Although the idea of global warming triggering an ice age may be couter-intuitive, the science is pretty solid. Melting icepack in Greenland results in the dumping of large amounts of fresh water right into the path of the North Atlantic warm water flow, resulting in the slowing and eventual cut-off of the circulation; this, in turn, results in lower temperatures in parts of the Northern Hemisphere, with Europe likely to be hit the hardest. Such a pattern has happened in the past due to the slower natural cycles of global temperatures; the fossil and ice core evidence suggests that the shift from a warm, wet environment to a cold, dry climate could take place over a matter of a few years.

Recent findings that the warm water flow may, in fact, be seeing a dramatic reduction has turned this concept from a theoretical possibility to a very real threat. But what would that world look like? Two studies give us very different images of what might happen.

Climate scientists at the Goddard Institute for Space Studies have been able to use a new model of ocean-atmosphere interaction to reproduce the effects of a massive freshwater dump into the North Atlantic that happened 8,000 years ago. The models results closely matched the geological record, and gave researchers new insights into the effects of a thermohaline circulation drop-off.

The researchers prodded their model with a freshwater pulse equal to between 25 and 50 times the flow of the Amazon River in 12 model runs that took more than a year to complete. Although the simulations largely agreed with proxy records from North Atlantic sediment cores and Greenland ice cores, the team's results showed that the flood had much milder effects around the globe than many people fear--including the dramatic shifts in climate depicted in the 2004 movie 'The Day After Tomorrow'.

According to the model, temperatures in the North Atlantic and Greenland showed the largest decrease, with slightly less cooling over parts of North America and Europe. The rest of the northern hemisphere, however, showed very little effect, and temperatures in the southern hemisphere remained largely unchanged. Moreover, ocean circulation, which initially dropped by half after simulated flood, appeared to rebound within 50 to 150 years.

"This was probably the closest thing to a 'Day After Tomorrow' scenario that we could model," said LeGrande. "The flood we looked at was even larger than anything that could happen today. Still, it's important for us to study because the real thing occurred during a period when conditions were not that much different from the present day."

So here's one end of the potential impact spectrum: a minor temperature drop, with ocean circulation disruption lasting no more than a century and a half (a blink of the geological eye). This would be a problem, to be sure, but not the global civilization-disrupting event seen in some scenarios. Unfortunately, other research suggests that a far worse result is also possible.

Flavia Nunes and Richard Norris, oceanographers at the University of California, San Diego, have produced an article for Nature describing the "Palaeocene/Eocene Thermal Maximum" (PETM) event of about 55 million years ago, during which ocean temperatures shot up by 7 or 8 degrees C, followed by a rapid change in ocean circulation. The Atlantic circulation saw a disruption of flow lasting a couple of thousand years, then a reversed flow that took 100,000 years to correct. The implication is that a major thermohaline circulation disruption would not be a brief event, but an extremely long-term problem.

Although scientists have found evidence of similar links between ocean current and climate in the last 200,000 years, [Woods Hole oceanographer Karen] Bice says that the PETM is a much better analogue for the climate change we see today, because it occurred in a world that was warming gradually in response to rising greenhouse gas levels. [...]

"It's a good indicator of what could happen in our own future," warns Bice.

It's notable that the PETM conditions are not known to have triggered an ice age. This is likely because the temperature increase was sufficiently dramatic as to melt the undersea methane clathrates, which in turn released a sufficient level of greenhouse gas to keep the global environment quite warm for potentially millions of years.

So which is it? Hopefully, the first scenario is the more accurate reflection of the current environment; the second scenario, with melting clathrates boosting global temperatures for millennia (at least), would be a far greater disaster than most of us could imagine. It's as yet unclear what would make the difference between a mild abrupt ice age and an effectively permanent climate disaster; the likely tipping point issue is whether the ocean temperatures get hotter fast enough to melt the clathrates before the cooling effects of the mini ice age can really kick in. That is to say, this is another piece of evidence making it clear that we need to stop making things worse.