From the timing, it looks as if an episode of marked oceanic oxygen deficiency during the Cretaceous was the result of undersea volcanism. Studies of such events are relevant to the warming world of today.
About 93 million years ago, Earth was shaken by an immense episode of volcanism. Massive piles of highly fluid lava accumulated under the seabed, forming much of the present-day Caribbean region in a geological heartbeat. On page 323 of this issue1, Turgeon and Creaser argue persuasively that these eruptions triggered an episode of oceanwide anoxia that led to mass extinctions. Moreover, at this time Earth was so hot that palm trees grew in Alaska and large reptiles roamed northern Canada. So this new work can provide valuable lessons about the way Earth responds to perturbations akin to those it is experiencing now.
The stack of lava flows that formed the Caribbean tectonic plate is known as a large igneous province (LIP)2,3. Other LIPs altered Earth's history by causing profound changes in the composition of the atmosphere: by increasing carbon dioxide levels and causing greenhouse warming; by altering the chemistry of sea water and ocean circulation; and by disrupting the global carbon cycle4. A LIP in Siberia that erupted 250 million years (Myr) ago led to the largest mass extinction ever, at the end of the Permian period5. Another event 120 Myr ago, the formation of the massive Ontong Java Plateau in the western Pacific Ocean6, led to extinctions of ocean plankton. Both of these LIPs caused the ocean circulation to slow, decelerating the cycling of carbon and oxygen, and ultimately leading to highly toxic, anoxic conditions on the sea floor.
Episodes of anoxia, known as oceanic anoxic events (OAEs), have occurred periodically during Earth's history, but none was more severe than that which occurred 93 Myr ago, during the Cretaceous period7,8. This OAE caused the extinction of large clams known as inoceramids and tiny protists called foraminifera that lived on the sea floor. Profound changes in ocean circulation also led to the production and preservation of enormous quantities of marine organic material that was subsequently transformed into oil during its burial. But the ultimate cause of the OAE has proved elusive. There is strong evidence that warming was involved, implicating greenhouse conditions9. Unusual metal enrichments in rocks deposited during the OAE suggest a link to volcanism10,11, but rocks that are organic-rich are often metalliferous as well, so this link remains unconfirmed.
The Caribbean lavas are now deeply buried in the ocean or found in mountain belts in places such as Haiti, where they have been exhumed during tectonic activity. Their eruption history is not as well known as that of the Siberian and Ontong Java LIPs, which are more widely sampled. The lavas' ages, measured using radioactive isotopes, span the interval 87–95 Myr ago, with suggestions that a large pulse occurred 93–94 Myr ago12. But given that the Caribbean is hundreds of thousands of cubic kilometres in extent, it is impossible to determine its eruption history with only a few age estimates. Instead, geologists must rely on other chemical signals in sedimentary rocks that provide continuous proxies for volcanic intensity through time.
Turgeon and Creaser1 used a sensitive proxy for volcanism based on two isotopes of osmium, 187Os and 188Os. The main sources of osmium to the ocean include detritus from rivers, volcanism and material from outer space such as meteorites. Thus, the osmium isotope composition of sea water is a measure of the weathering of the continents, volcanic activity and extraterrestrial input13. The abundance of both osmium isotopes in sea water is low, so their ratios adjust rapidly to changes in input.
Turgeon and Creaser's careful analysis of these isotopes in sedimentary rocks from drill cores off the coast of South America, and from mountains in Italy, provides clear evidence of a perturbation that immediately preceded the OAE 93 Myr ago. The isotope values show an unmistakable increase in the osmium contribution from a meteoritic or volcanic source. In the absence of other evidence for an extraterrestrial impact at this time, the data clearly point to a volcanic episode. And given that the shift in osmium isotopes suggests a factor of 30–50 increase in the osmium flux to the oceans, that episode was evidently on a huge scale. Although there are several candidates, the only LIP close to this age that is large enough to have caused this type of perturbation is in the Caribbean.
