Clathrate Gun Shot Down?

Greenland Ice Core_0

23rd July 2010.

Arctic and Antarctic ice cores provide a rich source of evidence that both temperature and atmospheric concentrations of methane and carbon dioxide have fluctuated over the past 800,000 years. These studies have also shown that the fluctuations are cyclic, and that higher temperatures are generally associated with higher levels of methane and carbon dioxide in the atmosphere. The source(s) of the periodic increases in methane concentrations are less clear, and have been the subject of much debate. Two competing hypotheses have been proposed: (1) that the extra methane comes from increased wetlands emissions (the ‘wetland hypothesis’) and (2) that methane hydrate disintegration is responsible for the increase in methane which in turn can trigger a dramatic temperature rise (the ‘clathrate gun hypothesis’).

A new study (Bock et al., 2010) of the hydrogen isotope ratios in methane from North Greenland ice-cores dated from 33,700 – 41,000 years ago, has now tipped the balance in favour of the wetland hypothesis for the series of climate fluctuations known as the Dansgaard-Oeschger events.

The study uses the fact that the hydrogen isotope composition of methane emitted from wetlands falls in the range -300 to -400‰ δD(CH4), distinctly more depleted in deuterium than that in hydrate-sourced methane, which averages ~ -190 δD(CH4). A number of other factors (e.g. precipitation, temperature-related Rayleigh distillation, kinetic fractionation associated with the OH sink) also impact on the hydrogen isotope ratio, but these are secondary effects which only alter the δD(CH4) by a few per mill

Bock et al. observed that over the 7,300 years represented by their ice cores, the methane from the interstadial (warmer temperature) periods was ~ 10 ‰ more depleted in deuterium than that from the cooler periods (stadials), and that this change in δD(CH4) could only be explained by the wetland hypothesis. Modelling indicated the hydrogen isotope measurements in the interstadials were consistent with a six fold increase in high latitude wetland emissions, from ~5 to ~ 32 Tg CH4 year-1, an increase of ~84 to ~118 Tg CH4 year-1 from tropical wetlands, and a constant rate of emission from marine hydrates of ~ 25 Tg CH4 year-1.

Hydrates expert Professor Mark Maslin, Director of the UCL Environment Institute, agrees that the paper “provides good evidence that gas hydrates were not involved in the initial methane rise…” during the Dansgaard-Oeschger events, but believes that there is “still considerable debate about whether tropical wetland, boreal wetlands or flooded shelf due to millennial scale sea level rise are the true source of the methane rise….” The authors of the paper concede that the methane cycle in this period remains underdetermined. However, the ruling out of one important potential methane source represents a significant advance in our understanding.


Michael Bock, Jochen Schmitt, Lars Möller, Renato Spahni, Thomas Blunier and Hubertus Fischer (2010), Hydrogen isotopes preclude marine hydrate CH4 emissions at the onset of Dansgaard-Oeschger events, Science, v.328, 1686-1689.

Mark Maslin, Matthew Owen, Richard Betts, Simon Day, Tom Dinkley-Jones and Andrew Ridgwell (2010), Gas hydrates: past and future geohazard? Phil. Trans. R. Soc. A, v.368, 2369-2393