropic ocean tides. J. Atmos. Oceanic Technol., 19 , 183204.

    • Search Google Scholar
    • Export Citation
  • Egbert, G. D., , A. F. Bennett, , and M. G. G. Foreman, 1994: TOPEX/Poseidon tides estimated using a global inverse model. J. Geophys. Res., 99 , 2482124852.

    • Search Google Scholar
    • Export Citation
  • Galperin, B., , L. Kantha, , S. Hassid, , and A. Rosati, 1988: A quasi-equilibrium turbulent energy model for geophysical flows. J. Atmos. Sci., 45 , 5562.

    • Search Google Scholar
    • Export Citation
  • Guo, X., , and T. Yanagi, 1998: Three-dimensional structure of tidal current in the East China Sea and the Yellow Sea. J. Oceanogr., 54 , 651668.

    • Search Google Scholar
    • Export Citation
  • Ichikawa, H., , and R. C. Beardsley, 2002: The current system in the Yellow and East China Seas. J. Oceanogr., 58 , 7792.

  • Kang, S. K., , S. R. Lee, , and H. J. Lie, 1998: Fine grid tidal modeling of the Yellow and East China Seas. Cont. Shelf Res., 18 , 739772.

    • Search Google Scholar
    • Export Citation
  • Kang, S. K., , M. G. G. Foreman, , H.-J. Lie, , J.-H. Lee, , J. Cherniawsky, , and K.-D. Yum, 2002: Two-layer tidal modeling of the Yellow and East China Seas with application to seasonal variability of the M2 tide. J. Geophys. Res., 107 , 3020. doi:10.1029/2001JC000838.

    • Search Google Scholar
    • Export Citation
  • Kantha, L. H., , I. Bang, , J.-K. Choi, , and M.-S. Suk, 1996: Shallow water tides in the seas around Korea. J. Korean Soc. Oceanogr., 31 , 123133.

    • Search Google Scholar
    • Export Citation
  • Katoh, O., , K. Morinaga, , and N. Nakagawa, 2000: Current distributions in the southern East China Sea in summer. J. Geophys. Res., 105 , (C4). 85658573.

    • Search Google Scholar
    • Export Citation
  • Kim, I.-O., , and H.-K. Rho, 1994: A study on China coastal water appeared in the neighbouring seas of Cheju Island. J. Korean Fish. Soc., 27 , 515528.

    • Search Google Scholar
    • Export Citation
  • Kim, K., , H. K. Rho, , and S. H. Lee, 1991: Water masses and circulation around Cheju-Do in summer. J. Oceanogr. Soc. Korea, 26 , 262277.

    • Search Google Scholar
    • Export Citation
  • Lee, H.-C., , A. Rosati, , and M. J. Spelman, 2006: Barotropic tidal mixing effects in a coupled climate model: Oceanic conditions in the northern Atlantic. Ocean Modell., 11 , 464477.

    • Search Google Scholar
    • Export Citation
  • Lee, S.-H., , and R. C. Beardsley, 1999: Influence of stratification on residual tidal currents in the Yellow Sea. J. Geophys. Res., 104 , (C7). 1567915701.

    • Search Google Scholar
    • Export Citation
  • Lefevre, F., , C. Le Provost, , and F. H. Lyard, 2000: How can we improve a global ocean tide model at a regional scale? A test on the Yellow Sea and the East China Sea. J. Geophys. Res., 105 , 87078725.

    • Search Google Scholar
    • Export Citation
  • Lie, H. J., , C. H. Cho, , J. H. Lee, , and S. Lee, 2003: Structure and eastward extension of the Changjiang River plume in the East China Sea. J. Geophys. Res., 108 , 3077. doi:10.1029/2001JC001194.

    • Search Google Scholar
    • Export Citation
  • Lien, R. C., , and M. C. Gregg, 2001: Observations of turbulence in a tidal beam and across a coastal ridge. J. Geophys. Res., 106 , 45754591.

    • Search Google Scholar
    • Export Citation
  • , X., , F. Qiao, , C. Xia, , J. Zhu, , and Y. Yuan, 2006: Upwelling off Yangtze River estuary in summer. J. Geophys. Res., 111 , C11S08. doi:10.1029/2005JC003250.

    • Search Google Scholar
    • Export Citation
  • Marchesiello, O., , J. C. McWilliams, , and A. Shchepetkin, 2001: Open boundary condition for long-term integration of regional oceanic models. Ocean Modell., 3 , 121.

    • Search Google Scholar
    • Export Citation
  • Mellor, G. L., , and T. Yamada, 1982: Development of a turbulence closure model for geophysical fluid problems. Rev. Geophys. Space Phys., 20 , 851875.

    • Search Google Scholar
    • Export Citation
  • Moon, J.-H., , N. Hirose, , J.-H. Yoon, , and I.-C. Pang, 2009: Effect of the along-strait wind on the volume transport through the Tsushima/Korea Strait in September. J. Oceanogr., 65 , 1729.

    • Search Google Scholar
    • Export Citation
  • Mourn, J. N., , D. R. Caldwell, , J. D. Nash, , and G. D. Gunderson, 2002: Observations of boundary mixing over the continental slope. J. Phys. Oceanogr., 32 , 21132130.

    • Search Google Scholar
    • Export Citation
  • Oey, L. Y., 1986: The formation and maintenance of density fronts on the U.S. southeastern continental shelf during winter. J. Phys. Oceanogr., 16 , 11211135.

    • Search Google Scholar
    • Export Citation
  • Qiu, B., , and N. Imasato, 1990: A numerical study on the formation of the Kuroshio countercurrent and the Kuroshio branch current in the East China Sea. Cont. Shelf Res., 10 , 165184.

