PROGRESS IN STUDIES ON THE CIRCULATION
AND REGIONAL AIR-SEA INTERACTION IN THE
SOUTH CHINA SEA DURING 1998-2002
WANG Dongxiao, DU Yan, WANG Weiqiang, CHEN Ju and XIE Qiang
Key Laboratory of Tropical Marine Environmental Dynamics, South China Sea Institute of Oceanology,
Chinese Academy of Sciences, Guangzhou 510301, China
In the present paper, by focusing on the dynamic process of the South China Sea (SCS), we review the progress about physical oceanography in the SCS carried out in the recent years (1998-2002), such as the general circulation and the large-scale air-sea interaction. The issues include the progress on the following topics: the seasonal basin-scale circulation, the mesoscale eddy, the interannual variability of the upper layer, the dynamical adjustment of circulation, the water mass exchange through the Luzon Strait and the circulation in the middle and deep layers. In addition, the progresses on the upper mixed layer and seasonal thermocline are also reviewed.
The South China Sea (SCS) is a typical tropical marginal sea, which can exchange water mass with the Pacific and the Indian Ocean via several straits. The semi-enclosed SCS basin spans about 20 degrees in the meridional direction, which makes the dynamics in the SCS similar to the open oceans. However, the curved coastline and numerous islands, as well as the shallow continental shelves and bays there have significant effects on the interior oceanographic process in the SCS. As a result, the characteristics of temperature and salinity distributions, tidal mixing process, mesoscale and small scale eddies as well as the meridional circulation are different from those of the open ocean. The spatial structure of the SCS circulation is of multiple scales, which includes the basin-scale structure, sub-basin scale structure and mesoscale structure. Moreover, the evolution of the SCS circulation is corresponding to multiple time-scales variations, such as the interannual variability, seasonal variability and some transient variations. To a certain extent, such kind of multi-scale feature on the spatial and temporary structures can be attributed to the regional air-sea interaction process, nevertheless, the monsoon forcing and the air-sea heat exchange should have tight connection with that too.
The SCS, which locates in the central region of the Asian-Australian monsoon, has a typical seasonally reversed feature. In recent years, many relevant studies suggest that the monsoon in the SCS has important regional characteristics. That is to say, there exists a tropical sub-system of the East Asian monsoon in the SCS, namely the SCS monsoon. The large-scale monsoon and its corresponding vorticity mainly contribute to the basin-scale circulation, which generates Ekman pumping and significantly influences the thermocline depth through the mass convergence or divergence. And the strong short-term weather phenomenon, such as typhoon, will form several strong mesoscale eddies in the upper ocean, which always leads to cooling of the upper ocean with the associated wind-induced evaporating at the sea surface. The Luzon Strait, which locates in the northeastern SCS and is 2500 m deep, is the mainly passage to the open sea. The Kuroshio and water exchange in the lower layers significantly influence the vertical structure and the interannual variability of the SCS circulation. Besides that, the surface heat fluxes and the sea surface fresh water fluxes related to monsoon variation have important influences on the thermodynamical and dynamical processes in the upper SCS.
By focusing our attention on the fields of the circulation dynamics and air-sea interaction, we will review the recent progress on the SCS oceanography made by the Chinese oceanographers from 1998 to 2002.
II. THE SCS REGIONAL OCEANOGRAPHY
Up to now people have already had numerous knowledge on the evolution of the upper layer circulation in the SCS (Su et al., 1999). When the large-scale circulation is concerned, the northeast monsoon will drive a basin-scale cyclonic circulation with each cyclonic gyre in the northern and southern SCS (NSCS, SSCS), respectively, and the circulation displays a double-gyre structure. In summer, the circulation in the NSCS will still maintain a weak cyclonic gyre. However, due to the southwest monsoon, the circulation in the SSCS displays an anti-cyclonic gyre, and the general circulation over the entire basin shows a dipole structure. The basin-scale circulation is of significant western intensification, which was bore out by many observations and numerical simulations (Qu, 2000; Liu et al., 2000; Cai et al., 1999; Wang et al., 2000).
The promotion of the satellite altimeter technique is worthy of noting. The altimeter sea surface height anomaly (SSHA) data have been used to study the spatial and temporary variability of the large-scale circulation in the SCS on both seasonal and interannual time scales (Mao et al., 1999; Shaw et al., 1999; Ho et al., 2000; Hwang et al., 2000; Wang et al., 2000; Liu et al., 2000, 2001).
1. Dynamical Interpretation to the SCS Circulation Pattern
When the mean circulation pattern is concerned, Wyrtki suggested that it should be mainly attributed to the broad topography and wind stress vorticity as early as 1961. Zeng et al. suggested that the SCS circulation can be taken as barotropic currents due to the coastline constrain and wind stress forcing. The result obtained by Wang et al. through an enclosed boundary simulation agrees well with the large-scale circulation pattern, as a consequence, monsoonal forcing plays a dominant role.By assuming the SCS as an enclosed basin, Liu et al. (2000), found that the basin-scale circulation pattern can be gotten from the Sverdrup stream function. And from the Sverdrup stream function in the interior region, the mass transport of the western boundary current (WBC) is 5-6Sv and 3-4Sv in winter and summer, respectively. They suggested that the upper layer basin-scale circulation can be mainly regarded as a wind-driven circulation forced by the wind stress vorticity, which indicates the SCS circulation having a strong regional characteristics.
