As mentioned in Ten dangers of Global Warming, one of the biggest dangers is that, without dramatic action, the atmosphere will reach certain tipping points beyond which sudden dramatic and catastrophic changes will take place.
As Earth warms up, tectonic plates will expand and some areas will come under increasing pressure, especially along fault lines where tectonic plates collide. As described in this comment and this post, this could lead to earthquakes. Thermal expansion of land and water could put more stress on areas prone to seismic activity, triggering earthquakes that can make the greenhouse effect much worse. The danger is that such seismic activity will cause slope failure in regions with methane hydrates that are already unstable and vulnerable due to global warming.
Ice and glaciers melting away
Links between climate change and geological and geomorphological phenomena were the theme of this 2009 conference. Several speakers addressed the danger that, as ice and glaciers in the mountains melt away, a substantial weight is disappearing, changing pressures that act on the Earth's crust and contribute to seismic activity. This link was confirmed in several scientific studies, such as this one dating back to 2003.
Hydrates disturbed by drilling and fracking
There is also an indirect risk. Melting of Arctic sea ice may open up sea routes to hydrates. Drilling and fracking in these hydrates could trigger earthquakes, especially if they're already under extra stress, resulting in the release of huge amounts of methane. This is particularly worrying in the Arctic, where waters can be very shallow, leaving less opportunity for methane to be broken down in the water.
Deep Ocean Warming
The ocean conveyor belt transports water--and heat--around the globe, as shown on the image left, from a NSF press release describing recent research by scientists at NCAR and the Bureau of Meteorology in Australia, which found that deep oceans can warm by 18% to 19% more during a period corresponding with a La Niña event. Global warming is likely to cause thermal expansion of the oceanic crust, putting stress on areas where tectonic plates meet. Such a warming peak deep in the ocean could put enough extra stress on these areas to trigger earthquakes that in turn disturb hydrates, resulting in huge amounts of methane to be released. The NOAA image below shows how the Mid-Atlantic Ridge continues into the Arctic Ocean.
Gakkel Ridge
One place to watch is Gakkel Ridge, the boundary between the North American Plate and the Eurasian Plate. Earthquake activity along Gakkel Ridge has been rising since 1970. Earthquakes in the Gakkel Ridge area could send shockwaves into the shallows of the Arctic Ocean.
Between 1999 and 2000 alone, there was an anomalously large number of earthquakes along the Arctic Gakkel Ridge (more than 250). In addition, two very unusual and extremely violent submarine pyroclastic eruptions occured in the central Gakkel Ridge region.
Of the earthquakes measured on the Arctic Gakkel Ridge between March 19th 1980 and the 31st December 2010, most (94%) were strong enough to cause widespread collapse of the methane hydrates and release of methane plumes into the water column and atmosphere. [source: globalwarmingmlight.blogspot.com]
Runaway Warming
In conclusion, global warming can accelerate in a number of ways, including thermal expansion of tectonic plates, causing landslides and shocks from earthquakes, while extra stress can be added due to deep ocean warming peaks and a change in weight as ice retreats on land. This could be ameliorated by drilling and fracking activities.
The danger is that this will put increasing stress on hydrates that can contain huge amounts of methane. If such hydrates are disturbed, huge plumes of methane can be released, causing supersaturation of waters with methane. As a result, further methane releases will enter the atmosphere without being oxidized in the water. The risk is that such methane releases lead to runaway global warming.
This risk is unacceptable, making it imperative to reduce emissions and bring atmospheric carbon dioxide down, which is best achieved by means of feebates and requires a number of geoengineering techniques, as discussed in Sustainable Economy. In Geoengineering the climate (Royal Society, 2009) various geoengineering methods are compared. These methods may differ in timescale, cost-effectiveness and wider impact (see e.g. this posts on Biomass), but the urgency to act on global warming is such that we may well need all of them to avoid runaway global warming.
