Ocean Sciences General Contributions II
Presiding: Y Song, Jet Propulsion Laboratory, California Institute of Technology; T Qu, IPRC/SOEST, University of Hawaii
OS51A-01 08:30h
Preliminary Study on the Variability of the Pacific Ocean and One of its Formation Mechanisms
Observations and air-sea models indicate that decadal to interdecadal variabilities exist in the Pacific Ocean. In the North Pacific, the decadal variability is expressed as PDO (Pacific Decadal Oscillation). In 1976-1977, there is a climate "Regime Shift" in the Pacific Ocean. So far oceanographers and meteorologists have not come to agreement about the physical mechanism of this decadal fluctuation. This paper has analyzed the temporal and spatial patterns of PDO on the basis of some variable data from JEDAC, NCEP/NCAR and COADS, and also discussed the effect of zonal wind to PDO, coming to the conclusion that the local wind stress in the North Pacific is a positive feedback to the PDO. Based on the above analysis, a nonlinear air-sea coupled model is constructed to analyze the formation and development mechanism of the decadal variability in the Pacific Ocean. There are indications that, in the midlatitude, the Sverdrup transport resulted from the zonal wind anomaly plays an important role to the heat budget in the high and low latitude. The difference in temperature between high and low latitude is a key factor to sustain this zonal wind anomaly in the midlatitude. When this difference is great to some extent, owing to the non-linearity in the air-sea coupled system, the zonal wind in midlatitude will jump from one state to another, which in turn effects the heat storage in the high and low latitude. Thus there will be a discontinuous oscillation in this coupled system. Therefore, decadal oscillation can also be formed by itself in the midlatitude of the Pacific Ocean. This formation mechanism is also available in the Atlantic Ocean. Besides, the above analysis suggests that the upper layer heat storage variability in 10� N-20� N is not entirely resulted from the subduction water originated in the north. The decadal variability in the tropical Pacific shares some common characteristics with the North Pacific. There also exist anomalous signals similar to the PDO period of more than 20 years. There is a good positive correlation between NINO3 index and PDO index. The east and the west tropical Pacific Ocean are positively and negatively correlated to the anomaly of the upper layer heat storage in the Kuroshio and Oyashio Extension, respectively. Air-sea model and data analysis also show that a decadal variability of the upper layer storage exists in the low latitude, which is negatively correlated to that in the high latitude. The anomaly of the upper layer heat storage in the low latitude may affect the equatorial easterly wind and results in the decadal anti-phase oscillation of the upper layer heat storage in the tropical Pacific Ocean. The negative correlation of the anomaly of the heat storage in the eastern Pacific may also probably come from the modulation of ENSO. So the origination of this oscillation in the tropical Pacific may be in the midlatitude. In addition, the upper heat storage and the thermocline in the tropical Pacific also contain the westward signals of 10-year period. Key words: Pacific Decadal Oscillation, nonlinearity, air-sea coupled system
OS51A-02 08:45h
A regional model of the Indonesian Seas circulation
It is shown that the developed regional model reasonably reproduces the basic features of the Indonesian Seas circulation. The model is based on the Princeton Ocean Model. The horizontal resolution of approximately 10km allows for proper resolution of flows within narrow passages and straits. The vertical resolution has been chosen to properly resolve the surface and bottom Ekman boundary layers and the salinity maximum usually located at 150-200m. The adaptation of the bottom topography based on ETOPO5 has been done by using carefully chosen smoothing. The motion in the whole Indonesian Seas area was assumed to be forced by the inflow and outflow of water due to well pronounced currents such as the Mindanao Current, New Guinea Coastal Surface Current and New Guinea Coastal Undercurrent, North Equatorial Countercurrent, and the major outflow through an appropriately chosen section in the Indian Ocean. So the model has 4 ports simulating these inflows and outflows. The total transports through these ports have been taken from observations. Simple distributions of the transport velocities across the ports have been assumed to provide the open boundary conditions for the barotropic velocities. At the entrance to the ports we used linearized momentum equations with modified friction and nudging to observed velocities. Such equations provide us with values for the baroclinic velocities at the open boundaries. The standard boundary conditions for temperature and salinity have been applied. Technically it appeared convenient to introduce the so-called port channels for tapering off the nudging and additional friction. Such a technique made it possible to do all needed adaptation outside the main region of interest thus not modifying any of the basic equations within this region. The basic features of the overall momentum balance are discussed.