2. Analysis of the New Cruise Observations
The direct current observation data still remains shortage in the SCS. Qiu et al. (2000) decompose the ADCP data into the barotropic and baroclinic components, which had been observed for 28.6 days running and at 27 levels (from 11, 12, … 115m) over the continental slope in the NSCS from March to April 1996. The harmonic analysis and energy spectrum method are applied to deal with the time series of the baroclinic current of each level.
Based on the data observed by two cruises of R/Vs <Kexue No. 1> and <Shiyan No. 3> from May to August 1998 during the SCSMEX-IOPs, Wang et al. (2001) analyzed the structure and variation of temperature and salinity at different sections prior and after the summer monsoon onset. The observation indicates that the central SCS is mainly dominated by a typical SCS water mass, however, in the NSCS, especially around the Luzon Strait, water mass in the surface and subsurface layers is influenced significantly by the western Pacific water mass. After the summer monsoon burst, the sea surface temperature (SST) in the NSCS increases clearly, whereas its variation in the middle and southern regions can be negligible. At the same time, the mixed layer in those areas in the NSCS develops. By applying the P-vector method to deal with the SCSMEX data, Pu et al. (2001) found that the Kuroshio intrudes westwards into the SCS; then it forms an anticyclonic loop current; and finally it shows the trend to flow out of the SCS. They also pointed out that the cool center of a cyclonic gyre located in the NSCS, whose strength decreases and whose spatial domain shrinks with the increasing depth, moves southwards with the increasing depth. Off the middle and east Vietnam coast there exist a significant anticyclonic eddy and a cyclonic eddy, which are very stable at these layers shallower than 200 m. Depending on the spectrum of the mooring ADCP data sampled in the northeastern SCS, which span 77 days in the 450-m level and 7 months in the 2000 and 2300 m levels respectively, Yuan et al. (2002) addressed that the upper layer currents in winter is stronger than that in summer and autumn.
Wang et al. (2002) analyzed the data sampled in the central and southern SCS in summer 2000. Their result indicates that the distributions of temperature and salinity in the upper ocean varies with the depth increasing. They also noticed that the temperature and salinity distribution in the middle layers (from about 250 to 400 meters) is much different from that in any other layers above and below. During the survey period, the deep circulation in the SSCS showed a weak cyclonic trend, however, the circulation in the upper layer displayed an anti-cyclonic trend. The circulations by the data-diagnostic method agree well with the geostrophic velocities derived from the Topex/Poseidon SSH during the survey, that is to say, there exist many mesoscale eddies.
3. Oceanography in the Southern SCS
The circulation in the SSCS is much different from that in the NSCS due to the existence of the Nansha Islands. Based on the previous observed data, the analytical results indicate that the circulation in the SSCS mainly consists of some cyclonic and anticyclonic eddies, which shows a multi-eddy structure. In general, a large scale gyre consists of two or more small scale eddies (Fang et al., 1998). He et al. (2000) calculated the circulation in the SSCS by using the temperature and salinity data from 1959-1988, and found that an anticyclonic circulation in winter dominates the region southeastern of the Nansha Islands, and the cyclonic circulation rest of area dominates. Guo et al. (2000) analyzed many in situ data and suggested the following three characteristics. The circulation in the upper layer in the region around the Nansha Islands has an independent enclosed structure. The deep circulation in the SSCS together with the central SCS forms an enclosed circulation system. The direction of circulation in the upper layer is always against that in the lower layer. The reversal of the upper and lower circulations indicates that the SSCS is dominated by a typical baroclinic structure; as a consequence, the stronger vertical motions will occur.
III. EDDIES IN THE SCS
The gyre and eddy structures in the SCS are reported by many hydrographic investigations, no matter how the data were obtained by scientists in China or foreign countries. Wang and Chern (1997) found that there was a cyclonic eddy located off the northwestern Luzon. Later, using his numerical model, Fang et al. (1998) also indicates that this eddy exists all the year round. Chu et al. (1997) mentioned that there obviously existed a cold eddy off the southeast Vietnam coast, which was named as the Vietnam cold eddy. The cross correlation analysis carried by them shows that the local air-sea feedback mechanism is the major cause to form the Vietnam cold eddy. Based on the P-vector method, Chu et al. (1998) obtained the basic characteristics of the multi-eddy structure of the SCS circulation. There exist cold eddies in the region east of Vietnam in the central SCS and in the region off the northwest Luzon Strait, and perennially exists a cold eddy in the region around Dongsha Islands. There also exist some stable eddies in the regions around the Nansha Islands and southeast of the Mekong estuary. On the basis of the altimeter data, Wang et al. (2001) pointed out that there exist some clear seasonal SSH signals of the Kuroshio's intrusion in the northeastern SCS. He et al. (2001) compared the geostrophic velocities derived from the altimeter SSHA data with that from the Argos buoy trajectories. They found that the buoy trajectories could be interpreted by the geostrophic current; as a consequence, by using the diagnosed geostrophic current we can explain most of the transient mesoscale eddy activities marked by the buoy trajectories with good validation.