The Pacific covers over half of the global ocean area and its wind- driven variability is of interest both for its consistency with wind stress observations and for potential air–sea feedbacks that could influence climate. Changes in Pacific Ocean circulation since the early 1990s to the present, from the subarctic gyre to the southern ocean, observed with satellite ocean data and in situ ocean measurements, are in good agreement and consistent with the expected dynamical response to observed changes in wind stress forcing.
The subarctic gyre in the North Pacific poleward of 40°N consists of the Alaska Gyre to the east and the Western Subarctic Gyre (WSG). Since 1993, the cyclonic Alaska Gyre has intensified while decreasing in size. The shrinking is seen in the northward shift of the North Pacific Current (NPC, the high gradient region centred about 40°N in Figure 3.10) and has been described using the satellite altimeter, XBT/hydrography, and, more recently, Argo profiling float data (Douglass et al., 2006; Cum- mins and Freeland, 2007). A similar 20-year trend is detected in the WSG, with the northern WSG in the Bering Sea having intensified while the southern WSG south of the Aleutian Islands has weakened. These decadal changes are attributable to strengthening and northward expansion of the Pacific High and Aleutian Low atmospheric pressure systems over the subarctic North Pacific Ocean (Carton et al., 2005).
The subtropical gyre in the North Pacific also expanded along its southern boundary over the past two decades. The North Equatorial Current (NEC) shifted southward along the 137°E meridian (Qiu and Chen, 2012; also note the SSH increase east of the Philippines in Figure 3.10 indicating the southward shift). The NEC’s bifurcation latitude along the Philippine coast migrated southward from a mean latitude of 13°N in the early 1990s to 11°N in the late 2000s (Qiu and Chen, 2010). These changes are due to a recent strengthening of the Walker circulation generating a positive wind stress curl anomaly (Tanaka et al., 2004; Mitas and Clement, 2005). The enhanced regional sea level rise, >10 mm yr –1 in the western tropical North Pacific Ocean (Timmer- mann et al., 2010, Figure 3.10), is indicative of the changes in ocean circulation. The 20-year time-scale expansion of the North Pacific sub- tropical gyre has high confidence owing to the good agreement seen in satellite altimetry, subsurface ocean data and wind stress changes. This sea level increase in the western tropical Pacific also indicates a strengthening of the equatorward geostrophic limb of the subtropical cells. However, the 20-year increase reversed a longer term weakening of the subtropical cells (Feng et al., 2010), illustrating the high difficulty of separating secular trends from multi-decadal variability.
Variability in the mid-latitude South Pacific over the past two decades is characterized by a broad increase in SSH in the 35°S to 50°S band and a lesser increase south of 50°S along the path of the ACC (Figure 3.10). These SSH fluctuations are induced by the intensification in the SH westerlies (i.e., the SAM; see also Section 3.4.4), generating positive and negative wind stress curl anomalies north and south of 50°S. In response, the southern limb of the South Pacific subtropical gyre has intensified in the past two decades (Cai, 2006; Qiu and Chen, 2006; Roemmich et al., 2007) along with a southward expansion of the East Australian Current (EAC) into the Tasman Sea (Hill et al., 2008). The intensification in the South Pacific gyre extends to a greater depth (>1800 m) than that in the North Pacific gyre (Roemmich and Gilson, 2009). As in the north, the 20-year changes in the South Pacific are seen with high confidence as they occur consistently in multiple lines of medium and high-quality data. Multiple linear regression analysis of the 20-year Pacific SSH field (Zhang and Church, 2012) indicated that interannual and decadal modes explain part of the circulation varia- bility seen in SSH gradients, and once the aliasing by these modes is removed, the SSH trends are weaker and more spatially uniform than in a single variable trend analysis.
The strengthening of SH westerlies is a multi-decadal signal, as seen in SLP difference between middle and high southern latitudes from 1949 to 2009 (Gillett and Stott, 2009; also Section 3.4.4). The multi-decadal warming in the Southern Ocean (e.g., Figure 3.1, and Gille, 2008, for the past 50 to 70 years) is consistent with a poleward displacement of the ACC and the southern limb of the subtropical gyres, by about 1° of latitude per 40 years (Gille, 2008). The warming and corresponding sea level rise signals are not confined to the South Pacific, but are seen globally in zonal mean fields (e.g., at 40°S to 50°S in Figures 3.9 I and 3.10). Alory et al. (2007) describe the broad warming consistent with a southward shift of the ACC in the South Indian Ocean. In the Atlantic, a southward trend in the location of the Brazil-Malvinas confluence (at around 39°S) is described from surface drifters and altimetry by Lump- kin and Garzoli (2011), and in the location of the Brazil Current sep- aration point from SST and altimetry by Goni et al. (2011). Enhanced surface warming and poleward displacement, globally, of the western boundary currents is described by Wu et al. (2012).
Changes in Pacific Ocean circulation over the past two decades since 1993, observed with medium to high confidence, include intensifica- tion of the North Pacific subpolar gyre, the South Pacific subtropical gyre, and the subtropical cells, plus expansion of the North Pacific sub- tropical gyre and a southward shift of the ACC. It is likely that these wind-driven changes are predominantly due to interannual-to-decadal variability, and in the case of the subtropical cells represent reversal of earlier multi-decadal change. Sustained time series of wind stress forcing and ocean circulation will permit increased skill in separating interannual and decadal variability from long-term trends (e.g., Zhang and Church, 2012).