Dynamic Structure Of Ocean Circulation In The Tropical Western Pacific Ocean And Its Simulation

Ocean circulation in the tropical western Pacific is a key element for subtropicalgyre and tropical gyre covering main part of the whole Pacific, which is also near thewarm pool in the western Pacific and plays an important role for formation andvariation of the warm pool. However, up to now, many problems about it in thetropical western Pacific have not been solved yet, including even dynamic structure.Therefore, studies on the circulation of the tropical western Pacific in this thesis, notonly have theoretical significance for circulation dynamics in the Pacific, but alsohave theoretical and practical value for examination and prediction of climate change.Bifurcation of the North Equatorial Current (NEC) in the Pacific is a veryprotruding phenomenon in the ocean, especially in the western boundary region of thePacific, which is the headstream of the subtropical and tropical gyres, determining thestructure of basin-scale ocean circulation of the Pacific to a great extent and playing acrucial role in ocean circulation dynamics. On the other hand, it determines themeridional transport of mass and heat and thus affects heavily climate change and isan important component in climate system. Therefore, bifurcation study of the NEC isthe first topic the thesis will work on.Since the 1930s, there are two kinds of studies on the bifurcation of the NEC,one based on observed hydrographical data and the other based on numericalmodelling. The results derived from observational data, are more consistent with factswith significant errors owing to poor resolution of the data used. The results fromnumerical models have higher temporal and spatial resolution, while they deviatefrom observations due to certain defects of the models themselves. Up to now, thesemodel results about bifurcation of the NEC, such as variation of bifurcation latitude,have great differences from each other. So it is the key to study bifurcation of theNEC how to find a data set with high enough resolution or to come up with anumerical model, which is based on real physical processes and can be validated byobservations.In this thesis, altimetry data with high resolution from October 1992 throughDecember 2004 are analyzed and the bifurcation latitude is determined accordingLagrangian trajectory method. Results show that, on annual average, the NECbifurcates at about 13.4°N. As for its seasonal variability, the southernmost latitude ofthe NEC bifurcation occurs at about 12.9°N in June and the northernmost latitude atabout 14.1°N in December. The mean state of 12-year data shows that the bifurcationlatitude in July is significantly larger than that in June and August, which should resultfrom local wind stress curl. In addition, the interannual variability of bifurcationlatitude is also studied. During El Ni?o years the bifurcation latitude of the NECmoves to the northernmost, while during La Ni?a years the bifurcation latitude of theNEC moves the southernmost.In the tropical western Pacific, especially in the region near the western boundarycurrents and the Indonesian Throughflow (ITF), there are many islands, the geometryand topography are very complex and historical hydrographical data and currents dataare very scarce. Thus it is extremely difficult to study ocean circulation in the regionby analyzing observational data, especially to comprehensively understand thesubsurface current structure. Therefore, a hybrid coordinate ocean model HYCOM,which is applicable in both open ocean and regions with steep topography, is used tosimulate the climatology and seasonality of currents in the tropical region, especiallywestern boundary currents and related flows, such as, the New Guinea CoastalUndercurrent (NGCUC), Mindanao Undercurrent (MUC), Equatorial Undercurrent(EUC) and the ITF. Simulated results well reproduce the main currents and theirseasonal variations. The sea surface height (SSH) field from modeling is similar toobservational SSH. Simulated current, temperature and salinity structures in some keysections agree well with observational data (Johnson et al., 2002). So, the simulatedresults are trustable and valid to analyze the circulations in the tropical western Pacific.The main results are as follows.(1) It is a key issue in equatorial circulation dynamics that where the EUC starts.Due to lacking of observational data and complex geometry and topography near theorigin area of the EUC, there are only a couple of studies on it, which claim the EUCstarts somewhere in between 135-137°E. Series of meridional sections of simulatedzonal velocity show that the EUC starts somewhere 129-130°E. In addition, based onthe horizontal velocity field at 200m, it can be obviously seen the EUC has foursources with differences in four seasons. (i) The NGCUC feeds to the EUC fromabout 135°E, which is agreed to Tsuchiya et al. (1989). (ii) The Mindanao Current(MC) feeds the EUC, which is similar to the conclusion of Lu and McCreary (1995),and Gu and Philander (1997). (iii) Waters from the South China Sea (SCS) throughthe Mindoro Strait, Sulu Sea and west part of Sulawesi Sea flow into the EUC fromthe north of Halmahera Island, especially in spring, summer and fall. (iv) The waterfrom Indian Ocean through Banda Sea and Maluku Sea flow northward along the eastcoast of Sulawesi Island and feeds into the EUC from the east of Halmahera Island(especially in spring). So, as for the water source of the EUC, it includes the waterfrom the South China Sea, Indian Ocean and east of Mindanao and in terms of itsorigin it should be thought starting from 129-130°E.(2) The MUC was found in the late 1980s. The origin of the MUC is veryimportant for circulation dynamics in the region, but few studies on it was madeowing to lack of enough observational data. In the thesis, simulated results show thata small part of the MUC seems to be from the MC in the layer of 300-500m. But it ismore important that a northward current along the western boundary of the MalukuSea (east of Sulawesi) is the main water source for the MUC, indicating considerablepart of the MUC coming from the southern hemisphere. At 600, 700m, especially at800, 900, 1000m, the velocity of the MC decreases rapidly and even diminishes, theMUC is very sticking out, and the water is mainly from the northward flow east of theSulawesi Island. As for the destination of the MUC, simulated results at 500, 600mshow that large part of MUC turns to the east near 12-13°N, while a small part turnsback to the south and feeds the MC. Below 700m, the MUC flows eastward alongabout 12°N, which is agreed to Hu and Cui (1989, 1991) and Wang and Hu (1998).(3) About the Indonesian Thoughflow. In mean state, HYCOM results show thatin the upper 250m layer, the ITF is mainly from the western Pacific to the easternIndian Ocean through Makassar Strait. However, below 300m, the water from theeastern Indian Ocean flows into the Pacific via Flores Sea, Banda Sea and Maluku Sea.According to the velocity field at 200m, the southward flow from the SCS throughMindoro Strait is the main source of the ITF in the Makassar Strait, and the othersmall part is from the MC.(4) About the South China Sea Branch of the Kuroshio (SCSBK). The SCSBKtakes place year round obviously in our simulation. It enters the SCS northwestwardfrom the north tip of the Luzon Island and flows southwestward again along about21°N with width 100-200km, in the upper layer to depth 300-400m and the maximumvelocity is over 20cm/s, even exceeding 40cm/s in winter. The annual mean ofSCSBK transport amounts to 6.5Sv with the largest value in winter and the smallestone in May and September, which proves the statement in Qiu et al. (1984) and Guoet al. (1985) about the SCSBK existence and to some extent explain quantitatively itsseasonal variations.