Characteristics And Mechanisms Of Sea Level Variations In The South China Sea And The Indian Ocean

Sea level changes in the South China Sea (SCS) and the Indian Ocean were analyzed bysatellite altimetry data from1993to2009. The dominant process affecting sea levelvariability vary in the studied area: steric response, the response to wind stress curl forcing,baroclinic Rossby wave and eddies’ effect are all important. Steric model (e.g, Buoyancymodel, Markov model), one-half-layer reduced-gravity model (RG) and eddy-track modelwere applied to investigate these processes. Furthermore, several diagnostic experiments wasdone by using ROMS, we emphasized in analyzing the influence of wind, shortwave radiationand freshwater flux on the sea level changes on the seasonal time-scale.There exist obviously spatial and temporal variations of the sea level in the SCS and theIndian Ocean. The sea level in the whole SCS basin is rising. The sea level in the most ofIndian Ocean, especially the southern Indian Ocean, is also rising. The rise trend in equatorialarea is not clear. The rise rates is low in the Arabian Sea and Bay of Bengal and their east-west spatial distribution is asymmetry. The rates in the west is lower than that in the east. Inthe Bay of Bengal, there is a significant feature which a down area located in the northwestand the decrease rate is2mm/yr. In South Indian Ocean (south of15S), the rise rates of sealevel are discrete, which are closely related to the eddy activities.The results of EOF analysis imply that the sea level changes primarily on the seasonal timescale associated with the monsoon in the the northwest of Luzon Island, the western Bay ofBengal and the Arabian Sea; Interannual variability of sea level are mainly distributed in theeastern Indian Ocean and the central southern Indian Ocean. There are two areas in the IndianOcean (80°E-92°E,14°S-8°S and55°E-70°E,12°S-5°S) where sea level changes significantlyreflect the annual and inter-annual variation, respectively. And sea level changes on the inter-annual time scale in these two region are coherent with ENSO/IOD based on the lead-lagrelationship. There are notable seasonal and interannual signal of the spatial average sea levelchanges in the SCS, and its interannual variability is strongly related to ENSO.Steric height changes are the most important part of sea level changes. Overall, it is obviousto the steric processes distributed of the eastern and northern SCS, the southern Arabian Sea, and the Indian Ocean (north of12°S). Steric effects are bad in south of12°S and the sea levelchanges in this region are markly caused by Rossby waves and eddies.The response to buoyancy flux dominates seasonal sea level in some area. Compared withthe heat flux, freshwater flux could be neglected because of its low contribution. Through theanalysis of heat flux, middle regions of the SCS, the Arabic Sea, and a little part of the IndianOcan, were induced significantly. Heat flux mainly affects the annual signal in the sea levelvariation. Lagged response existed between sea level change and heat flux from thenorthwestern part outside Australian to the western tropical Indian Ocean (northwestern tosoutheastern direction).Responses to the local wind stress are the main mechanism of seasonal sea level variationof the SCS and the northern Indian Ocean (including the Bengal Bay and the Arabic Sea). On-half-layer reduced-gravity model can explain the band area between12°S and8°S well. Alongthe zonal direction of12°S, Rossby waves break down in the area between50°E and70°E.Sea level variations in the Indian Ocean (southern of15°S) are mainly influenced by eddyactivities. By using edd-track model, anticyclonic eddies accounts for more than10%whilecyclonic only contributes5%. Especially, total contribution of the eddies to the sea levelchange could reach up to25%in the southern of20°S.The interannual variability of sea level in the SCS are closely related to the subsurfacewater (depth about100-300m). ENSO/PDO/NPGO signals affect the temperature and salinitychanges of the subsurface water and further lead to interannual variations of sea level.Diagnostic experiments of ROMS numerical simulation further reflect the impacts of thewind stress, short wave radiation flux and freshwater flux on sea level change. Removing thewind stress, the spatial distribution patterns of sea level on the four seasons have greatchanges. The response to wind stress dominates seasonal sea level variability in the SCS, theBay of Bengal, and the Arabian Sea. Effects of wind filed can not be ignored in these regions.Removing the short wave radiation flux, the spatial distribution patterns of sea level on thefour seasons have no great changes, only the values of sea level have small changes in somelocal area.Removing the freshwater flux, the spatial distribution patterns of sea level on thefour seasons change little.Overall, among three factors affecting sea level changes, wind fieldsis the most important one, short wave radiation flux plays a secondary role, and freshwaterflux contributes least.