    • Search Google Scholar
    • Export Citation
  • Shchepetkin, A., , and J. C. McWilliams, 1998: Quasi-monotone advection schemes based on explicit locally adaptive dissipation. Mon. Wea. Rev., 126 , 15411580.

    • Search Google Scholar
    • Export Citation
  • Shchepetkin, A., , and J. C. McWilliams, 2005: The Regional Oceanic Modeling System: A split-explicit, free-surface, topography-following-coordinate ocean model. Ocean Modell., 9 , 347404.

    • Search Google Scholar
    • Export Citation
  • Simpson, J. H., , and J. R. Hunter, 1974: Fronts in the Irish Sea. Nature, 250 , 404406.

  • Song, Y., , and D. B. Haidvogel, 1994: A semi-implicit ocean circulation model using a generalized topography-following coordinate. J. Comput. Phys., 115 , 228244.

    • Search Google Scholar
    • Export Citation
  • Visser, A. W., 1997: Using random walk models to simulate the vertical distribution of particles in a turbulent water column. Mar. Ecol. Prog. Ser., 158 , 275281.

    • Search Google Scholar
    • Export Citation
  • Wensu, W., 1988: Yangtze brackish water plume—Circulation and diffusion. Prog. Oceanogr., 21 , 373385.

  • Wright, D. G., , and J. W. Loder, 1985: A depth-dependent study of the topographic rectification of tidal currents. Geophys. Astrophys. Fluid Dyn., 31 , 169220.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Horizontal distributions of surface salinity (a) in July 1986 obtained by Chen et al. (2008), (b) in August 1996 observed by National Fisheries Research and Development Institute (NFRDI, in South Korea), and (c) in July 1997 obtained by Lie et al. (2003).

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 2.
Fig. 2.

Model domain bottom topography (m). Rotated square box indicates the model domain. The solid line and dashed box are the location of the vertical section (CJ line) in section 3 and the area for the freshwater volume calculation (WJ area) in section 4, respectively.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 3.
Fig. 3.

Distribution of daily mean surface current and salinity in the East China Sea for the initial condition (t = 0) after 30 days of spinup. The model was spun up with constant Changjiang river discharge of 60 000 m3 s−1, harmonic tidal forcing, and climatological fields for 30 model days.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 4.
Fig. 4.

Horizontal distributions of daily mean surface salinity every 4 days from 12 to 32 model days after spinup. Contour interval is 1.0.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 5.
Fig. 5.

Vertical distributions of daily mean salinity every 4 days from 12 to 32 model days after spinup. Contour interval is 1.0.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 6.
Fig. 6.

Vertical sections of salinity distribution between the Changjiang mouth and Jeju Island obtained in August 1997 (data from Cheju National University). These sections are close to our model section CJ line of Fig. 2.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 7.
Fig. 7.

(a) Time–space plot of the surface salinity along the S line of Fig. 4a and (b) the sea level variation at the center of the S line.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 8.
Fig. 8.

(a) Trajectory of a particle released at the depth of 20 m and surface salinity distribution on day 16 after the spinup. Time series of (b) depth and (c) salinity for the particle trajectory. The particle was released at day 13 during spring tide. The dashed line indicates 30-, 40-, and 50-m isobaths.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 9.
Fig. 9.

(a) Trajectory of a particle released at the depth of 20 m and surface salinity distribution on day 25 after spinup. Time series of (b) depth and (c) salinity for the particle trajectory. The particle was released at day 23 in neap tide. The dashed line indicates 30-, 40-, and 50-m isobaths.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 10.
Fig. 10.

Vertical structure of mean (a),(d) salinity, (b),(e) vertical velocity w, and (c),(f) vertical eddy viscosity Km along the CJ line at (left) day 16 in spring and (right) day 24 during neap tide. Contour interval is 1.0 in salinity, 0.2 × 10−4 m s−1 in vertical velocity w, and 1.0 × 10−2 m2 s−1 in vertical eddy viscosity Km.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 11.
Fig. 11.

(a) Sea surface temperature derived from a satellite image taken on 17 Jun 1992 (Lee and Beardsley 1999). (b) Model-predicted sea surface temperature at day 16 during spring tide: sea surface salinity (white contour line) is superimposed for comparison.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 12.
Fig. 12.

Time–space plot of sea surface salinity along the S line of Fig. 4a simulated by the M2 tide.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 13.
Fig. 13.

Vertical structure of mean (a),(d) salinity, (b),(e) vertical velocity w, and (c),(f) vertical eddy viscosity Km along the CJ line at day 16 during (left) spring tide and (right) the M2 tide. Contour interval is 1.0 in salinity, 0.2 × 10−4 m s−1 in vertical velocity w, and 1.0 × 10−2 m2 s−1 in vertical eddy viscosity Km.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 14.
Fig. 14.

(a) Time–space plot of the surface salinity along the S line of Fig. 4a simulated for half-period of the spring–neap tide and (b) the sea level variation at the center of the S line.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 15.
Fig. 15.

Response of the low-salinity water to wind magnitudes at model day 20, 28, and 36. Winds blow northwestward with magnitude (a)–(c) 0.02, (d)–(f) 0.035, and (g)–(i) 0.05 N m−2. Contour interval for salinity is 1.0.

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Fig. 16.
Fig. 16.

Time series of the freshwater volume across the WJ line in Fig. 2 with a northwestward wind of magnitude 0.02 (dotted line), 0.035 (solid line), and 0.05 N m−2 (dashed line).

Citation: Journal of Physical Oceanography 40, 5; 10.1175/2010JPO4167.1

Save