The Current That Stopped: Greenland's Ice Is Rewriting Atlantic Circulation

Laura Jackson was reviewing salinity measurements from an autonomous float in the Labrador Sea when she noticed something that made her set down her coffee. It was March 2025, and the float — one of 4,000 scattered across the world's oceans — had recorded a layer of freshwater so pronounced that it should not have existed. 'I thought the sensor had malfunctioned,' Jackson, a physical oceanographer at the National Oceanography Centre in Southampton, told me. She pulled data from neighbouring floats. They showed the same thing. The cold, dense water that normally sinks near Greenland and drives the Atlantic's great conveyor belt was becoming lighter. It was refusing to sink.

What Jackson was seeing was the signature of a climate system crossing a threshold. The Atlantic Meridional Overturning Circulation — AMOC, the planetary pump that carries warm tropical water northward and cold deep water southward — is weakening at a pace that has surprised even researchers who spent decades predicting it would. The question is no longer whether this collapse is happening, but how fast it will proceed and what it will break along the way.

The Pump That Warms Europe

To understand why oceanographers are alarmed, you need to understand what AMOC does. Picture the Atlantic as a vast loop. In the tropics, surface water absorbs heat from the sun and flows northward along the Gulf Stream. By the time it reaches the waters between Greenland and Norway, it has warmed Western Europe's climate by several degrees — the reason London has palm trees while Labrador, at the same latitude, has permafrost. Near the Arctic, that water releases its heat to the atmosphere, grows cold and dense, and sinks nearly two miles to the ocean floor. Then it flows southward again, completing a cycle that takes roughly a thousand years.

This circulation moves more water than all the world's rivers combined — about 17 million cubic metres per second. It redistributes heat, influences rainfall patterns from the Amazon to the Sahel, and drives the productivity of fisheries that feed hundreds of millions of people. When paleoclimatologists study ancient climate swings, they find AMOC's fingerprints everywhere. Twelve thousand years ago, a sudden slowdown triggered a return to ice-age conditions in Europe within a decade.

The thing is, scientists knew this was coming. Climate models have long predicted that global warming would slow AMOC by melting Greenland's ice sheet, flooding the North Atlantic with freshwater that is too light to sink. What they did not predict was the speed. The models said this would happen gradually, over centuries. The data now suggest it is happening within decades.

The Uncomfortable Data

Direct measurements of AMOC began only in 2004, when scientists deployed an array of moorings called RAPID across the Atlantic at 26.5°N. In 2024, a consortium of European research institutions supplemented RAPID with a second array, OSNAP, in the subpolar North Atlantic. Together, they provide an unprecedented picture of how the circulation is changing — and what they show is troubling.

The decline is not smooth. AMOC fluctuates year to year, and some researchers argue the trend is within natural variability. But the 2020–2025 data from the combined arrays show something new: the variance itself is increasing. The circulation is becoming unstable, flickering between stronger and weaker states in ways that were not observed before 2020.

Stefan Rahmstorf has studied AMOC for three decades. When I spoke with him in late March, he was careful to distinguish between what the data show and what they imply. 'We cannot say with certainty that AMOC will collapse,' he said. 'But we can say that the warning signs are appearing faster than our models anticipated. And our models may have been optimistic.'

The Greenland Factor

The culprit is above the ocean, not within it. Greenland's ice sheet contains enough frozen water to raise global sea levels by seven metres. For most of human history, it lost roughly as much ice each year as it gained from snowfall. That equilibrium ended in the 1990s. Since then, the ice sheet has been shedding mass at an accelerating rate — approximately 270 billion tonnes per year over the past decade, according to NASA's GRACE satellite mission.

Here is what this means for the ocean. Saltwater is denser than freshwater. When freshwater from Greenland's glaciers pours into the Labrador and Irminger Seas, it forms a buoyant layer that resists sinking. The cold, dense water that once plunged to the ocean floor now hovers near the surface, blocking the downward leg of the conveyor belt. Laura Jackson's anomalous readings were a snapshot of this process in action.

What makes scientists particularly uneasy is the nonlinearity. AMOC does not weaken gradually and then recover. Paleoclimate records show it has two stable states: on and off. The transition between them can happen within years, not centuries. In 2024, a team at Utrecht University published a study in Science Advances using a high-resolution climate model to calculate when that transition might occur. Their central estimate: between 2037 and 2065, depending on emissions trajectories.

What a Collapse Would Break

The consequences of an AMOC collapse would not be evenly distributed. Europe would cool by several degrees — a bitter irony in a warming world. Britain's average winter temperature could drop by 5°C within a decade. Agricultural patterns built over millennia would fail. The wine regions of France and Germany would shift southward. Growing seasons would shorten across Scandinavia.

But the tropics would suffer more. AMOC's northward flow pulls warm water away from the equator, helping drive the monsoons that sustain agriculture across South Asia and West Africa. Modelling studies suggest an AMOC collapse would weaken the South Asian monsoon by up to 30%, threatening food security for more than a billion people. The Amazon — already under stress — would see reduced rainfall in its eastern reaches, potentially accelerating the dieback that separate research has identified as a looming tipping point.

Sea level rise would also become asymmetric. AMOC's flow currently pulls water away from the eastern United States. If the circulation weakens significantly, sea levels along the U.S. Atlantic coast could rise an additional half-metre beyond what thermal expansion and ice melt would cause — a devastating addition for cities like Miami, New York, and Boston that are already struggling to adapt.

The Scientific Debate

Not everyone is convinced that collapse is imminent. The Intergovernmental Panel on Climate Change's Sixth Assessment Report, published in 2021 and 2022, assigned 'low confidence' to the possibility of an abrupt AMOC collapse before 2100. Some researchers argue that the Utrecht study's model, while sophisticated, uses boundary conditions that may not reflect the real climate system. Others point out that 20 years of direct measurement is too short to distinguish a long-term trend from decadal variability.

'We need to be careful about extrapolating from a brief observational record,' said Penny Holliday, a physical oceanographer at the National Oceanography Centre who was not involved in the Utrecht study. 'AMOC has natural cycles that operate on timescales of decades. What looks like a trend might be part of a longer oscillation.'

But even sceptics acknowledge that the direction of travel is concerning. The disagreement is not about whether AMOC is weakening — the data are clear on that — but about how fast and how far. 'The question is not if but when,' Holliday told me. 'And the honest answer is that we do not know. We are watching a planetary system approach a threshold, and we cannot say precisely where that threshold lies.'

What We Still Don't Know

Laura Jackson has spent the months since her discovery leading a team that is deploying additional sensors in the Labrador Sea, trying to understand whether the freshwater anomaly she observed was a one-time event or the beginning of a pattern. The data are still coming in. Early results suggest the layer has persisted.

The fundamental uncertainty is this: Earth's climate system contains feedback loops that models do not fully capture. Some feedbacks might accelerate AMOC's decline — Greenland melt weakening circulation, which warms Greenland, which accelerates melt. Others might stabilise it — cooling in Europe could shift atmospheric patterns in ways that slow the process. Scientists are watching a vast, slow experiment unfold, and they cannot run it twice.

What the research does make clear is that this is not a future risk to be managed by future generations. The changes are underway now, in data being collected this year, in water masses that are already refusing to sink. The open question is whether the world will treat this as an urgent warning or as another data point in a warming planet's long accumulation of bad news. In the Labrador Sea, the buoys are still transmitting. The measurements are still surprising. And Laura Jackson is still watching, coffee growing cold beside her, as the numbers come in.