In the 2004 disaster film The Day After Tomorrow, climate change causes a major disruption of ocean currents in the Atlantic Ocean, which in turn brings about a sudden ice age in New York City.
That scenario was ridiculous and overwrought. Still, the underlying idea that global warming could mess with some important ocean circulation systems actually isn't that far-fetched. Such an event wouldn't blanket Manhattan in ice, but it could wreak havoc on fisheries or speed up sea-level rise in cities like Boston and New York.
In particular, scientists have been paying close attention to the Atlantic Meridional Overturning Circulation (AMOC), an ocean pattern that transports warm water from the tropics to the North Atlantic. This Gulf Stream system is why Europe has a mild climate despite being so far north. But it's also a source of concern: evidence suggests the circulation has slowed abruptly in the very distant past, leading to sharp climate shifts regionally. More recently, the AMOC has been seen weakening and strengthening, but scientists figured it was just fluctuating naturally.
Now, however, a new study in Nature Climate Change argues that the Atlantic overturning circulation isn't just fluctuating up and down. The researchers, led by Stefan Rahmstorf of the Potsdam Institute for Climate Research, created an index of the AMOC's behavior going back centuries. They conclude the system has weakened to its lowest level in 1,100 years, perhaps due to an influx of freshwater from Greenland's melting ice sheet.
In other words, global warming may already be slowing down the Gulf Stream system — sooner than anyone expected it would. If so, that could mean the AMOC will keep weakening in the decades ahead, potentially bring things like harsher winters in Europe and faster sea-level rise along the US East Coast.
And what about the worst-case scenario — a sudden shutdown? For now, most scientists still don't think that will happen anytime soon. Climate models do expect the AMOC to eventually weaken if the Earth keeps warming up, but those models generally don't show drastic, abrupt changes this century.
Yet there's a catch: If this latest study holds up and melting ice in Greenland is indeed driving a slowdown earlier then an expected, that could suggest there's a lot we still don't understand about how the AMOC works. "The slowdown we see in the data is not what you see in the climate models," Rahmstorf told me. "If climate models have underestimated the decline, we might be closer to that threshold [where abrupt changes happen] than we thought."
The melting of Greenland may be slowing the Atlantic circulation
The Atlantic Meridional Overturning Circulation is essentially a giant conveyor belt of water. Ocean currents bring warm water up from the tropics to the North Atlantic. There, the water gets colder and saltier (thanks in part to the formation of sea ice) and sinks, pushing its way back down south. The NASA video above shows the process.
Previous research has suggested that this system has been fluctuating naturally in recent decades — in particular, it seemed to weaken sharply in the 1970s and then strengthen in the 1990s. But it wasn't clear if there were any long-term trends.
That's what Rahmstorf and his co-authors wanted to figure out. Unfortunately, this isn't easy to do, since scientists have only been observing the AMOC for a few decades. So they created a proxy record, looking at climate conditions believed to be affected by fluctuations in the circulation, such as the difference in sea-surface temperatures between the southern and northern hemispheres. (This can be measured far back using corals and other proxies.)
What they discovered was that the AMOC isn't just fluctuating up and down. It appears to have weakened dramatically since the 1970s — to a level not seen in the last 1,100 years:
So what's happening? One possibility is that global warming is driving the decline.
In their study, Rahmstorf and his co-authors point out that Greenland has been melting significantly in the past few decades. That influx of cool freshwater can reduce the surface density of ocean water in the North Atlantic, preventing it from sinking. That, in turn, would weaken the AMOC's conveyor belt system.
Scientists who weren't involved in the study said this new finding was potentially very important, but they wanted to see additional confirmation first. Gerald Meehl, a senior scientist of the National Center for Atmospheric Research, notes that reconstructing the behavior of the Atlantic circulation over the last 1,100 years is extremely difficult to do. "They're essentially inferring what the circulation has been doing from other processes," he says.
