Australia: The Land Where Time Began

A biography of the Australian continent 

Antarctic Climate Change During the Last Interglacial and Local Orbital Forcing

The climate of the Antarctic region changed during the last interglacial before that of the Northern Hemisphere. According to Kim et al. large local changes in precession forcing could have produced this pattern if a rectified response in the cover of sea ice had occurred. When a coupled sea ice-ocean general circulation model was tested for 3 intervals around the last Interglacial the results supported this hypothesis. It is suggested that such a mechanism may play an important role in contributing to phase offsets between climate change in the Northern and Southern hemispheres for other intervals.

One of the problems in climatology of the Pleistocene involves the factors responsible for climate change in Antarctica. The processional component of orbital forcing is almost out of phase with between the Northern Hemisphere and the Southern Hemisphere, though variations in orbital insolation play a major role in driving climate changes in the Pleistocene (1), so any conditions that favour glaciation and deglaciation in the Northern Hemisphere should result in the opposite response in the Southern Hemisphere. It has been known for more than 20 years is that though cooling in the Southern Hemisphere during the Pleistocene accompanies glaciation in the Northern Hemisphere, Southern Hemisphere climate led climate in the Northern Hemisphere into and out of the last, as well as other, glaciations. I.e., the Southern Hemisphere warmed and cooled before the Northern Hemisphere (1). Before the glacial retreat in the Northern Hemisphere carbon dioxide also increased (2). Yet it is rare for standard explanations for climate change in the Southern Hemisphere to focus on local changes in forcing around Antarctica. According to Kim et al. most explanations involve processes that are more remote such as changes the atmospheric concentrations of CO₂ (3), changes in North Atlantic Deep Water  (NADW) transport of heat to the Antarctic  (4), or the lowering if sea level leading to expansion of the Antarctic ice sheet. There is, however, a modest contribution of about 1^oC from the mean annual changes in the local radiation budget at the highest latitudes as a result of synchronous Northern Hemisphere-Southern Hemisphere obliquity changes at the 41,000 year period (5).

In this paper Kim et al. show that local forcing at the precessional period (19,000 and 23,000 years), which is out of phase between the Northern Hemisphere and the Southern Hemisphere climate changes. The study of Kim et al. was based on the hypothesis that seasonal changes in the Antarctic summer may be more important, proportionally, than in the Antarctic winter because sea ice is much closer to the freezing point in summer. This relation may allow a rectified response to variations in orbital insolations and may account for some of the phase offsets in climate change between the Northern Hemisphere and the Southern Hemisphere. Their model was tested with a coupled sea ice-ocean general circulation model (OGCM).

Kim et al. used the Hamburg Ocean Primitive Equation (HOPE) model (6), which is based on the primitive equations, with a prognostic free surface (7). Based on the Arakawa-E grid (8), the equations are discretised, and the model has horizontal resolution of a 3.5^o x 3.5^o, with 11 vertical layers. Included in the model is a comprehensive dynamic-thermodynamic sea ice model (9). Climatological monthly mean winds force the ocean (10), apart from the sea ice of the Southern Ocean, which is forced by daily winds from the European Centre for Medium Range Weather Forecast analyses. The treatment of the temperature of the surface and salinity is dependent on the presence of sea ice. In grid cells that are ice free, sea surface temperature and salinity are relaxed to air temperature (11) and salinity (12) that are prescribed.

Kim et al. used a linear version of a linear version of an energy balance model (EBM) (13) in order to obtain the surface temperature response to the change in orbital insolation in the Pleistocene. The temperature response to seasonal insolation forcing as it is modified by geography by the EBM, which is a 2-D model. It is indicated by many sensitivity experiments (14) that its response to orbital insolation changes is approximately the same as that of atmospheric general circulation models, though the EBM is a simplified model. Kim et al. chose 3 time periods: 106 ka and 125 ka, at which local summer insolation is at a minima (15), and 135 ka, at which the local summer insolation was at a maximum. These time intervals were chosen because the CO₂­ and temperature increased before the Northern Hemisphere ice sheet melted (2,16), and temperature then decreased before the ice growth in the Northern Hemisphere (1). In the Southern Hemisphere the mid- to high latitudes cooled almost as much as the as at the glacial maximum at 106 ka. It is suggested by earlier linear EBM calculations that the local orbital forcing could play an important role in phase shifts and seasonal cooling for these time periods, though in order to translate the forcing into mean annual temperature changes, such as are estimated for the Vostok site (17), some feedback would be required.

