The Oligocene to the Quaternary: climate and biology - Climate Change: Biological and Human Aspects - page 142

the seabed on the slopes of the continental margins (see section 3.3.9 on the Eocene).
Estimates of the total quantity of methane stored as hydrates vary mainly between
l 10 19 g. This is about 4000 times the amount in the atmosphere

today. Because methane hydrates contain a smaller proportion of the 13 C isotope
compared to atmospheric carbon dioxide - you will recall that the enzymes driving
photosynthesis prefer 12 Cto 13 C - these clathrates are biogenic in origin (see section
3.2.2). As noted in the last chapter when discussing the end-Permian extinction, the
physicochemical properties of methane hydrates are that they are meta-stable and
will easily dissociate into hydrogen and water with an increase in temperature or a
decrease in pressure.
In 2000 James Kennett and his team, from the Universities of California and
California State, discovered oscillation in carbon isotopes in forams that could be
explained by methane releases from hydrates. The team discovered four major releases
including one just before the beginning of the Holocene interglacial, and these took
place at the same time as four major Heinrich events at the beginning of Dansgaard-
Oeschger cycles.
In 1998 a team led by Guy Rothwell of Britain's Southampton Oceanography
Centre suggested that the biggest climate change over the last glacial-interglacial
cycle was possibly attributable to methane release from marine methane hydrates.
Seismic studies of the Balearic abyssal plain to the west of Sardinia revealed
an extensive ancient submarine landslide three times the size of Wales involving
500 km 3 of mud. Rothwell's team radiocarbon-dated sediments above and below the
landslide and got an estimated age of 22 000 years. This date ties in with the LGM.
Could it be that the landslide was the result of a massive methane release well in
excess of half a billion tonnes? This would have represented a very significant input
of gas to the Earth's atmosphere at a time when it contained just 0.8-1.0 billion t. As
a greenhouse gas methane is far more powerful, molecule for molecule, than carbon
dioxide, but lasts for less time in the atmosphere (see section 1.2). Even so, such a
methane release could easily have given the Earth's climate system a jolt. A decade
or so of warming would trigger other changes. For example, continuing with meth-
ane, warming could have released more of the gas (and other forms of carbon) from
permafrosts, and there would be implications for atmospheric and ocean circulation.
What is known for certain is that there was a short period of warming about this time,
in the depth of the last glacial, even though temperatures did not return to anywhere
near interglacial levels.
But what triggered the massive landslide? It could have been coincidence as such
landslides happen by chance. Whereas chance is certainly a factor in marine land-
slides, in that they are fairly common, what made such a large area collapse all at
once? The suggestion is that as the glacial progressed and ice caps grew, sea level
dropped. As the sea level dropped the pressure of water stabilising the meta-stable
hydrates declined. Hydrates in this depth zone risked dissociating and all it took was
one trigger event (an initial slide) to trigger others. Of course, sea levels did not
just decline in the Mediterranean, so if this climate change mechanism has validity
one would expect other major slides worldwide but perhaps spread out over a few
centuries. Interestingly, Antarctica and Greenland ice cores reveal that there was a
major methane overshoot prior to the beginning of the previous interglacial, and this

l0 19 × and 2 ×