Statistical Analyses On Mesoscale Eddies In The South China Sea And The Northwest Pacific

Mesoscale eddies exist almost everywhere in the ocean. Because they contain large kinetic energy with 80%, or even 90% of the total energy of the ocean circulation, they play very important roles in the whole ocean circulation in the world. There are abundant eddies in the South China Sea (SCS) and the Northwest Pacific(10°N-35°N,120°E-150°E) (NWP), but in the past, for the lack of detailed data, the integrated characteristics of eddies in these regions had not been lucubrated. In the thesis, eddies are identified and traced in the SCS and the NWP from 1993 to 2001 using Topex/Poseidon merged ERS-1/2 altimeter data through several criteria. Eddies' characteristics of generation and moving, occurrence probability and contribution to sea level undulation are statistically analyzed. In addition, we discuss eddy generation mechanism in the SCS. There were about 18 mesoscale eddies annually generated averagely in the SCS during this period, and warm eddies were a bit more than cold eddies. Eddies in the SCS mainly generated locally, and few passed from the Luzon Strait. While in the NWP, about 60 eddies were formed annually, including 35 warm eddies and 25 cold eddies, and about 1/5 of them came from east of this region. In these two regions, the radii diminished poleward and followed up with the baroclinic Rossby deformation radii. Eddies' lifetimes were generally less than 180 days and mainly ranged from 30 to 60 days. About 80% eddies moved westward in the SCS. Mesoscale eddies in the SCS could be separated into two relatively independent systems bounded by 17°N, where eddies could not propagate across. In the NWP, about 90% eddies moved westward. The westward moving speeds slowered in the poleward direction, even turned eastward. Around the strong currents, like North Equator Current (NEC) and the Kuroshio Extension (KE), meridional variation of the eddy zonal moving speed was much closer to that of the phase speed of Rossby Wave including effects of the zonal mean current. Between 20°N and 30°N where the mean current is weak, the westward moving speed was about 4cm/s larger than the phase speed of Rossby Wave and the westward component the of mean current. In the SCS, eddies were active along a band from east of Southern Vietnam to Southwest of Taiwan in the northeast-southwest direction. Statistically calculated on 2o×2o grids, areas of relatively large eddy occurrence probabilities (OPs) were found located to the west of the Luzon Strait (named R1) and to the east of Southern Vietnam(named R2) with probabilities of about 23% and 25%, respectively. From November to May, OPs to the Southwest of Taiwan (namely Northern R1) were larger than those of other months, which mainly attributed to the warm eddies. From May to October, OPs to the west of the Luzon Island (namely Southern R1) were larger than those of other months. Cold and warm eddies both contributed to the OPs. In R2, OPs from March to November were larger than those of other months. Cold eddies' contributions were larger than the warm's from September to November, which would appear an absolutely opposite state during March and July. In the NWP, OPs of eddies showed seasonal variation with the largest in spring and smallest in autumn. There were three zones of OPs: one was to the east of 145°E, one was the zone extending westward to the vicinity of Luzon Strait between 19°N and 22°N, and the other was to the north of 25°N and to the west of 141°E, with maximum of were 25%, 17% and 25%, respectively. The areas of high OPs of cold eddies and warm eddies were different. Noticeably, in the place with depth >200m in the SCS and Pacific, the mean contributions of eddies to the sea level undulation were respectively about 36% and 56%. The maximum contribution could reach as high as 80% where the rms of SLA was very high. In the region where eddies occurred frequently in the SCS, high rms of SLA and large contributions were almost occurred at the same places. While in the NWP, large contributions were not always coherent with high OPs, for examp...