Bubbles from the deep

A study suggests that hydrocarbons released from sedimentary basins formed part of a climatic feedback mechanism that exacerbated global warming during the Eocene epoch.

Gas released from sedimentary basins has been proposed to be a key player in many of the rapid climate changes that occurred in the past 250 million years. The types and sources of the released gases are still debated, but they probably included gases formed by the heating of organic matter around hot magma1, and gases released by the dissociation of gas hydrates (solid compounds that trap gas molecules) found in deep-ocean sediments2. The influence of hydrates is greater in a warming world because higher ocean temperatures can melt methane-bearing gas hydrates, releasing the gas. The methane can then be oxidized to carbon dioxide, causing ocean acidification, and can contribute to global warming because both methane and carbon dioxide are greenhouse gases. Writing in Geophysical Research Letters, Kroeger and Funnell3 suggest another way in which global warming can lead to increased gas emissions from the sea floor.

Sedimentary basins have large accumulations of biological, inorganic and clastic deposits (which consist of fragments of pre-existing rocks), and are host to more than 99.9% of the organic carbon in Earth's crust. This amounts to a staggering 15,000,000 gigatonnes (Gt) of carbon; for comparison, 3,300 Gt of carbon are stored in all known hydrocarbon and coal reserves4. Any viable mechanism for transferring significant quantities of sediment-bound carbon to the atmosphere on a short timescale may thus perturb the global carbon cycle and lead to global warming.

A series of such perturbations has been suggested5,6 to have happened in the Eocene epoch (which ran from 55.8 million to 33.9 million years ago) to explain several short-lived climatic anomalies that occurred during a period of otherwise steady warming from 58 to 50 Myr ago (Fig. 1). The most prominent of these events was the Palaeocene–Eocene thermal maximum (PETM), which was characterized by global warming of 5–10 °C and subtropical conditions in the Arctic. One cause of the PETM is commonly assumed to have been methane release to the atmosphere that was triggered either by the melting of gas hydrates2 or by the heating of rocks rich in organic material in sedimentary basins (caused by widespread volcanic activity in the northeast Atlantic region1,7).

Figure 1: Ancient climate change.

The graph depicts ocean temperatures during the Palaeocene and early Eocene epochs, estimated from the abundance of oxygen isotopes in microfossils5,6. A warming trend from about 58 to 50 Myr ago was punctuated by short-lived climatic anomalies known as hyperthermals, of which the Palaeocene–Eocene Thermal Maximum (PETM) was the greatest. The warmest climate occurred during the Early Eocene Climatic Optimum (EECO) about 52 to 50 Myr ago, after which climate cooling occurred. The PETM may have been caused in part by the release of methane to the atmosphere, triggered by the melting of gas hydrates (solids that trap gases in ocean sediments) and/or by the heating of organic matter by intruding magma in sedimentary basins. Kroeger and Funnell3 propose that the production of hydrocarbons (petroleum and natural gas) in sedimentary basins increased during and after the PETM, and that seepage of hydrocarbons from such basins into the ocean and atmosphere contributed to global warming.

The extended period of warming during the Eocene culminated in the Early Eocene Climatic Optimum (EECO) 52 to 50 Myr ago, the hottest prolonged climatic episode since the Cretaceous period 145.5 to 65.5 Myr ago. Kroeger and Funnell3 focus on the EECO in their work. Using computer models of four sedimentary basins in the southwest Pacific, they simulated the increase in hydrocarbon generation in the basins as the warming Eocene oceans transferred heat to the sea floor and to deep-seated, organic-rich rocks.

The sedimentary basins studied by the authors all contained rocks that could produce petroleum given the right temperature conditions and sufficient time. Organic matter within the sedimentary rocks can convert into petroleum at temperatures of 60–120 °C (the oil window), or into predominantly natural gas at temperatures of 100–200 °C (the gas window). On the basis of their models, Kroeger and Funnell estimate that the warming ocean during and after the PETM eventually raised the temperature of a 300-metre-thick layer of sediments into the oil and gas windows. A marked rise in hydrocarbon production would therefore have occurred in the 4 to 5 Myr following the PETM, peaking during the EECO.

