New paper: Warm Archean oceans reconstructed from oxygen isotope composition of early-life remnants

The following summary has been written by Dr Romain Tartèse, who is a visitor in the SEES University of Manchester Isotope Group.


There is strong evidence in the geological record suggesting that life started on the Earth more than 3 billion years ago. However, deciphering the conditions at the surface of the Earth at that time has remained a highly controversial topic for decades. In this new paper published in Geochemical Perspectives Letters, we have analysed the oxygen isotope composition of microfossil remnants extracted from rock samples up to 3.4 billion years old. These data have allowed us to reconstruct the oxygen isotope composition of seawater through time, which combined with that of sedimentary chemical precipitates – the so-called cherts – suggests that temperatures at the surface of the Earth have dropped by 50-60 °C over the past 3.5 billion years.

Isotopes are variants of a particular chemical element that differ in terms of their number of neutrons. The lighter element, hydrogen, has two main isotopes 1H and 2H – the former does not have any neutron while the latter has one. Oxygen has three main isotopes 16O, 17O and 18O, 16O being naturally the most abundant (there are about 2500 16O atoms for five 18O atoms for one 17O atom). Importantly, since these isotopes have slightly different masses their natural abundances can be modified during chemical and physical processes, which we refer to as ‘isotopic fractionation’.

For more than 50 years, and the seminal paper ‘The thermodynamic properties of isotopic substances’ published in 1947 by Harold Urey in the Journal of the Chemical Society, we have known that the magnitude of isotopic fractionation between a compound and the aqueous solution from which it precipitated varies with temperature. From this point, researchers realised that constraining the isotopic composition of oxygen in rocks precipitated from oceanic waters could allow them to reconstruct the temperature of the oceans throughout geological times.


Flint nodule from West Horsley, Surrey, UK. Scale bar represents 5 cm. Image from the Imperial College Rock Library

One of these rocks that has been extensively used for paleo-temperature reconstructions is chert, notably because it is found throughout the geological record in formations up to around 3.5 billion years old, and because it is relatively resistant to alteration after it has crystallised. Chert essentially comprises silica (SiO2) – flints found in Cretaceous chalk formations throughout the UK constitute a famous example of chert (see right).

Chert precipitation from silica and oxygen initially dissolved in water obeys thermodynamics and this process, therefore, induces isotopic fractionation of oxygen isotope. In other word, the 18O/16O ratio of oxygen in SiO2 is different than the initial 18O/16O of oxygen in H2O from which it formed, by an amount that depends on the temperature at which chert precipitation occurred. Thanks to a calibration proposed 40 years ago by Paul Knauth and Samuel Epstein, we know that if chert formed at 10°C its 18O/16O ratio will be around 35 ‰ (since isotopic variations are often small we commonly use a part per thousand notation – here this corresponds to 3.5 %) higher than that of the water from which it precipitated, while it would be around 20 ‰ higher if chert formed at around 80 °C.


The top panel displays the oxygen isotope composition of ‘unaltered’ chert samples up to 3.5 billion years old, and the two end-member scenarios of (i) a constant 18O/16O ratio of 0 ± 2 ‰ for seawater through time vs. (ii) a progressive increase of the seawater 18O/16O ratio through time. The bottom panel illustrates two ocean temperature reconstructions corresponding to these two end-member scenarios . Figure from Tartèse et al. (2016).

Half a century of analysis of the oxygen isotope composition of chert samples collected across the world in rocks ranging in age from the present-day up to around 3.5 billion years old shows that overall the chert 18O/16O ratios have regularly increased from ~20 ‰ to ~35 ‰ higher than the present-day average oxygen isotopic composition of the oceans (see figure at right). Taken at face value this can be interpreted as reflecting a decrease of ocean temperatures from about 70-80 °C around 3.5 billion years ago to 5-10 °C today, considering that all these cherts precipitated from seawater (Figure). However, this relies on the key assumption that the oxygen isotope composition of the oceans has not changed through time, which has never been directly measured. As a result, this interpretation of very warm surface conditions during Archean times (> 2.5 billion years old) has been challenged, notably because such elevated temperatures seem hard to reconcile with our Sun being ~20 % fainter than today 2.5 billion years ago. Thus, some have argued for an alternative scenario in which the 18O/16O ratio of the oceans has increased regularly by ~15 ‰ for the past 3.5 billion years (see figure above right). In this scenario the oxygen isotope fractionation between chert and water remains the same through time, corresponding to temperatures around 15-30 °C (see figure above right).

