Geophysical Research Letters

Antarctic ozone depletion causes an intensification of the Southern Ocean super-gyre circulation



[1] Recent climate trends over the Southern Hemisphere (SH) summer feature a strengthening of the circumpolar westerly and a weakening of the midlatitude westerly extending from the stratosphere to Earth's surface. Much of the change is attributable to Antarctic ozone depletion. However, the consequential ocean circulation changes are unknown. Here I demonstrate that the observed surface wind changes have forced a southward shift and spin-up of the super gyre, which links the subtropical South Pacific, Indian and Atlantic Ocean circulation, advecting more warm water southward. The circulation change includes a strengthening of the East Australian Current (EAC) flow passing through the Tasman Sea. The southward shift may be responsible for the observed unusually large warming in the SH midlatitude ocean and may contribute to the reported range extension to the south of many marine species in the South West Pacific.

1. Introduction

[2] Since the late 1970s climate change in the SH has been marked by a strengthening of the circumpolar westerlies extending from the stratosphere [Hurrell and van Loon, 1994; Waugh et al., 1999; Zhou et al., 2000; Thompson and Solomon, 2002] to the troposphere and Earth's surface [Hurrell and van Loon, 1994; Thompson and Solomon, 2002; Gillett and Thompson, 2003]. The greatest changes in the stratosphere have occurred during the spring months [Waugh et al., 1999; Zhou et al., 2000; Thompson and Solomon, 2002; Gillett and Thompson, 2003; Randel and Wu, 1999], however, closer to Earth's surface, the largest changes have occurred during the summer months [Thompson and Solomon, 2002; Gillett and Thompson, 2003]. This time lag was elucidated using ozonesonde observations (S. Solomon et al., Four decades of ozonesonde measurements over Antarctica, submitted to Journal of Geophysical Research, 2005). These changes have manifested as a bias toward the high-index polarity of the southern annular mode (SAM), the leading mode of SH midlatitude variability [Thompson et al., 2002; Hartmann and Lo, 1998; Kidson, 1988; Karoly, 1990], operating on all time scales.

[3] Although simulations conducted with increasing CO2 produced trends that are of the same sign as the observed trend [Fyfe et al., 1999; Kushner et al., 2001; Cai et al., 2003], the simulated trend is much smaller than the observed. Based on observations, Thompson and Solomon [2002] argued that the summer trend is consistent with forcing by stratospheric ozone depletion, a hypothesis tested by Gillett and Thompson [2003], who concluded that the anthropogenic emissions of ozone-depleting gases primarily account for the observed trends of surface flows at the midlatitudes (see Figure 1). The contribution of ozone depletion has been confirmed through analysis of model simulations submitted for the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (R. L. Miller et al., Forced variations of annular modes in the 20th century IPCC AR4 simulations, submitted to Journal of Geophysical Research, 2006), which further supports that much of the observed trend is induced by ozone depletion.

Figure 1.

(top) Ozone depletion-induced and (bottom) observed near surface wind changes averaged over SH warm months for the past decades. This figure is reproduced, with permission, from Gillett and Thompson [2003, Figure 4]. Copyright 2006 AAAS.

[4] However, examinations of the impact by ozone depletion have largely been limited to Earth's surface and the impact on the ocean circulation and marine ecosystem has not been explored. Utilizing recent Argo data and hydrographic observations in early 1990s, Roemmich et al. [2006] found that there is a spin-up of the South Pacific gyre, and suggested that it is linked to wind stress changes associated with the trend in the SAM since early 1990s. Here I demonstrate that the trend in surface wind stress since late 1970s has generated pronounced changes to the southern subtropical and midlatitude ocean circulation using a well-established Sverdrup model [Godfrey, 1989]. Recent studies [Saenko et al., 2005; Cai et al., 2005] examining the response of ocean circulation to greenhouse warming show that much of the ocean circulation change, including a strengthening and a poleward extension of the subtropical gyre circulation, can be calculated using Sverdrup balance forced by wind changes.

2. Data

[5] I focus on the warm months of December through to May, when climate trends are strongest and statistically significant [Thompson and Solomon, 2002; Gillett and Thompson, 2003; Marshall, 2003] and explore the consequential ocean circulation change. These months encompass the period January to March, when the SH subtropical gyre circulation extends furthest south [Ridgway and Godfrey, 1997]. Linear trends of the surface wind stress and its curl from 1978 onward have been obtained from the National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis [Kalnay et al., 1996], which has assimilated satellite sounder data since the late 1970s [Hines et al., 2000], producing a post-1978 reanalysis that compares well with station-based observations [Marshall, 2003] and radiosonde data [Thompson and Solomon, 2002].

3. Results

[6] Previous studies did not examine the impact on the SH subtropical and midlatitude oceanic circulation probably because the maximum change of zonal wind stress is located far to the south (Figure 1), at approximately 60°S [Thompson and Solomon, 2002; Gillett and Thompson, 2003]. However, it is the wind stress curl, not the wind stress itself, that drives the large-scale ocean circulation. The wind stress curl at southern midlatitudes is dominated by the meridional gradient of zonal wind stress. As a result, the center for the maximum wind stress curl change is located at approximately 45°S–50°S (Figure 2 (top)).

