A global project that’s been instrumental in shaping scientists’ understanding of how the oceans affect our climate celebrated its tenth birthday recently.
Over its short lifetime, the project has thrown up a few surprises. Parts of the Atlantic circulation seem to have slowed down, though whether that’s down to human activity remains to be seen.
Carbon Brief talks to one of RAPID’s founding scientists, Prof Harry Bryden from the National Oceanography Centre in Southampton, about the project.
Global heat transport
Above about 1,000 metres in the North Atlantic, warm water flows northwards from the equator towards the poles, releasing heat as it goes. The water cools and sinks at high latitudes, returning southwards towards the equator at much deeper depths.
This is known as the Atlantic Meridional Overturning Circulation (AMOC) and forms part of a global ocean conveyor belt that transports heat all around the world.
The Atlantic Meridional Overturning Circulation (AMOC). Warm water flows north in the upper ocean (red arrows) then sinks and returns south as deep cold water (blue arrows) Source: Srokosz & Bryden ( 2015) Supplementary material
The Gulf Stream – another component of the AMOC – is driven by the wind. Heat released to the atmosphere as the warm Gulf Stream moves northward gives northwest Europe its mild climate.
All components of the AMOC together transport about 18 million cubic metres of water per second – equivalent to a hundred times the flow from the Amazon river. The heat carried with it means North Atlantic sea surface temperature is about 5C warmer than in the North Pacific at similar latitudes.
In the beginning
Understanding the AMOC has been at the heart of oceanography since the 1800s, Bryden tells Carbon Brief. He says:
“It has been a topic that goes way back…But I think [it] was difficult for oceanographers back then because they typically measured on regional scales, they went on small expeditions.”
The centuries have brought a wealth of scientific advances, making the monitoring of such a vast ocean system possible. But there was a particular reason for beginning the RAPID project when they did, Bryden says.
“I think what specifically triggered our interest was the National Academy of Sciences produced a report about inevitable surprises.”
That view is now thought to be too simplistic. But the AMOC’s strong influence on North Atlantic climate and the potential for it to decline prompted the first attempt to monitor its behaviour.
Sea level north of New York rose by 12cm after the AMOC slowed by 30% in 2009-10. Credit: Lewis Tse Pui Lung via Shutterstock.com
The RAPID array
At a latitude of 26.5 degrees north, the RAPID sequence of moorings tethered to the sea floor stretches from the Bahamas on one side of the Atlantic to Africa’s continental shelf on the other.
The moorings’ instruments continuously measure temperature and salinity at different depths, allowing scientists to calculate how much water is crossing that line of latitude. Bryden explains:
“In effect, the ability to make sections across the oceans allows you to make a snapshot estimate. If you do a hydrographic section across 26 North, you can say what the overturning circulation is at that point, at that moment.”
But while a snapshot tells you about a point in time, the RAPID project aimed to make longer term measurements to uncover how the AMOC might be changing, Bryden says.
A decade of surprises
In just the first year of the RAPID project, the scientists found that the amount of water being transported by the AMOC in a single year was four times greater than ship data suggested.
The first four years of observations also showed that the seasonal cycle of the AMOC reached its maximum in autumn, rather than in summer as had been expected.
The AMOC appeared to be relatively stable for the ensuing five years. But soon there was “another surprise in store”, the new paper says.
From 2009-2010, the strength of the AMOC declined by 30% from one spring season to the other, before recovering again in 2010. This pronounced dip, which you can see in the graph below, was far outside the variations in strength the researchers had come to expect.
10-year time series of the strength of AMOC, measured as water transported in Sverdrups (millions of cubic metres of water per second). Graph shows 10-day average measurements (grey line) and 180-day average (red line). Source: Srokosz & Bryden ( 2015)
Scientists aren’t certain what triggered the sudden weakening, but several explanations have been suggested. It’s partly down to another Atlantic process – the North Atlantic Oscillation (NAO ), the researchers say. This natural fluctuation was in an extremely negative phase during the 2009-10 winter, which affected the amount of water transported.
With a few more years’ data, the researchers started to look for patterns beyond the year-to-year fluctuations. Climate models predict a slowdown in the overturning circulation as the climate warms, but the RAPID researchers found a much bigger change than expected.
From 2004-2012, the AMOC weakened by around 0.5 Sverdrups a year – that means each year, the conveyor belt transported half a million cubic metres less water every second. That’s a ten times bigger change per year than the Intergovernmental Panel on Climate Change ( IPCC) predicted.
But with only 10 years of data, it’s still too soon to say whether the decline is down to human-induced climate change. It may be a result of a decade-to-decade cycle of the overturning circulation, or natural fluctuations such as the Atlantic Multidecadal Oscillation (AMO ). Bryden says:
“It’s a 10-year trend and we don’t know whether it’s climate change. No-one’s ever measured for 10 years before.”
The need to distinguish between short- and longer-term variability is just one reason why continued measurements of the AMOC are needed, the paper concludes.
Decline since 2004 in the northward transport anomaly (red) at 26.5N relative to the mean annual cycle. Source: Smeed et al., (2014)
Another reason the RAPID project is important is the potential impacts for Europe and the Eastern US of changes in the overturning circulation. Bryden says it looks like the 2009-10 dip in strength caused local changes in sea level. He says:
“The event we had in 2009-2010 with a 30% slowdown, seems to be associated to 12cm sea level rise north of New Yorkâ?¦That’s quite an impact.”
Warmer than usual water in the tropics south of 26.5N may have also have triggered a particularly strong hurricane season in the US, the paper notes. If the AMOC is slowing down over time, that could also have consequences, Bryden explains.
“Ultimately, we think it will affect…the warming of the atmosphere by the ocean. And potentially the temperature over northwestern Europe.”
But a common misconception is that the Gulf Stream has slowed, says Bryden. The Gulf Stream is primarily wind-driven and most of the observed decline in AMOC strength is happening in the other components, he says:
“What’s changed is the amount of water recirculating southward in the interior of the ocean, in the so-called gyre circulation. That has increased, so that more of the Gulf Stream water is recirculating, rather than heading north towards Iceland.”
For the RAPID scientists, five main questions remain.
Will changes in the AMOC continue, or will it bounce back to its earlier strength? Are the same changes occurring at other latitudes in the Atlantic? How unusual was the 2009-10 dip? Could the AMOC ‘switch off’ entirely? And if so, what would the impacts be?
On these remaining questions, the paper notes:
“Irrespective of whether the present decline in the AMOC continues, ends, or reverses, the observations will provide a stringent test of different climate models’ abilities and whether their projections will prove valid.”
The RAPID scientists say they can be sure of one thing, at least. There are likely to be many more surprises along the way.
Main image: Surf on rocky coast, Henningsvaer, Austvagoya, Lofoten, Norway.
Srokosz, M.A. and Bryden, H.L. (2015) Observing the Atlantic Meridional Overturning Circulation yields a decade of inevitable surprises, Science, doi:10.1126/science.1255575