OS51A-03 09:00h
A Numerical Simulation of the East Asian Seas in March 2002: Effect of Vertical Grid Choice
The effect of vertical grid choice in the hybrid Navy Coastal Ocean Model (NCOM) is discussed for the Yellow Sea, East Asian Sea, and Japan/East Sea domain. In particular, the logarithmically stretched hybrid vertical profile used operationally at the Naval Research Laboratory is compared to six variations. The variations include a full z-level run, a full sigma-coordinate run, and other hybrid constructs that modify the hybrid's transitions depth or the structure of the operational grid. The results are compared to each other and some limited observations. The comparisons show that the operational vertical grid structure is a good first guess.
OS51A-04 09:15h
SST Simulations From HYCOM in the Equatorial Pacific Ocean During ENSO Events Since 1990
A 0.72 degree HYbrid Coordinate Ocean Model (HYCOM) is set up for the Equatorial Pacific and used for predictions of sea surface temperature durin El Nino and La Nina events from 1990 to 2003. The model domain spans 30N-30S in latitude and 70W to 110E in longitude, and the latitudinal resolution is increased to 0.36 degrees near the equator to improve model's equatorial dynamics. Model simulations are performed using high resolution (6 hourly) atmospheric forcing from European Centre for Medium-Range Weather Forecasts~(ECMWF). Model simulations include no assimilation of any SST data, and there is no relaxation to any SST climatology. Performance of HYCOM in predicting daily SST is examined using five different mixed layer models: (1) K--Profile Parameterization (KPP), (2) Goddard Institute for Space Studies (GISS) model, (3) Mellor-Yamada 2.5 turbulence closure (MY2.5), (4) Kraus-Turner (KT) model, and (5) Price-Weller-Pinkel (PWP) model. SST simulations from HYCOM are validated against those from moored buoy observations in the equatorial Pacific using various statistical metrics.
http://oceanmodeling.rsmas.miami.edu
OS51A-05 09:30h
Deepwater Overflow through Luzon Strait
This study examines water property distributions in the deep layer of the South China Sea using all available hydrographic data. Our analysis reveals that below about 1500 m there is a persistent baroclinic pressure gradient driving flow from the Pacific into the South China Sea through the Luzon Strait. Applying hydraulic theory with assumptions of zero potential vorticity and flat bottom to the Luzon Strait yields a transport estimate of 2.5 Sv (1 Sv=106 m3 s-1). This result suggests a residence time of only about 30 years in the deep layer of the South China Sea, well below the earlier estimates based on radioactive tracer measurements. Evidence also exists to suggest that topographic control is the correct dynamics of the deepwater overflow. Upon entering the South China Sea via the deep Bashi Channel in the Luzon Strait, water of Pacific origin tends to flow southwestward along local isobaths, and this probably induces a cyclonic circulation in the deep layer of the South China Sea.
OS51A-06 09:45h
Subarctic Ocean-Bottom-Pressure Oscillation and Its Link to the Tropical Pacific ENSO-Related Mass Oscillation
The Gravity Recovery and Climate Experiment (GRACE) satellites have observed strong ocean bottom pressure (OBP) signals in the high-latitudes, but data has been only available since their launch in May 2002. Here we also use a non-Boussinesq ocean model to simulate the OBP from 1948 to 2004 using NCEP/NCAR reanalysis data. It is found that the OBP time-series, from both GRACE and the model, have pronounced interannual variability signals, closely correlating with the Nino3.4 SST index. Such OBP oscillating features have not been reported before. Specifically: 1) There is a strong north-south OBP oscillation between the subarctic and subtropical gyres in the North Pacific Ocean. The bottom pressure difference between the two gyres is found to lead the Niño3.4 index by more than 12 months with a correlation of 0.52 at the 99% significant interval. 2) There is also a moderate east-west OBP oscillation in the tropical Pacific. Unlike the subarctic oscillation, the bottom pressure difference between the eastern (Niño3.4 region) and the western tropical Pacific (warm-pool region) is in phase with the Niño 3.4 index with a correlation 0.82. These two observations indicate ENSO has a direct effect on oceanic mass redistribution. Vorticity analysis leads to the explanation of a possible link between the north-south subarctic mass oscillation and the east-west tropical mass oscillation, due to the presence of bottom pressure torques, that contribute to the vorticity balance in the same sense as the wind stress curl. The subarctic OBP oscillation tends to decrease/increase the western boundary currents, adding/removing water mass in the western Pacific, consequently altering west-east pressure gradient along the Equator, and is probably a contributor or barotropic response of the onset/demise of El Niño. The GRACE and model results have been cross-verified by available TOPEX/Poseidon data and in-situ measurements.