Figure 2.

Pattern of wind stress curl (N m−3, scaled by a factor of 10−6) for warm months (December–May). (top) The trends are calculated using linear trend-fitted surface wind stresses from the NCEP/NCAR reanalysis [Kalnay et al., 1996] for the period from 1978 to 2002. (bottom) The trends of increasing positive curl in the midlatitudes represent significant changes to the curl field prior to the onset of the trends, implying a southward strengthening of the subtropical gyres.

[7] The curl trend represents significant changes to the curl field before the onset of the trend (Figure 2 (bottom)), which reflects several distinct features. Positive curl (Figure 2 (bottom)) in the southern subtropical and middle latitudes implies a counter-clockwise gyre circulation in all three oceans. In midlatitudes, the contour of zero curl, which defines the southern boundary of the gyre circulation, is situated approximately at 50°S and extends across the three ocean sectors. This feature determines the pan-Southern Ocean connectivity of the basin gyres. Overall, the structure of the positive curl change (Figure 2 (top)) suggests an intensifying circulation at the southern edge of the subtropical gyres.

[8] The spatial pattern of the wind-driven circulation change is determined by Godfrey's Island Rule model of depth-integrated transport stream function [Godfrey, 1989] forced by the trend-fitted surface wind stress. This well-established Sverdrup model [Sverdrup, 1947] is elegant in that given global fields of surface wind stresses, the wind-driven ocean circulation can be determined. The model has all the assumptions of the Sverdrup balance, including that the depth of no wind-driven motion is shallower than the total water depth. As the model requires an eastern boundary, I carry out the calculation to the southern-most latitude, 54°S.

[9] The oceanic circulation trends show a pan-Southern Ocean scale of circulation change (Figure 3 (top)). To put the change into an appropriate context, the total wind-driven circulation prior to the onset of the trend is calculated (Figure 3 (bottom)). In the South Pacific, the wind-driven EAC averaged over December to May bifurcates at approximately 20°S [Church, 1987; Webb, 2000; Ridgway and Dunn, 2003]. The majority of the EAC flow passes through the Tasman Sea. The flow then veers northwest into the Great Australian Bight and the Indian Ocean and then to the Atlantic Ocean [Godfrey, 1989; Ridgway and Dunn, 2003], before it retroflects. The entire flow forms the Southern Ocean inter-basin super gyre linking the South Pacific, Indian and Atlantic Oceans, a feature determined by the interbasin connectivity of the wind stress curl (Figure 2 (bottom)). To the south, the Antarctic Circumpolar Current (ACC), in part determined by Sverdrup balance north of the southern-most tip [Godfrey, 1989; Baker, 1982] of South America, is about 120Sv. The circulation change (Figure 3 (top)) indicates that the EAC flow through the Tasman Sea has strengthened and that the super gyre has intensified and shifted southward, the largest change being in the South Atlantic. The ACC at the southern-most tip of the South America increases by 11Sv.

Figure 3.

Transport stream function (Sv, 1 Sv = 106 m3 s−1) calculated using Godfrey's Island Rule model [Godfrey, 1999], forced by the linear trend-fitted surface wind stress [Kalnay et al., 1996] averaged over the warm months (December to May). The trends show (top) pronounced changes to the Southern Ocean super-gyre circulation prior to (bottom) the onset of the trends, representing a significant intensification of the gyre circulation. The EAC increases by about 20%. The flow directions are indicated with arrows.

[10] South of 30°S, the EAC increases markedly, by about 9Sv or about 20% at 36°S. The large increase is due to the fact that the lateral boundaries of the South Pacific gyre are either close to or extend into the zonal strip of strong curl changes. This feature allows the South Pacific gyre to tap into the effect of the curl changes integrated from the South American coast to New Zealand, and to near the west boundary. The resultant circulation changes include (1) stronger northward flows in the interior and off New Zealand's northeast coast, and (2) swifter southward EAC flows through the Tasman Sea, as the super-gyre circulation shifts southward. One direct consequence of the circulation change is strong warming along the path of the EAC intensification, and this provides an explanation for the strong ocean warming [Pittock, 2003] off Maria Island (148.16°E, 42.36°S) near Tasmania, as the EAC intensifies. As the super-gyre circulation shifts southward, the EAC decreases north of 30°S.