In the Southern Hemisphere summer is the time when the EBM temperature response occurred. In January at about 80^oS, simulated temperature at 106 ka and 125 ka where about 3.9^oC and 1.8^oC lower than at present, respectively, whereas at 135 ka the temperature was about 1.4^oC higher than at present. These temperature differences were imposed by Kim et al. on the climatological atmospheric forcing of sea ice-ocean model in order to investigate the change in the area covered by Antarctic ice and the overturning circulation. In order to isolate the mid- to high latitude response of the Southern Hemisphere to changes in orbital insolation the temperature changes were imposed only south of 45^oS. The model reached a quasi-steady state after 600 years of integration; i.e., at all model levels the climate drift in temperature and salinity was within the range of 0.01^oC and 0.001 psu per century.

At present the area covered by sea ice in the Antarctic varies from about 4.5 x 10⁶ km² in the austral summer to 17 x 10⁶ km² in the austral winter. In the model the mean annual area of Antarctic ice increased by 3.2 x 10⁶ km² and 1 x 10⁶ km², respectively, and at 135 ka it decreased by about 1.1 x 10⁶ km². In the austral summer, October to April, the largest difference occurred, with a change of +80%, +40%, and -40% for 106 ka, 125 ka ND 135 ka respectively. It is indicated by model time series (19) that there is a systematic offset from the control run in the area of sea ice; i.e., the differences do not appear to reflect model drift or centennial variability. A mean annual cooling was also obtained for this region in a coupled model run (20), though the full (global) orbital insolation change was used in that experiment for this time interval.

Changes also occurred in the thermohaline circulation that was modelled. The intrusion of Antarctic bottom water (AABW) across the equator is about 12 sverdrups (1 Sv = 1 x 10⁶ m³/sec), which is close to the flow that was obtained in a previous model study (21) and it compares well with recent estimates (22,23). The formation at the present of NADW (36 Sv) is overestimated by the model verses recent estimated from sections that were observed (27 Sv) (23), though at 30^oS overflow of NADW (19) is 17 Sv, which is realistic. The intrusion of Antarctic Bottom Water increased by 3.3 Sv for the 106 ka simulation, while North Atlantic Bottom Water production decreased by 1.2 Sv. Intrusion of the Antarctic Bottom Water increased by 1.5 Sv at 125 ka and decreased slightly by 0.6 Sv at 135 ka. The largest change in the deep Antarctic outflow is recorded by the Pacific Basin, though this response is suggested by Kim et al. to possibly model dependent.

Geochemical data has been interpreted as an indication that variations in the North Atlantic Deep Water contributed to meltback of Antarctic sea ice during terminations (4). The role of North Atlantic Deep Water in terms of forcing Southern Hemisphere climate has been challenged (24). It is suggest by the results of the study by Kim et al. that the changing outflow of deep water from the Antarctic could change the relative abundance of the northern component water at any particular site, which leads to potential interpretations of past variations of the North Atlantic Deep Water.

The results of Kim et al. therefore support the hypotheses that the response of the sea ice to local changes in Milankovitch forcing could affect both the timing and magnitude of climate change in the Southern Ocean. Modelled changes in sea ice could also affect concentrations of CO₂ (25), which would then further modify sea ice (3). At about 130-135 ka this latter response might be particularly relevant to the early rise in CO₂, though additional feedbacks would be required to amplify the modest response obtained by Kim et al. for this time interval. E.g., weaker winds in the Southern Ocean (26), that are caused by reduced sea ice should decrease loss of heat from the ocean (27) and increase sea surface temperatures. Kim et al. suggest these problems would have to be addressed with models that are fully coupled, which at present do not simulate high latitude climates in the Southern Hemisphere well (28) and are too computer-intensive to be used for a series of quasi-equilibrium sensitivity experiments.

Sources & Further reading

Kim, S.-J., et al. (1998). "Local Orbital Forcing of Antarctic Climate Change During the Last Interglacial." Science 280(5364): 728-730.

1. J. D. Hays, J. A. Imbrie, N. J. Shackleton, Science

194, 1121 (1976); J. Imbrie et al., in Milankovitch and

Climate, A. Berger et al., Eds. (Reidel, Dordrecht,

Netherlands, 1984), pp. 269–305; D. G. Martinson

et al., Quat. Res. 27, 1 (1987); J. Imbrie, A. McIntyre,

A. Mix, in Climate and Geosciences: A Challenge for

Science and Society in the 21st Century, A. Berger,

J.-C. Duplessy, S. H. Schneider, Eds. (Reidel, Hingham,

MA, 1989), pp. 121–164; L. Labeyrie et al.,

Paleoceanography 11, 57 (1996).

