墨西哥湾流在一次古代寒潮期间突然向北移动——这对我们的未来是一个警示。

Around 13,000 years ago, as the world was emerging from the grip of the last ice age, much of the North Atlantic region plunged back into near-glacial conditions.

Sea ice expanded across the North Atlantic, reaching as far south as the Shetland Islands. Glaciers began to regrow in the Scottish Highlands, while winter temperatures across Europe and North America plummeted. Yet off the coast of Atlantic Canada, the ocean did the opposite.

In our new study, published in the journal Nature Communications, we found evidence that waters off Nova Scotia, Canada, warmed as the Gulf Stream shifted hundreds of kilometres northward, while deep circulation also changed.

It is the first direct evidence that this vital current responded in such a way during a period of abrupt climate change that rearranged Atlantic Ocean circulation.

The finding lends support to the climate models that predict a similar northward shift in the future if the Atlantic Meridional Overturning Circulation (Amoc) weakens – a trend that has probably already begun.

Why the Gulf Stream matters

The Gulf Stream transports warm tropical waters northwards along the eastern coast of North America before turning north-east towards Europe. In doing so, it forms part of the Amoc, a vast system of ocean currents that redistributes heat, nutrients and carbon around the Atlantic Ocean. Consequently, the Amoc plays a major role in regulating the climate. In particular, the northern arm of the Gulf Stream helps keep western Europe much milder than other regions at similar latitudes.

Scientists are increasingly concerned about the future of this circulation system. As the climate warms and extra freshwater (from melting ice) enters the North Atlantic, surface waters become less dense and therefore less able to sink. Most climate models project that these changes weaken the Amoc. Observations suggest that this weakening has already begun, but it is predicted to weaken much more as the 21st century progresses. However, direct evidence showing how the system responds to such major disruptions remains relatively limited.

To answer that question, paleoceanographers like us turn to the past.

A natural experiment from the end of the last ice age

The Younger Dryas was one of the most dramatic episodes of abrupt climate change in Earth’s recent history. As the planet emerged from the last ice age, warming trends across much of the North Atlantic region abruptly reversed. European summer temperatures declined by around 4°C–8°C in less than a century, while Greenland cooled by up to 10°C within just a few decades. The effects rippled far beyond the North Atlantic, weakening monsoon systems across Africa and Asia.

To understand how the ocean responded, we analysed sediment extracted from the seabed off Nova Scotia. Microscopic fossil shells and sediment grains preserved within this marine mud can reveal what the sea would have been like at the time it formed. We then reconstructed changes in both surface and deep Atlantic circulation before, during, and after the Younger Dryas.

An unexpected warming signal

What we found surprised us. While Greenland and much of the subpolar North Atlantic cooled rapidly, waters off Atlantic Canada warmed instead, by as much as 4°C–5°C.

The most likely explanation is that the Gulf Stream migrated northwards, bringing warm subtropical waters closer to the Canadian coastline.

Previous climate-model simulations had predicted that a weakening of one of the Amoc’s deep currents could trigger exactly this response. Until now, however, there had been little direct geological evidence that it had happened before.

Our study provides real-world evidence for a process that climate models have long proposed. That matters because it shows that large reorganisations of Atlantic circulation are not just theoretical possibilities – they have happened before.

What can the past tell us about the future?

No past climate event is a perfect analogue for modern climate change. The Younger Dryas occurred under very different conditions from today. Massive ice sheets still covered much of Canada and Scandinavia, and the sea level was tens of metres lower than at present.

Nevertheless, the physical links connecting the different components of the North Atlantic circulation system are likely to be the same.

Our study does not suggest that the Amoc completely collapsed during the Younger Dryas, nor does it tell us whether such a collapse is likely in the future. Instead, it reveals a more nuanced picture in which various components of the North Atlantic circulation system changed in different ways. Rather than producing a uniform response, this reorganisation created a patchwork of warming and cooling across the North Atlantic.

Similar patterns have also emerged over the last 150 years, with a relative “warming hole” developing in the ocean south of Greenland while regions closer to the Gulf Stream have warmed more rapidly. Our findings provide real-world evidence that these contrasting patterns are closely linked to changes in ocean circulation.

Looking to the future, scientists are concerned that continued human-caused warming could trigger major changes in North Atlantic circulation, leading to shifts in ocean temperature patterns, which would disrupt weather and climate across the globe. Examining how the Atlantic behaved 13,000 years ago can help us recognise the warning signs of major changes before they happen again.

Critically, our study suggests that such reorganisations can unfold over about a century, with individual components of the circulation changing within just a few decades – within a human lifetime.

By showing how different parts of the Atlantic circulation interacted during a past episode of abrupt climate change, our findings provide an important benchmark for testing climate models. The deeper understanding we have gained into how the interconnected Atlantic system behaves will also help us with the very challenging task of developing early-warning systems for future circulation changes and potential climate tipping points.

Alice Carter-Champion receives funding from the Natural Environment Research Council (NERC) , the Royal Society’s project “Rethinking Palaeoclimate for Society” and the Leverhulme Trust.

Fangjingcheng Zhu receives funding from the INSPIRE NERC Doctoral Training Partnership and NERC project ReconAMOC.

Jack Wharton receives funding from the Natural Environmental Research Council (NERC) , the European Union’s Horizon Europe project EPOC, and the Advanced Research + Invention Agency (ARIA) project VERIFY.