The formation of the Caribbean is of great interest to geologists, but it also has much broader implications. Abrupt warming events in the geological record are of great significance for scientists working on modern global warming. The ancient episodes provide a complete picture of the processes operating during various stages of a global warming event. For example, at some stage photosynthetic plankton will draw down CO2 from the atmosphere, producing organic matter that will be buried in rocks, and leading to global cooling that possibly heralds the end of the warming event8,9. In this regard, the event of 93 Myr ago can serve as a test run for the possibility of seeding the modern ocean with nutrients to promote photosynthesis and CO2 reduction.
But a better understanding of that event and its relevance to modern global change requires knowledge of the scale of the volcanism (and so the rate and amount of CO2 input), and how it is related to the OAE. These questions are only partially addressed by Turgeon and Creaser1. They show that there is a temporal relationship between the LIP and the OAE that implies a causal connection, but they offer no proof for the nature of that connection.
One possibility is that the large amounts of metal-rich fluids produced during volcanism may have seeded the ocean with micronutrients such as iron, stimulating plankton to produce large quantities of organic matter11. Oxidation of this organic matter would have stripped the ocean of oxygen, leading to the OAE (Fig. 1). Whether such a process alone could cause a global anoxic event has yet to be determined, but I remain doubtful that it could. Perhaps that effect was exacerbated by oceanic stratification (Fig. 1), a result of the warming produced by volcanic CO2 that severely inhibited mixing in the ocean.
Especially significant is the time lag between the initial climatic perturbation, produced by the emission of huge quantities of CO2 into the atmosphere, and its long-term consequences. Based on the rate at which the sedimentary rocks are known to have accumulated, Turgeon and Creaser1 estimate that the volcanic pulse preceded the OAE by 23,000 years. Although the lag between trigger and catastrophe might be heartening to those concerned about the impacts of global warming, the time it takes signals to mix in the oceans today (about 1,500 years) implies that the lag time was shorter; other abrupt warming events in the geological record have similar, shorter lags. If the response time to the LIP volcanism was indeed longer, it suggests either that ocean circulation was extraordinarily sluggish before the event, or that there was a complicated connection between the trigger and the OAE. Determining how the volcanism of 93 Myr ago wreaked havoc on the warm Earth is another challenge for geologists endeavouring to understand the complex behaviour of our planet.
Turgeon, S. C. & Creaser, R. A. Nature 454, 323–326 (2008).
Coffin, M. F. & Eldholm, O. Rev. Geophys. 32, 1–36 (1994).
Kerr, A. C. in Treatise on Geochemistry Vol. 3 (ed. Rudnick, R. L.) 537–565 (Elsevier, 2003).
Larson, R. L. Geology 19, 547–550 (1991).
Bowring, S. A. et al. Science 280, 1039–1045 (1998).
Tarduno, J. A. et al. Science 254, 399–403 (1991).
Jenkyns, H. C. J. Geol. Soc. Lond. 137, 171–188 (1980).
Arthur, M. A., Dean, W. E. & Pratt, L. M. Nature 335, 714–717 (1988).
Forster, A., Schouten, S., Moriya, K., Wilson, P. A. & Sinninghe Damsté, J. S. Paleoceanography 22, PA1219, doi:10.1029/2006PA001349 (2007).
Orth, C. J. et al. Earth Planet. Sci. Lett. 117, 189–204 (1993).
Snow, L. J., Duncan, R. A. & Bralower, T. J. Paleoceanography 20, PA3005, doi:10.1029/2004PA001093 (2005).
Sinton, C. W. & Duncan, R. A. Econ. Geol. 92, 836–842 (1997).
Peucker-Erhenbrink, B. & Ravizza, G. Terra Nova 12, 205–219 (2000).
About this article
The Cenomanian–Turonian boundary in the northwestern part of the Adriatic Carbonate Platform (Ćićarija Mtn., Istria, Croatia): characteristics and implications
Earth and Planetary Science Letters (2017)
Cenomanian-Coniacian Sea-level Change and Dissolved Oxygen Fluctuations in Tethys-Himalaya: Evidences from Benthic Foraminifera of Gamba, Tibet
Acta Geologica Sinica - English Edition (2013)
Latitudinal differences in the amplitude of the OAE-2 carbon isotopic excursion: pCO2 and paleo productivity