2. T. Sowers et al., Paleoceanography 8, 737 (1993); T.

Sowers and M. Bender, Science 269, 210 (1995).

3. S. Manabe and A. J. Broccoli, J. Atmos. Sci. 42,

2643 (1985).

4. P. K. Weyl, Meteorol. Monogr. 8, 37 (1968); J. Imbrie

et al., Paleoceanography 8, 699 (1993).

5. D. A. Short et al., Quat. Res. 35, 157 (1991).

6. J. Wolff, E. Maier-Reimer, S. Legutke, Report No. 13,

Deutsches Klimarechenzentrum (1997), p. 81.

7. S. Drijfhout, C. Heinze, M. Latif, E. Maier-Reimer, J.

Phys. Oceanogr. 26, 559 (1996).

8. A. Arakawa and V. R. Lamb, Methods Comput.

Phys. 17, 173 (1972).

9. The sea ice dynamics adopt the viscous-plastic constitutive

law of W. D. Hibler III [J. Phys. Oceanogr. 9,

815 (1979)]; sea ice thermodynamics are adopted

from W. B. Owens and P. Lemke [J. Geophys. Res.

95, 9527 (1990)].

10. S. Hellerman and M. Rosenstein, J. Phys. Oceanogr.

13, 1093 (1983).

11. S. D. Woodruff, R. J. Slutz, R. L. Jenne, P. M.

Streurer, Bull. Am. Meteorol. Soc. 68, 1239 (1987).

12. S. Levitus, NOAA Prof. Pap. 13 (1982).

13. G. R. North, J. G. Mengel, D. A. Short, J. Geophys.

Res. 88, 6576 (1983).

14. T. J. Crowley, D. A. Short, J. G. Mengel, G. R. North,

Science 231, 579 (1986); W. T. Hyde, T. J. Crowley,

K.-Y. Kim, G. R. North, J. Clim. 2, 864 (1989); T. J.

Crowley, S. K. Baum, W. T. Hyde, J. Geophys. Res.

96, 9217 (1991).

15. A. L. Berger and M. F. Loutre, Quat. Sci. Rev. 10,

297 (1991).

16. J. M. Barnola, D. Raynaud, Y. S. Korotkevich, C.

Lorius, Nature 329, 408 (1987).

17. J. Jouzel et al., ibid., p. 403; J. Jouzel et al., ibid. 364,

407 (1993).

18. This is slightly overestimated over passive microwave

observation [P. Gloersen et al., Arctic and Antarctic

Sea Ice, 1978–1987 (Scientific and Technical

Information Program, NASA, Washington, DC,

1992), p. 290], which shows that the Antarctic sea

ice varies from ;2 3 106 km2 in austral summer to

;15 3 106 km2 in austral winter.

19. S.-J. Kim, T. J. Crowley, A. Sto¨ ssel, data not shown.

20. M. Montoya, T. J. Crowley, H. von Storch, Paleoceanography

13, 170 (1998).

21. W. Cai and P. G. Bains, J. Geophys. Res. 101,

14073 (1996); A. Sto¨ ssel, S.-J. Kim, S. Drijfhout, J.

Phys. Oceanogr., in press.

22. M. M. Hall, M. McCartney, J. A. Whitehead, J. Phys.

Oceanogr. 27, 1903 (1997).

23. A. M. Macdonald and C. Wunsch, Nature 382, 436

(1996).

24. T. J. Crowley, Paleoceanography 7, 489 (1992); T.

Blunier et al., Geophys. Res. Lett. 24, 2683 (1997).

25. J. R. Toggweiler and J. L. Sarmiento, in The Carbon

Cycle and Atmospheric CO2: Natural Variations Archean

to Present, E. T. Sundquist and W. S. Broecker,

Eds. (American Geophysical Union, Washington,

DC, 1985), pp. 163–184; R. Franc¸ ois et al.,

Nature 389, 929 (1997).

26. M. DeAngelis, N. I. Barkov, V. N. Petrov, Nature 325,

318 (1987).

27. T. J. Crowley and C. L. Parkinson, Clim. Dyn. 3, 93

(1988).

28. J.-S. von Storch et al., J. Clim. 10, 1525 (1997); a

climate drift also occurs in the National Oceanic and

Atmospheric Administration/Geophysical Fluid Dynamics

Laboratory–coupled model [S. Manabe and R. J.

Stouffer, ibid. 9, 376 (1996)], but the reason has not

been identified.

29. Supported by NSF grants OCE96-16977 and ATM

95-29109 and funding from Texas A&M University.

19 December 1997; accepted 16 March 1998