then freezing New York (why is it always New York?) after global warming destabilises the circulation of the Atlantic Ocean. Although the flash-freezing depicted in the movie is thermodynamically impossible, the scenario of a collapsing Atlantic current is not complete science fiction. All the models examined by the expert group led by Tim Lenton showed a tipping point in the North Atlantic where warmer, fresher waters could shut down the circulation pattern that brings comparatively balmy temperatures to the eastern US and high-latitude Western Europe. This shutdown would not trigger a new ice age, but temperatures in these regions could fall for several decades, causing serious impacts on societies and ecosystems alike.37 Again unlike the Hollywood movie, which showed temperatures dropping in seconds, the full transition towards an Atlantic Ocean circulation shutdown would likely take a century or more. More good news is that avoiding this tipping point is still possible: the scientists conclude from studying their models that a global warming of 3–5˚C would be needed to put us in the danger zone, well above the 1.5˚C maximum warming implied by our 350 ppm planetary boundary.
Another candidate on the tipping-point list is the Amazonian rain-forest. For years now many scientists have warned that global warming could trigger a collapse of the forest if rising temperatures lead to severe drought in western Brazil. This scenario seems even more of a danger given the recent droughts experienced in Amazonia in both 2005 and 2010, where entire river systems in this normally wet forest dried up for hundreds of kilometres. The problem here is that models don’t concur: some show a warmer Amazon getting wetter, whilst the most pessimistic forecasts for Amazon die-back are based on the projections of just one model, the HadCM3 model produced by the UK Met Office’s Hadley Centre. However, half of the 19 different models examined by a team of scientists led by Oxford University’s Yadvinder Mahli in 2009 did show a shift towards more seasonal forest, and a quarter showed that the rainforest could dry out sufficiently to collapse into a savannah-type ecosystem instead.38 Keeping global temperatures below 3˚C – very likely if our 350 ppm planetary boundary is achieved – should be enough to avoid this transition, but just as important will be respecting the other planetary boundaries on land use and biodiversity loss. The Amazon rainforest today is probably more threatened by deforestation and agriculture than it is by rising temperatures.
If the Amazon rainforest did collapse, huge quantities of carbon would be released in the process, giving a further boost to global warming. But the biggest carbon stores of all lie not in the tropics, but in the sub-polar continental regions where frozen permafrost holds enormous carbon stores tens of metres thick in Siberia and other high-latitude land areas. The threat to permafrost stability is possibly global warming’s biggest tipping point, because if this frozen carbon store begins to thaw, vast quantities of both carbon dioxide and methane will be released. According to a 2008 study in the journal BioScience, the carbon locked up in the Northern permafrost zone totals more than 1.5 trillion tonnes, double the entire carbon content of the atmosphere.39 Even if only 10 per cent of this permafrost thaws, another 80 ppm of CO2 will have accumulated in the atmosphere by 2100, raising the planet’s temperature by an additional 0.7 degrees40 – and making the eventual attainment of the 350 ppm climate change boundary much more difficult.
Scientists have already begun watching with some alarm a recent upward trend in atmospheric methane, some of which may be coming from the Arctic.41 Not all this methane – a greenhouse gas 25 times more potent than CO2 – is likely to bubble out of swamps on land; vastly more is contained in subsea sediments in the form of ice-like methane hydrates. If these hydrates melt rapidly as the oceans warm up, then all global warming bets are off – a scenario that has already sparked scary newspaper headlines. So how afraid should we be? Researchers have already reported seeps of methane leaking from the seabed offshore from eastern Siberia and the Norwegian Arctic islands of Svalbard, in both cases possibly in response to warmer ocean waters.42 But the experts are cautious. ‘Methane sells newspapers, but it’s not the big story,’ writes David Archer on the excellent RealClimate blog.43 ‘CO2 is plenty to be frightened of, while methane is frosting on the cake.’
Work by Archer and colleagues modelling the Earth’s response to climate change suggests that methane hydrate release could add another half-degree or so to the total warming, but only over several thousand years, and only if the released methane is not dissolved or oxidised first in the ocean before it has time to escape into the atmosphere.44 This is a ‘slow tipping point’, Archer concludes: it takes a long time for warming to penetrate the oceans, even longer for this to melt and release hydrates, and longer still for this methane to warm the atmosphere and the oceans further in a positive feedback loop. Happily, this is a tipping point we have still not crossed – ‘We have not yet activated strong climate feedbacks from permafrost and CH4 [methane] hydrates,’ reported a team of scientists in 2009.45 In the case of methane hydrates, respecting the climate boundary is not necessarily about protecting ourselves or even our children, but the stability of the Earth system over the very long term – for this tipping point, while slow to activate, would be essentially irreversible once crossed.
350: PAST EVIDENCE
If current observations of accelerating climate change and worries about tipping points in the future make two very good reasons why 350 ppm is the right place for a climate change planetary boundary, even stronger evidence comes from the Earth’s more distant climatic past. Climate models projections such as those published by the IPCC tend to project nice smooth – albeit upward-pointing – curves of likely future temperature trends. But a glance back in time, courtesy of ice-core records drilled in Greenland and Antarctica, shows that gentle, slow changes are far from being the norm in the Earth’s past. Instead, these records of past climate – which now reach back almost a million years – show climatic swings of extraordinary and terrifying abruptness. One extremely sudden warming took place in Greenland 11,700 years ago; it involved a temperature rise of 10 degrees Celsius within just three years.46 Rapid shifts are observed elsewhere too: 12,679 years ago, according to sediments recovered from a lake in western Germany, the European climate saw a sudden transition to more stormy conditions between one year and the next.47 The lesson is clear. Abrupt climate change is not the exception in the past, it is the norm. As the veteran oceanographer Wally Broecker says: ‘The climate is an angry beast, and we are poking it with a stick.’
Although current CO2 levels are higher than they have been for a million years, if we look even further back into the geological past there are episodes when both carbon dioxide and temperatures were far above where they are now. But rather than suggesting we have nothing to worry about, they further strengthen the evidence for counting 350 ppm as the crucial planetary boundary. For example, during the Pliocene epoch, about 3 million years ago, sea levels were 25 metres higher than today because the major ice sheets were much smaller than now due to a warmer climate. The CO2 concentration then? About 360 ppm – a line we crossed in 1995.48
The Earth was completely ice-free – and sea levels 80 metres or more higher – until about 33 million years ago, early in the geological epoch called the Oligocene. After having been at 1000 ppm or higher throughout the Cretaceous, Eocene and Paleocene, this was the moment when CO2 levels dropped past a crucial threshold allowing continental-scale ice sheets to form on Antarctica for the first time in perhaps a hundred million years.49 This CO2 level was 750 ppm, a level expected to be crossed again in about 2075 if carbon emissions continue to rise unabated. For the following 31 million years, only Antarctica held substantial ice sheets – until, late in the Pliocene,