"The key to this whole paper is whether this index really reflects the behavior of the AMOC," agreed Tom Delworth, a research scientist at NOAA's Geophysical Fluid Dynamics Laboratory. "If it holds up, it's a tremendous achievement. But as a scientist you'd always like to see more lines of evidence."
"I'm more of the opinion that a lot of what we've seen in the 20th century is internal variability," Delworth added. "But if the evidence says that it's forced" — that is, if the Atlantic overturning is now being weakened due to global warming — "then we have to follow that."
Why it matters if the Gulf Stream system is now weakening
If this study holds up, it would imply that the AMOC could weaken even further in the next decade or two. That would have all sorts of messy consequences for both Europe and the United States.
"If the slowdown of the Atlantic overturning continues, the impacts might be substantial," Rahmstorf said. "Disturbing the circulation will likely have a negative effect on the ocean ecosystem, and thereby fisheries and the associated livelihoods of many people in coastal areas. A slowdown also adds to the regional sea-level rise affecting cities like New York and Boston."
Here's what he means: Sea levels around the world are already expected to rise 1 to 3 feet or more this century if global warming continues — both because ocean water expands as it warms and because ice caps will melt, particularly in Greenland and Antarctica. But sea levels won't rise uniformly everywhere, since local conditions are also affected by winds and currents.
Right now, the Atlantic overturning circulation is essentially suppressing some sea-level rise along the Northeast US by keeping waters relatively cool, so they don't expand as rapidly. But if the AMOC keeps weakening, sea levels will rise faster along the East Coast than they otherwise would. One recent study found that abrupt AMOC changes in 2009-2010 led to a five-inch spike in sea levels along the Northeast US.
And those are just the effects scientists know about. If the circulation keeps weakening, it would also mean that a region known as the "subpolar gyre," a patch of water to the south of Greenland, will continue to warm more slowly than the rest of the world:
This relatively cool patch could affect weather patterns in North America and Europe, possibly by affecting what's known as the North Atlantic Oscillation. At least one study has suggested a link between a weaker AMOC and harsh winters in Europe. This, however, is an area that scientists still don't understand very well.
Could the Atlantic circulation abruptly shut down?
The big question is whether the Atlantic overturning circulation could weaken so much that it abruptly shuts down, possibly with dramatic effects on regional climate systems. Indeed, some scientists have argued that a sudden shutdown may have led to a serious cold spell in Europe 12,900 years ago.
There is some reason for optimism here. As a 2013 review by the National Research Council found, most climate models agree a shutdown was unlikely this century. The Atlantic overturning circulation is expected to eventually weaken as the planet heats up, the models say, but it should remain stable for the foreseeable future.
But what if those models are wrong? The Nature Climate Change paper points out that current climate models don't suggest the AMOC should already be weakening the way their paper suggests it is. What's more, computer models may be missing important dynamics, like runoff from the ice caps in Greenland, which are now melting faster than anyone expected.
"Our results could suggest that we're closer to that point of extreme sensitivity in the AMOC than state-of-the-art climate-model simulations suggest," says Michael Mann of Penn State, a co-author of the Nature Climate Change paper. "It could be another example, as if we needed one, where climate projections may be, if anything, erring on the conservative side."
For now, other scientists are still skeptical that we're likely to see an abrupt shift anytime soon, regardless of what's happening now. "I think the models do get the big features right," says Meehl. "They show the AMOC will continue to slow down, but then it actually becomes a bit more stable. An abrupt change, where the whole AMOC just stopped, would take a huge amount of freshwater, and it's hard to imagine Greenland destabilizing that quickly."
Delworth offered a similar assessment: "Right now we don't have any good evidence that there's likely to be an abrupt change [this century]," he said. "But there are still uncertainties, in part because our models don't capture how the AMOC works in great detail."
Indeed, in its 2013 review, the National Research Council suggested that "it is important to keep a close watch on this system" — possibly with more observation devices. That way we can better understand the impacts of a slowdown, as well as keep an eye on the possibility of big, drastic shifts.
Further reading: Stefan Rahmstorf has an insightful discussion of the new paper over at RealClimate.