The Oligocene to the Quaternary: climate and biology - Climate Change: Biological and Human Aspects - page 143

might signify a number of major landslides occurring more or less simultaneously,
geologically speaking (within a century or two).
Before we leave methane it is worth mentioning here that the Intergovernmental
Panel on Climate Change (IPCC) have warned policy-makers to be wary of surprises
and not to rely solely on climatic forecasts (we shall return to this in Chapter 6). How-
ever, one such surprise we may encounter as the Earth continues to warm might be a
major release of methane from marine hydrates as temperature destabilises them. Of
course, since in the 20th century sea levels globally have been rising marine methane
hydrates are under more pressure rather than less and this will confer greater stability.
Yet if the temperature of the water around the hydrates were to increase by about 5 ◦ C
then there may be a massive release of methane, as was likely back in the Eocene. One
recent estimate of the methane that could be released this way globally was provided
by a geological team led by Matthew Hornbach (2004). They suggest that a release
of some 2000 Gt of methane is possible. To place this into the context of present-day
anthropogenic warming from carbon dioxide releases, this is roughly 30 times the
mass of carbon that was released as carbon dioxide from fossil fuels in the final
decade of the 20th century. Furthermore, from Chapter 1, bearing in mind methane's
higher global warming potential (GWP) then the warming possible over two decades
of such methane release would be far greater. However, methane in the atmosphere
over several years gets oxidised to carbon dioxide, which is a comparatively weaker
greenhouse gas but which stays in the atmosphere for a century or two. In short,
should we experience a theoretical release of methane today from some hypothetical
dissociation of all methane clathrates, then we would see a very considerable initial
warming followed a longer period when it would be less warm than during the initial
burst. Fortunately, a 5 ◦ C rise in abyssal waters - which is necessary to destabil-
ise methane hydrates - is highly unlikely this century (although we should heed
the IPCC's warning of surprises and other destabilising mechanisms). Nonetheless,
such a risk becomes increasingly relevant should anthropogenic warming and fossil
fuel consumption continue to grow and especially if it does into the next century.
What we do not want to happen is to see an event analogous to the Initial Eocene
Thermal Maximum/Palaeocene-Eocene Thermal Maximum (IETM/PETM; see

section 3.3.9).
Turning away from the marine environment, it is also worth emphasising that
wetlands are not the only terrestrial methane source. Peatlands represent a huge
store of carbon. High-latitude peatlands, with between 180 and 455 Gt of carbon,
represent up to about a third of the global soil carbon pool. Most have been formed
since the LGM and so represent a major terrestrial carbon sink with 70 Gt of carbon
being sequestered since that time. Much of this initial peatland formation took place in
Siberia's western lowland before their counterparts in North America: all told, Russia's
peatlands contain perhaps half of the planet's peat. Peatlands are, in interglacial times,
an atmospheric methane source but much is still unknown as to how exactly they fit
into the dynamic carbon cycle over glacial-interglacial cycles (Smith et al., 2004).
Indeed, another carbon cycle unknown is the degree to which these high-latitude
peatlands and permafrost soils might release their tremendous store of carbon if
warmed (see Chapters 1 and 7). These are just two more reasons why climate computer
models are not complete (see section 5.3.1).