Low-down on a rhythmic high

Changes in the amount of solar energy reaching Earth account for certain climate cycles at high and low latitudes. Surprisingly, the effects of a high-latitude cycle evidently reached into the tropics.

During the 1920s, Milutin Milankovitch, a Serbian mathematician, calculated the effects of alterations in Earth's motion around the Sun on the amount of solar energy reaching different latitudes1. Since then, some of the long-period, cyclic changes seen in archives of past environmental conditions on Earth have been explained by these changes in insolation. For example, a 41,000-year cycle in a climate record is commonly ascribed to changes in the planet's tilt, which affects insolation at high latitudes, and a 19,000–23,000-year cycle is ascribed to changes in the planet's wobble, which dominates insolation at low and middle latitudes2,3.

Eighty years after Milankovitch's innovative work, Liu and Herbert (page 720 of this issue4) have unearthed surprising evidence that the views based on his conclusions are incomplete. The authors provide a record of low-latitude sea surface temperature (SST) that is in phase with the 41,000-year rhythm reminiscent of high-latitude insolation. Although their record represents only the past 1.8 million years, a small portion of Earth's history, their findings force us to think further about a climate system that we already knew to be complex.

How does the geological record document past climate change? Perhaps the most ubiquitous recorders of past climate change are the fossil shells of foraminifera, calcareous marine organisms. To a large degree, the oxygen-isotope ratio of the calcium carbonate shell of foraminifera reflects the oxygen-isotope ratio in sea water, which in turn is a function of the hydrologic cycle and thus the amount of fresh water stored as ice at the poles. Because ice-sheet growth and decay are affected by cycles in insolation, the H218O/H216O ratio of sea water varies at the same frequency as does the 18O/16O ratio of the carbonate shells of the foraminifera in the world ocean. Over time, foraminifers accumulating on the bottom of the ocean build an archive of information about glacial to interglacial climate change.

These types of record provide the backbone of palaeoclimate studies. For example, over the past 5 million years perhaps the most profound change in Earth's climate history was the onset of glaciation in the Northern Hemisphere, recorded by the increase in foraminiferal 18O/16O ratios between 3 million and 2 million years ago (Fig. 1). However, the isotopic composition of sea water, and hence global ice extent, is not the only variable determining foraminiferal 18O/16O ratios; water temperature affects the preference of one isotope over the other incorporated in calcium carbonate during calcification. So although foraminiferal 18O/16O ratios by themselves do provide information about large-scale climate change as described above, they cannot contribute unequivocally to our understanding of the relationship between individual climate variables such as ice volume and water temperature.

Figure 1: Global climate change from 5 million years ago to the present.

This record is inferred from the foraminiferal oxygen-isotope archive from the eastern equatorial Pacific12, with 18O/16O ratios plotted as a per mil deviation from a standard — δ18O(‰). Over this time there has been a general trend towards cooler climate; at the moment, we are in a peak warm interval within the overall cooling trend. The increase in foraminiferal δ18O values between 3 million and 2 million years ago signifies the (poorly understood) growth of ice sheets in the Northern Hemisphere. Liu and Herbert4 have constructed a separate record for the eastern equatorial Pacific from 1.8 million years ago (shaded), and provide evidence that cyclic processes operating at high latitudes produce a surface-ocean response at low latitudes.

Against the backdrop of global climate change, Liu and Herbert4 introduce an SST record from the eastern equatorial Pacific Ocean that spans the past 1.8 million years. The data are derived from measurements of the biochemistry of certain species of marine phytoplankton that are preserved in deep-sea sediments. At the molecular level, the structure of these organisms is a function of temperature5. The record generated by Liu and Herbert thus provides a unique opportunity to set SST variations in the eastern equatorial Pacific in the context of global climate change — as, for instance, assessed by the oxygen-isotope composition of foraminifera shown in Fig. 1.

Liu and Herbert are not the first to apply this ‘palaeothermometer’ to the geological record6,7,8, but they provide the first tropical SST record to go as far back as 1.8 million years. This record shows strong 41,000-year fluctuations in its early part (1.8–1.2 million years ago). Liu and Herbert observe that the fluctuations correlate closely with variations in foraminiferal 18O/16O ratios and, hence, global ice volume. This is perhaps a result of feedbacks between ice extent and the amount of insolation being reflected from Earth's surface. But as the authors point out, SST minima just precede ice-volume maxima (and vice versa), which means that the ice sheet cannot be the initial driving mechanism. Liu and Herbert allow that it is surprising that this connection is strongest between 1.8 million and 1.2 million years ago, when Northern Hemisphere ice sheets were still relatively small. Importantly, however, SST minima and maxima in the eastern equatorial Pacific are coincident with minima and maxima in insolation at high latitudes, and thus more suggestive of a coupling between the low and high latitudes via circulation in the atmosphere or upper ocean, or both. Regardless of the mechanism, Liu and Herbert present new evidence for a high-latitude signature in a low-latitude climate record.

These results are particularly intriguing in light of the search for the causes of glaciation in the Northern Hemisphere between 3 million and 2 million years ago (Fig. 1). Current explanations focus on tectonic events that had a large effect on patterns of ocean circulation, such as the closure of the Central American9 and Indonesian10 seaways. It has also been proposed that the onset of glacial–interglacial climate cycles around 3 million years ago should have been accompanied by an increased response of the tropical oceans to changes in insolation caused by the planet's tilt; this should be recorded as 41,000-year cycles in low-latitude climate records11. Liu and Herbert's SST record from the eastern equatorial Pacific supports this idea. Ultimately, the hypothesis needs to be tested by constructing similar records that include this entire climate transition. But for the moment, Liu and Herbert have provided an enticing example of the strength of the connection between high and low latitudes.


  1. 1

    Milankovitch, M. Serb. Akad. Beogr. Spec. Publ. 132 (1941).

  2. 2

    Hays, J. D. et al. Science 194, 1121–1132 (1976).

  3. 3

    Berger, A. L. J. Atmos. Sci. 35, 2362–2367 (1978).

    ADS  Article  Google Scholar 

  4. 4

    Liu, Z. & Herbert, T. D. Nature 427, 720–723 (2004).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Wefer, G., Berger, W. H., Bijma, J. & Fischer, G. in Use of Proxies in Paleoceanography (eds Fischer, G. & Wefer, G.) 1–68 (Springer, Berlin, 1999).

    Google Scholar 

  6. 6

    Emeis, K. C., Doose, H., Mix, A. & Schulz-Bull, D. Proc. ODP Sci. Results 138, 605–614 (1995).

    Google Scholar 

  7. 7

    Sikes, E. & Keigwin, L. D. Paleoceanography 9, 31–45 (1994).

    ADS  Article  Google Scholar 

  8. 8

    Sachs, J. P. & Lehman, S. J. Science 286, 756–759 (1999).

    CAS  Article  Google Scholar 

  9. 9

    Driscoll, N. W. & Haug, G. H. Science 282, 436–438 (1998).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Cane, M. A. & Molnar, P. Nature 411, 157–162 (2001).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Philander, S. G. H. & Fedorov, A. V. Paleoceanography 18, 10.1029/2002PA000837 (2003).

  12. 12

    Mix. A. et al. Proc. ODP Sci. Results 138, 371–412 (1995).

    Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Billups, K. Low-down on a rhythmic high. Nature 427, 686–687 (2004).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.