Although the preceding discussion has attempted to describe the configuration of the mean flow within the Pacific Sector, the currents actually exhibit substantial spatial and temporal variability. In his description of the Polar Front, Mackintosh (1946) noted that "it forms twists and loops that may extend as much as 100 miles north or south, and it possibly even forms isolated rings." Since the jets within the ACC extend to great depths, interactions with bottom topography influence the nature and intensity of current variability. For example, Gordon (1967b) noted the unusual complexity of the Polar Front downstream of the Udintsev and Eltanin Fracture Zones which he characterized as a "double" polar front. He later described similar complexity downstream of the Macquarie Ridge (Gordon, 1972a) and speculated that it might be due to rings shedding from the ACC at the ridge and advecting downstream. His speculation was supported by the results of a numerical model (Boyer and Guala, 1972).
The belief that current jets within the ACC tend to meander and occasionally form detached rings has subsequently been confirmed by ship surveys (Joyce and Patterson, 1977; Savchenko et al., 1978; Patterson and Sievers, 1979/80; Joyce et al., 1981; Peterson et al., 1982), moored current meter arrays (Sciremammano, 1979; Hofmann and Whitworth, 1985) and satellite imagery (Legeckis, 1977; Peterson, 1985).
Extensive meandering of the fronts within the ACC has been observed within Drake Passage and south of New Zealand. Within the restricted confines of Drake Passage, meanders can shift the location of the Subantarctic Front by more than 100 km in a few weeks (Hofmann and Whitworth, 1985). In the open ocean, such shifts in frontal position might be several times larger. In the seven months between two surveys of a region southeast of New Zealand, Bryden and Heath (1985) found that the flow just north of the Subantarctic Front had changed direction by 180°. That current meanders extend to the bottom was confirmed by Sciremammano (1979), who detected such a feature in the Polar Front from current meter data at 2,700 m, and found the signal to be coherent with the surface expression as revealed by satellite images (Legeckis, 1977).
Bryden (1983) reviewed eddy observations in the Southern Ocean and noted that a typical eddy south of New Zealand or in Drake Passage has a radius of 3050 km and surface speeds of 30-40 cm sec."1. The frequency of formation of detached rings is not known, but using eight-month current records from central Drake Passage, Pillsbury and Bottero (1984) inferred the presence of five cold-core rings and one warm-core ring. Also, during a one-year study in Drake Passage, one warm-core and three cold-core rings were detected passing through an extensive array of current, temperature and pressure sensors (Hofmann and Whitworth, 1985).
In the open South Pacific, there are fewer direct observations of mesoscale variability. However, distributions of buoy-derived eddy kinetic energy (Patterson, 1985), the variance in geopotential anomaly (Lutjeharms and Baker, 1980), and the variance in altimeter-derived sea surface height (Cheney et al., 1983) all exhibit large values near the Macquarie Ridge and the Udintsev and Eltanin Fracture
Zones (e.g., Fig. 3.9). Current meter data in northern Drake Passage and southeast of New Zealand reveal eddy kinetic energies at depth comparable to those found within the Gulf Stream and Kuroshio (Nowlin et al., 1981; Bryden and Heath, 1985).
Was this article helpful?