The authors' results help us to understand the dynamics and temperature-dependence of hydrocarbon (petroleum and gas) generation in sedimentary basins. But there is more to the story than that. If oil and gas generated during the Eocene somehow escaped the sedimentary basins and leaked out to the ocean and the atmosphere in sufficient quantities, this may have contributed to global warming at the time. In other words, Kroeger and Funnell propose a climate feedback mechanism: global warming causes increased hydrocarbon production that leads to prolonged global warming.

The idea that the conversion of organic matter into petroleum following a temperature increase in sedimentary basins may form part of a climate feedback mechanism was first posited8 by Kroeger and colleagues last year. The modelling in Kroeger and Funnell's present work3 remarkably predicts and quantifies a peak in hydrocarbon production that overlaps in time with the EECO. The authors' results show that about 37 Gt of oil and 8 Gt of gas were generated from the four basins during this period, which is 50% more than would have occurred in the absence of extra ocean warming. This alone would not have been sufficient to affect the Eocene climate9, but many other basins around the world would presumably have increased in temperature at the same time. The mass of hydrocarbons generated globally could therefore have been considerably higher than that predicted in the authors' study, although the exact amount is difficult to quantify.

Is hydrocarbon seepage at the sea floor on a climate-changing scale a realistic possibility? Kroeger and Funnell argue that it is, because more oil and gas are generated in sedimentary basins than are trapped10. Studies11 of sedimentary basins around the globe have shown that gas seepage is currently a common phenomenon, and there is no reason why this should not have been the case earlier in Earth's history12.

But did the hydrocarbons predicted3 to have been generated during the Eocene really leak out to the ocean and atmosphere, or did they stay trapped in the subsurface? A way to test this is to search for carbonate deposits in old sea-floor sediments, because sea-floor hydrocarbon seepage leaves behind deposits that have distinct geochemical signals that can be attributed to their origin13. Unfortunately, the carbonate record is currently too poorly investigated to confirm the extent of seepage from sedimentary basins during the Eocene, and so more studies are needed. Nevertheless, one thing seems clear: the transfer of carbon from sedimentary rocks to the atmosphere is an important component of climate change, both past and future.


  1. 1

    Svensen, H. et al. Nature 429, 542–545 (2004).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Dickens, G. R., O'Neil, J. R., Rea, D. K. & Owen, R. M. Paleoceanography 10, 965–971 (1995).

    ADS  Article  Google Scholar 

  3. 3

    Kroeger, K. F. & Funnell, R. H. Geophys. Res. Lett. (2012).

  4. 4

  5. 5

    Zachos, J. C., Pagani, M., Sloan, L., Thomas, E. & Billups, K. Science 292, 686–693 (2001).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Sluijs, A., Bowen, G. J., Brinkhuis, H., Lourens, L. J. & Thomas, E. in Deep Time Perspectives on Climate Change: Marrying the Signal from Computer Models and Biological Proxies (eds Williams, M., Haywood, A., Gregory, J. & Schmidt, D.) 267–293 (Geol. Soc. Lond., 2007).

    Google Scholar 

  7. 7

    Storey, M., Duncan, R. A. & Swisher, C. C. III Science 316, 587–589 (2007).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Kroeger, K. F., di Primio, R. & Horsfield, B. Earth Sci. Rev. 107, 423–442 (2011).

    ADS  CAS  Article  Google Scholar 

  9. 9

    Cui, Y. et al. Nature Geosci. 4, 481–485 (2011).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Kvenvolden, K. A. & Cooper, C. K. Geo-Mar. Lett. 23, 140–146 (2003).

    ADS  CAS  Article  Google Scholar 

  11. 11

    Serié, C. S., Huuse, M. & Schødt, N. H. Geology (2012).

  12. 12

    Berndt, C. Phil. Trans. R. Soc. A 363, 2855–2871 (2005).

    ADS  Article  Google Scholar 

  13. 13

    Mazzini, A., Aloisi, G., Akhmanov, J., Parnell, B. & Murphy, P. J. Geol. Soc. Lond. 162, 815–827 (2005).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Henrik Svensen.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Svensen, H. Bubbles from the deep. Nature 483, 413–415 (2012).

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.