Past and present trends in methane and clathrates

Variations in _CH4 content measured in the EPICA Dôme-C ice core in Antarctica can be traced back as far as 800,000 years BP. Long-term fluctuations can be explained by the superposition of cyclical components corresponding to the parameters of the Earth's orbit (precession, obliquity and eccentricity). Astronomical forcing has caused seasonal insolation to vary at different latitudes, leading to migrations of the inter-tropical convergence zone and changes in monsoon systems. These climatic variations have influenced methane sources and sinks in the intertropical zone.

By subtracting the cyclic component from the raw _pCH4 record, a high-frequency component can be extracted. This clearly correlates with the rapid temperature variations observed in the northern hemisphere (Dansgaard-Oeschger (DO) events identified in Greenland ice). This rapid _CH4 variability is also correlated with changes in the Asian monsoon recorded in Chinese stalagmites. The most recent DO event (DO1 or Bölling-Alleröd) enables us to study with great precision the phase shift between atmospheric _CH4 and Greenland temperature. The results show that the two parameters have evolved almost synchronously, with a time lag of less than 25 years.

By considering the interhemispheric gradient of _pCH4 over the same period (the difference between Greenland and Antarctic records), it is possible to propose _CH4 emission scenarios involving both boreal sources linked to warming, and tropical sources linked to the reactivation of wetlands. To further discriminate sources, the ^13C/12Cratio was measured in parallel with _pCH4 for the last ice age. Surprisingly, variations in the ^13C/12Cof _CH4 are not correlated with those of _pCH4, but follow fluctuations in atmospheric _pCO2. This suggests that_CO2 also has a controlling effect on _{CH4-producing} tropical ecosystems, notably on the ratio of C3 to C4 plants, the former being favoured by increases in _pCO2.

As a counterpoint to these mechanisms involving continental sources, paleoceanographer Jim Kennett has proposed an alternative hypothesis involving marine _CH4 hydrates. This suggestion is based on regional observations in sediments off California. Sedimentary records show ^13C/12Canomalies and the presence of certain molecular tracers (diplopterol, for example) that are signatures of marine _CH4 releases during warm phases of DO variability characterized by _pCH4 maxima.

In a clathrate (or gas hydrate), _CH4 is trapped in a crystalline structure with water molecules forming a cage around the gas molecules. The methane formed by methanogenesis in the sediment is stabilized in hydrate form at low temperature and high pressure (e.g. 0 °C and 30 bars). As clathrates are less dense than water, they can only be preserved in sediment. However, as temperature increases with depth (general geothermal gradient), clathrates are no longer stable at a certain depth in the sediment (< 1 km). A layer of clathrates is thus formed, which can be located and mapped by seismic analysis of the reflector linked to the solid-gas contrast. Marine clathrates are mainly located on continental margins at depths greater than 500 m. The current inventory of _CH4 hydrates is approximately 1,800 Gt of carbon, which is huge compared with the atmospheric _CH4 stock (≈ 4 GtC).

The D/H ratio of _CH4 from marine clathrates is much higher than that of sources from continental wetlands. The D/H of _CH4 from polar ice air bubbles was therefore measured to test the hypothesis that marine clathrates are responsible for the glacial variability of atmospheric _CH4. In fact, D/H decreases were recorded during DO events, which is incompatible with a marine source. This conclusion is confirmed by the ^14Ccontents of _CH4 in air bubbles occluded in Greenland ice, which were analyzed for the last late glacial transition (Recent Dryas-Preboreal). Instead of decreasing sharply during the transition, _14CH4 remains stable, consistent with _CH4 from continental wetlands.

Although clathrates do not appear to be responsible for DO climate variability, it is important to continue studying their destabilization mechanisms, which could increase atmospheric _pCH4 in the future.

It should first be pointed out that, after clathrate destabilization, the _CH4 that diffuses into the water column is progressively reoxidized into_{CO2 by}methanotrophic bacteria. This is notably the case for _CH4 fromcold seeps, which are deep-sea sites where numerous gases such as hydrogen sulfide (_H2S)and hydrocarbons emanate at low temperatures (< 40 °C). For _CH4 to reach the atmosphere, destabilized hydrates must be located in shallow sediments. However, these sediments are generally poor in clathrates : only a few percent of the global stock is found at depths of less than 500 m.

In direct relation to current global warming, the ocean is absorbing a preponderant share of the heat that is gradually diffusing at depth. This raises legitimate concerns about the future of clathrates trapped in ocean sediments. The destabilization of _CH4 hydrates is a slow process, as heat progresses slowly through sediments. A temperature disturbance at the sediment surface will take around a millennium to reach a depth of 200 m, with only half the effect.

A number of cases were discussed regarding the penetration of a temperature change in sediments under varying water depths. For a theoretical warming of 1 °C over a century, destabilization would be minimal for clathrates located at depths of over one km. Only the rare hydrates below 500 m would be affected. A case of oceanic _CH4 release has been demonstrated off Spitsbergen at depths of a few hundred meters, which have warmed over recent decades.

In addition to oceanic clathrates, there are also continental _CH4 hydrates located beneath permafrost in the high latitudes of the Northern Hemisphere. These clathrates account for around 1 % of the _CH4 hydrate inventory. Even with cold surface temperatures, permafrost thickness is limited to a few tens or hundreds of meters due to the geothermal gradient.

In phase with the warming of the Arctic zone, evidence of melting permafrost has been observed with the formation of thermokarst lakes. A number of studies of organic matter fermentation and _CH4 release in arctic lakes distinguish between several types of _CH4 seeps, both superficial and deep(subcap seeps). _CH4 isotope analyses (D, ^13Cand ^14C) are used to distinguish the different carbon origins.

Systematic fluxes of _CH4 to the atmosphere have been observed in the Laptev Sea, and much of Siberia's coastal waters are supersaturated with _CH4. This is not due to the destabilization of marine clathrates, but to the release of hydrates located beneath the permafrost that was flooded during the last post-glacial marine transgression. As a result of this phenomenon, a strong temperature change propagated through the permafrost, which is now covered by a thin layer of marine water.

Although the contribution of hydrates to atmospheric _CH4 is still only a natural response to post-glacial warming, this should not blind us to the fact that global warming is amplified in the polar zone. This should accelerate the destabilization of continental hydrates, which could contribute to an increase in atmospheric _CH4 content. Nevertheless, the current _CH4 flux linked to hydrates in general still seems very low (≈ 1 %) compared with other natural and anthropogenic _CH4 sources.