[11] The large change in the South Atlantic, with an increase in the Brazil Current of some 40%, is determined by the feature that in the latitude band of 43°–54°S the two sides of the South American continent form the major east and west boundaries. The annular-scale integration from the east to the west boundaries of the strong wind stress curl changes (Figure 2 (top)) is a narrow but swift flow off the west boundary as a result of Stommel's westward intensification [Stommel, 1948]. Thus the greatest change occurs off the west boundary of the South Atlantic. Note that, however, this change is somewhat overestimated, because it takes longer than 25 years for the wind effect off the eastern boundary to reach the western boundary. Nevertheless, it is worth highlighting that the intensification of the super-gyre circulation is underpinned by increasing trends and a southward extension of isopleths of the zonally-averaged positive curls (Figure 4), including that of zero-curl, which indicates the southern boundary of the super gyre.

Figure 4.

Zonal mean of wind stress curl averaged over December to May (color contours, N m−3, scaled by a factor of 10−6) from the NCEP/NCAR reanalysis [Kalnay et al., 1996]. Superimposed are contours of linear trends of the zonal-mean wind stress curl. They show an increasing trend of the midlatitude positive curl with isopleths shifting southward, indicating an intensifying, southward-shifting subtropical gyre circulation.

4. Discussion

[12] The results described here have important implications for climate change and marine ecosystem of the Southern Ocean. Over the past decades, strong warming has taken place over in the southern midlatitude ocean from the surface to a great depth [Gille, 2002] particularly in the South Atlantic and South Indian Ocean sectors. Over the past 15 years, the southern midlatitudes have recorded the greatest heat content increase compared to all other latitudes [Willis et al., 2003]. These latitudes coincide with the latitudes of strong positive wind stress curl changes (Figure 2 (top)). The cause for the unusually strong warming remains unknown. Our result provides a plausible explanation: the positive wind stress curl trend drives an intensifying, southward shifting super-gyre circulation leading to a greater southward influx of warm water in all three oceans, and contributing to the greater rate of warming.

[13] There are no adequate observations of ocean column warming for the period examined to reveal the spatial pattern of change. However, one would expect that a positive wind stress curl change over the interior ocean will lead to warming of the ocean column. In fully coupled climate models with climate change forcing, such warming features are indeed produced in association with an upward trend of the model SAM [Cai et al., 2005].

[14] Comparisons of data from recent Argo floats and other observations dated back 12 years ago reveal that there is a spin-up of the South Pacific gyre circulation [Roemmich et al., 2006], and that it is strongly influenced by wind stress curl at midlatitudes associated with the SAM over the same period. The intensification of the South Pacific gyre circulation manifests as an increasing trend in sea surface heights around New Zealand [Qiu and Chen, 2006]. Our results suggest that these circulation trends have started earlier.

[15] The ocean circulation changes presented here may be responsible for the substantial changes in the boundaries of the South Pacific marine biodiversity over the past several decades. The New South Wales native sea urchin, Centrostepphanus rodgersii, off Australia's east coast, has been extending its range to the Tasmanian east coast since the late 1970s [Edgar, 1997], as has been the introduced shore crab, Carcinus maenas, from Victoria to Tasmania [Thresher et al., 2003]. Other examples of the poleward spread include a number of subtropical fish that have extended southward their habitat to north-east Tasmania [Pittock, 2003]. These changes are consistent with the expected response of these species to the large warming and to an intensifying EAC flow through the Tasman Sea. Given the annular scale of the circulation change, we expect a far broader impact on the marine ecosystem beyond the South Pacific.

[16] One of the most robust features of the SH response to increasing CO2 simulated by climate models is an upward trend of the SAM [Fyfe et al., 1999; Kushner et al., 2001; Cai et al., 2003], with wind and wind stress curl changes similar to those observed in the past decades. Over the next fifty years, anthropogenic emission of ozone depleting gases is projected to reduce while CO2 is projected to increase. If the net effect of a slow ozone recovery and a faster pace of CO2 increase leads to a continuation of the observed trend over the past decades, then the ocean circulation changes similar to what we have described here will continue into the future [Cai et al., 2005].

5. Conclusions

[17] The results of this study demonstrate that changes in atmospheric surface flows over the past decades have induced a southward intensification of the Southern Ocean super-gyre circulation, including a strengthening of the EAC flow through the Tasman Sea. Our results offer a plausible explanation for the unusually large warming [Gille, 2002] of the southern midlatitude ocean over the past decades and for the observed changes in boundaries of marine biodiversity in the South Pacific [Pittock, 2003; Koslow and Thresher, 1999; Edgar, 1997; Thresher et al., 2003]. The findings reported here suggest that if stratospheric ozone depletion is indeed the primary cause for the wind changes since the late 1970s [Thompson and Solomon, 2002; Gillett and Thompson, 2003], then it has a vertically-integrated climate impact from the stratosphere through to the ocean with significant impacts on marine ecosystem.


[18] I thank Susan Solomon for her generous comments, John Church, John Gould, Paul Holper, David Karoly, Vincent Lyn, Barrie Pittock, Ken Ridgway, Steve Rintoul, and Ron Thresher for useful discussion, and Tim Cowan, Ge Shi, and Stuart Godfrey for their help. This work is part of the CSIRO's Wealth from Oceans Flagship and is partially funded by the Australian Greenhouse Office.