Oceans represent an important global carbon sink, absorbing 25% of annual man-made CO2 emissions and helping to slow the rate of climate change. The Southern Ocean in particular is known to be a significant oceanic sink, and accounts for 40% of all carbon entering the deep oceans. And yet, until now, no-one could quite work out how the carbon gets there from the surface waters.

"We thought wind was the major player," said lead author of the new study, Jean-Baptiste Sallée of the British Antarctic Survey. "The ocean is like an onion – in layers – and there is very little connection between the surface and deep layers," he explained. When strong winds displace a large slab of surface water and cause it to accumulate in a specific region, the localized bloat in the surface layer gets injected downwards into the ocean's interior. But this kind of wind action alone should have a fairly uniform effect over vast swathes of ocean – which is not what the scientists measured.

Subduction hotspots

Scrutinizing 10 years of temperature, salinity and pressure data from a fleet of 80 small robotic probes dotted around the remote Southern Ocean, the researchers discovered that surface waters are drawn down – or subducted – at a number of specific locations. This occurs due to the interplay between winds, dominant currents and circular currents known as "eddies". "You end up with a very particular regional structure for the injection of carbon," said Sallée, describing 1000 km-wide funnels that export carbon to the depths.

The team pinpointed five such zones in the Southern Ocean, including one off the southern tip of Chile and another to the south-west of New Zealand. Elsewhere, currents return carbon to the surface in a process known as "reventilation", but overall, the Southern Ocean is a net carbon sink.

Carbon bottleneck

The mechanisms governing atmosphere-to-ocean carbon transfer – the mechanical mixing action of wind and waves, and biological uptake by micro-organisms in the sunlit top layer of water – are already well understood. The step that determines the rate of the oceanic uptake of carbon, according to co-author of the study, Richard Matear of Australia's Commonwealth Scientific and Industrial Research Organization, is the physical transport of this dissolved carbon from the surface waters into the ocean interior. "Our study identifies these pathways for the first time," he said.

Understanding these subduction pathways fully is key to predicting how climate change might alter the Southern Ocean's carbon sequestering capabilities. Both global warming and the Antarctic ozone hole increase the temperature gradient between the equator and the pole, which intensifies the southern hemisphere winds. Climate models predict that stronger winds could stir up deep waters, especially in violent seas such as the Southern Ocean, and result in a net release of carbon back into the atmosphere.

"What we don't know yet is the impact of climate change on eddy formation," said Sallée. Eddies arise from oceanic instabilities caused by extreme gusts of wind, intense surface heating or cooling, or strong currents meeting uneven bottom topography, but tend to escape the granularity of even the most detailed climate models. "We can speculate that if wind increases, there will also be more eddies to counterbalance its effects," considered Sallée. "But it's a question we don't know how to answer yet. And it's a big incentive for climate models to refine their grid."

Improving climate models

Ocean-carbon-cycle expert Corinne Le Quéré, director of the Tyndall Centre for Climate Change Research, UK, who was not involved in the study, echoes Sallée's call for improved understanding of wind-eddy interplay. "Southern Ocean winds have increased in the past 15 years in response to the depletion of stratospheric ozone," she explained, adding "There's a lot of discussion right now about how [wind-induced changes] are then counterbalanced by changes in eddies."

Because of this, today's climate models diverge when it comes to predicting the future carbon sequestration response of the Southern Ocean. "I think this [paper] is really the first time that we have such small-scale resolution in the exchange of carbon in the ocean from observation directly," said Le Quéré. "The natural next step will be to take climate models and see how well they're performing spatially and [temporally]...this study can really help constrain which are the good models."

The research is published in Nature Geoscience.