Complementary studies by separate teams have explored the interactions between melting ice in the North Atlantic and the flow of a crucial ocean current. One intensifies the alarm many oceanographers have already expressed: that more rapid melting will cause a crucial part of the Gulf Stream system to slow or even stop, with disastrous consequences. The other reveals that if that happens, one of the few beneficial effects will be a decrease in the Arctic melting that caused the problem – but probably not enough.

Some aspects of climate change are simple: for example, more carbon dioxide and certain other gases released means heat gets trapped closer to the Earth’s surface, making the lower atmosphere warmer. However, others involve complex interactions between phenomena, so that changes to one cause cascading effects that are hard to model. Changes to the Earth’s interconnected system of ocean currents are one of the most difficult but most important of these, and newly published papers look at it from different angles.

First to the more familiar, and scarier, part. A warmer world means faster ice melt, particularly since a key aspect of the enhanced Greenhouse effect is to raise temperatures faster at the poles. Ice melt is fresh, so despite being cold it doesn’t sink like cold water normally does in the ocean, instead sitting on top of the saltier, denser water below.

That’s potentially a big problem, because sinking cold water helps drive the Atlantic Meridional Overturning Circulation (AMOC), which is the most vulnerable cog in the global thermohaline circulation of ocean currents. For years, studies have gone back and forth on how much melting is affecting this, and how soon the consequences will get serious. Last week, 44 leading climate scientists called on the world to stop playing Russian roulette with the possibility.

One way to explore this is to look at what happened during other warming periods, and Dr Mohamed Ezat of iC3 Polar Research Hub is part of a team that has bad news. “Our finding that enhanced melting of Arctic sea-ice likely resulted in significant cooling in northern Europe in the earth’s past is alarming,” Ezat said in a statement. “This reminds us that the planet’s climate is a delicate balance, easily disrupted by changes in temperature and ice cover.” 

The conclusion is based on measurements from the early part of the last interglacial era, 128,000-126,500 years ago. The team collected sediment cores from locations midway between Norway, Scotland, and Iceland and looked for evidence of whether summers were ice-free in the area, allowing ocean life to flourish briefly each year. In combination with oxygen isotope ratios, and ratios of barium to calcium in microorganism shells, this creates a picture of the local climate of the era.

They found that at a time when the world as a whole was unusually warm, the Norwegian Sea was colder, which they attribute to reduced warm water delivered by AMOC. By causing cold water to flow south at great depths, AMOC not only affects the North Atlantic, but starts a chain reaction of global currents, which could suffer if it stops.

Ocean currents redistribute warmth around the globe, and without them the poles would be much colder and the tropics hotter. A disruption in this flow makes both more unpleasant for humans, and the steeper temperature gradient could also lead to more intense storms. 

Northern Europe is the greatest beneficiary of this, far warmer in winter than any other region at similar latitudes because of warm water flowing up from the equator. If the interruption is repeated, parts of Europe could become uninhabitable from cold at the same time as people are fleeing other regions from too much heat or rising sea levels.

“We hope that our study provides a benchmark for climate modelers to utilize this time period to better constrain the impacts of ice changes on regional and global climate,” Ezat said.

On the other hand, if the currents stop bringing warmth north, this might slow the melting and therefore restart the currents. Climatologists who have questioned the scale of the problem rely on this, but working out the size of the effect is very hard.

University of California Riverside graduate student Yu-Chi Lee is part of a team that has tried. They calculate that a slowing AMOC could keep Arctic temperatures 2° C (3.6° F) cooler than they would be in 2100 if AMOC maintained current flow. That sounds encouraging – until you realize that under their calculations the region warms by a shocking 8° C (14.4° F) instead of a catastrophic 10° C (18° F).

“The AMOC is a critical component of our climate system because it moves heat around the globe,” Lee said in a different statement. “We found that its weakening reduces the amount of heat reaching the Arctic, which slows down the rate of warming.” The moderation in temperature rise is a result of multiple factors, such as reduced sea-ice loss and changes in cloud cover.

If greenhouse emissions stop, and particularly if we find viable ways to draw earlier emissions out of the atmosphere, melting might slow, potentially restoring currents. Something similar probably happened in the era Ezat and colleagues studied, although the timing is still unclear. However, if drivers of a hotter world keep rising fast, a slower AMOC will probably be insufficient to prevent ice melt until Greenland has no more ice to lose.

Meanwhile, Lee and colleagues note evidence that without AMOC the Intertropical Convergence Zone would move south, potentially cutting much of the population of Africa off from the rains they need to farm.

“The AMOC slowdown may offer some temporary relief in the Arctic, but this is not a simple good-news story,” Lee’s co-author Dr Wei Liu said. “The overall impact on ecosystems and weather patterns, both in the Arctic and globally, could still be severe.”

The paper by Ezat et al is published open access in the journal Nature Communications. The paper by Lee, Liu et al is published in the journal Proceedings of the National Academy of Sciences.