To provide new constrains on this hot debate, together with colleagues from the Muséum National d’Histoire Naturelle in Paris, the Institut de Physique du Globe in Paris, and the Centre de Recherches Pétrographiques et Géochimiques in Nancy, we have measured the oxygen isotope composition of organic matter residues ranging in age from 580 million years up to 3.4 billion years, and which represent remnants of bacterial life that inhabited oceans on the early-Earth, instead of that of the chert matrix that hosts them. In doing so we were hoping to get independent constraints on the evolution of the oxygen isotope composition of seawater through time. This is because contrary to thermodynamic processes, isotopic fractionation associated with biological processes seems to be largely independent on temperature. Overall, organic matter residues have consistent 18O/16O ratios through geological times, ~20 ± 5 ‰ higher than that of present-day seawater, which we argue indicates that the 18O/16O isotopic composition of seawater has globally remained constant for the past 3.5 billion years.


Estimates of water temperatures calculated using the chert oxygen isotope record for a constant seawater 18O/16O ratio through time, the chert silicon isotope record (Robert and Chaussidon, 2006), and resurrected proteins of ancient bacteria (Gaucher et al., 2008). Figure from Tartèse et al. (2016).

This new constraint precludes a progressive increase of the 18O/16O isotopic composition of the oceans through time. Therefore, the global chert oxygen isotope record likely indicates that ocean temperatures have progressively decreased by 50-60 °C for the past 3.5 billion years (see figure above). This decrease is consistent with other estimates, such as those derived from the silicon isotope composition of cherts, and the temperature of stability measured for resurrected proteins presumably akin those of Precambrian bacteria (see figure above). Importantly, temperatures calculated for the past 1.5 Ga remain below ~30 °C, which is within the range allowing development of complex eukaryotic life.

Sustaining elevated surface temperatures of around 40-60 °C during Archean times, when solar luminosity was 20-25 % lower than today, required an effective greenhouse atmosphere that may have been controlled by elevated partial pressures of N2 and CO2. Finally, this progressive decrease of temperatures at the surface of the Earth through time is inversely correlated with the progressive emerging of the continents since ~3.0-3.5 billion years ago, which suggests that the first order control on temperatures at the surface of the Earth over billions of years is the consumption and sequestration of atmospheric CO2 by weathering of continental surfaces.


Reference for the full open access paper is: R. Tartèse, M. Chaussidon, A. Gurenko, F. Delarue and F. Robert (2017) Warm Archean oceans reconstructed from oxygen isotope composition of early-life remnants. Geochemical Perspectives Letters 3, 55-65.


Further reading

  • Clarke A. (2014) The thermal limits to life on Earth. International Journal of Astrobiology 13, 141-154.
  • Gaucher, E.A., Govindarajan, S., Ganesh, O.K. (2008) Palaeotemperature trend for Precambrian life inferred from resurrected proteins. Nature 451, 704-707.
  • Jaffrés, J.B.D., Shields, G.A., Wallmann, K. (2007) The oxygen isotope evolution of seawater: A critical review of a long-standing controversy and an improved geological water cycle model for the past 3.4 billion years. Earth-Science Reviews 83, 83-122.
  • Knauth, L.P., Epstein, S. (1976) Hydrogen and oxygen isotope ratios in nodular and bedded cherts. Geochimica et Cosmochimica Acta 40, 1095-1108.
  • Marty, B., Zimmermann, L., Pujol, M., Burgess, R., Philippot, P. (2013) Nitrogen isotopic composition and density of the Archean atmosphere. Science 342, 101-104.
  • Robert, F., Chaussidon, M. (2006) A palaeo-temperature curve for the Precambrian oceans based on silicon isotopes in cherts. Nature 443, 969-972.
  • Urey, H.C. (1947) The thermodynamic properties of isotopic substances. Journal of the Chemical Society, 562-581.

About Katherine Joy

Hello! I am Katherine Joy. I am part of the University of Manchester Isotope Geochemistry and Cosmochemistry group. More details about my research interests can be found at
Gallery | This entry was posted in Earth, Laboratory and tagged , , , . Bookmark the permalink.