The media in this post is not displayed to visitors. To view it, please log in.
Although it can be hard to tell from looking at the often placid waters of the Earth’s oceans, their currents carry immense amounts of water around the globe on a daily basis, underlying a dynamic system that – much like the Earth’s atmosphere – plays a major role in everything from weather systems to local climates and ecosystems.
Of all these ocean currents the Atlantic meridional overturning circulation (AMOC) is perhaps the most famous, as it is basically the sole reason why Europe has the mild climate that it does today, courtesy of it carrying thermal energy from the equator all the way to the coast off Scandinavia.
Although collapsing an ocean current seems as improbable as stopping the jet streams in the upper atmosphere, it’s actually significantly easier due to how much ocean currents rely on factors that we can fairly easily influence. Over the past decades we have seen worrying signs that the AMOC is indeed weakening, with the million-dollar question being what scenario we’ll be looking at.
While collapsing the AMOC within a decade may be theoretically possible, current models seem to point towards a weakening by about half by the end of this century, with a recent research article by Valentin Portmann et al. in Science Advances going over the various statistical models to come to this conclusion.
The AMOC
Differences in temperature and salinity drive the ocean currents, causing the transport of water from one area to another as the system seeks to equalize itself. While this may bring to mind the atmosphere’s unrelenting jet streams, or the precarious and harrowing traversal of the area near the Cape of Good Hope where the Agulhas and Benguela currents meet, the AMOC is rather slow and ponderous, taking centuries to circulate.
The North Atlantic Current (NAC). (Credit: Goddard Space Flight Center)
Although it’s obviously a circular system, you can put its beginning at the point of energy input, which for the AMOC is the warm water of the Gulf of Mexico, from where the Gulf Stream flows through the Straits of Florida, following the coast line until it splits up into various smaller currents. The largest of these is the North Atlantic Current (NAC) that provides Europe with warmth and nutrients, while another significant branch is the Canary Current which brushes along the west coast of Africa as part of the North Atlantic Gyre.
As the warmer water travels along the surface of the ocean, it will gradually lose heat to the cooler air, especially once it reaches the North Atlantic. This thermohaline circulation (THC) follows the same pattern around the world’s oceans, with AMOC and the Southern Ocean overturning circulation (SOOC) forming its two main components, distributing heat and nutrients from the equatorial regions to the rest of the world’s oceans.
When the cooling water reaches the limits, the cooling water undergoes a density change that results in it sinking. This is caused by the process of brine rejection, which is the phenomenon where the freezing of saltwater rejects the salts from the forming ice matrix. This produces very salty brine, which is more dense than the surrounding ocean water, ergo it will sink and thereby terminate its branch of the THC.
Salinity Changes
Observed AMOC collapse over time with mild freshwater forcing. (Credit: Van Westen, Oceanography, 2024)
An obvious conclusion one can draw from this brine rejection mechanism is that it can conceivably be interrupted, such as when enough freshwater slows down or disturbs the process. This collapse of the AMOC by freshwater forcing has been the topic of many studies over the years, with a 2024 article in Oceanography by René van Westen et al. investigating evidence that the AMOC is indeed on course for such an event.
At the core of this research are coupled models such as those of the Coupled Model Intercomparison Project (CMIP), which has been developed in phases since 1995. This a cooperative project in which researchers from around the world attempt to create the most complete climate model possible, in order to improve our understanding of the current climate and to make projections about what effects certain changes would have.
The Community Earth System Model (CESM) is one of the contributing models to the CMIP5, using which Van Westen et al. tried to find evidence of a so-called tipping event in the AMOC. What’s notable about their study is that they didn’t attempt freshwater forcing using very large volumes of said freshwater, but still saw a gradual weakening over centuries.
With freshwater forcing consistently reducing the amount of salinity transferred via the NAC, this weakens the effect of brine rejection, thus weakening the AMOC until it eventually collapses. The balance here is the freshwater budget of the Atlantic Ocean, with rivers and melt-off from glaciers and such affecting said budget.
This is where we run into the conundrum that the analysis done by Portmann et al. on the CMIP6 predictions suggest a consistent weakening of the AMOC of about 50% by the year 2100, whereas Van Westen et al. observed a tipping point and rapid collapse of the AMOC to effectively zero using the CESM and a gradual freshwater forcing until said tipping point in the Atlantic freshwater budget is reached.
While a gradual weakening of the AMOC would obviously be bad, a full-blown collapse would obviously be significantly worse, potentially occurring over the span of a few decades and with any recovery deemed either unlikely or taking multiple millennia, as modelled by e.g. Curtis et al. in a 2024 study (free access PDF).
Implications
Whether the AMOC merely weakens by about half or collapses completely, the implications will be severe, with the same 2024 paper by Van Westen et al. providing a good overview, including a summary graphic:
Climate implications of AMOC collapse. (Credit: Van Westen et al. Oceanography, 2024)
Europe in particular would be hit, experiencing far colder seasons along with a sharp drop-off in precipitation. Yet other regions would not be left untouched either, with the Amazon region in particular experiencing a big shift in its climate patterns. In particular the periods when it’d experience cooler, wet weather and vice versa would be flipped, while even Africa and Australia would see a shift in precipitation levels.
In effect, there would likely be severe consequences for the ecosystems in South-America, while Europe would largely turn into a significantly more arid and colder region, similar to e.g. parts of Canada that are on the same latitude. With most of Canada’s population doing its utmost to avoid its more northern latitudes for rather reasonable reasons, it seems fair to say that a full-blown collapse of the AMOC would spell disaster for most European nations.
Keeping The AMOC Healthy
Ice core data estimates of Atmospheric CO2 over the last 800 millennia. (Credit: Tomruen, Wikimedia)
Now that we know the mechanism behind the AMOC and other parts of the THC, the solution to its weakening seems rather obvious: all we have to do is prevent excess freshwater forcing that risks diminishing the Atlantic Ocean’s freshwater budget and we should be golden. Doing so requires identifying the sources of this excess forcing, with a recent study by Oliver Mehling et al. making clear that where the freshwater forcing occurs matters a lot, with many models overestimating the time that we have left until an AMOC tipping point for this reason.
We can also look back on historical climate data courtesy of Antarctic ice cores that go back about a million years, though the Greenland ice cores are the golden standard for e.g. the Dansgaard-Oeschger event that occurred at the end of the last ice age. Since a warming climate naturally results in more freshwater forcing due to meltwater run-off into the oceans, we might be able to find some historical data that shows how the AMOC fared over the past millennia.
What we know is that the AMOC first formed about 34 million years ago when the continents shifted sufficiently to create the THC that we know today began to form. Since then the AMOC has apparently operated continuously, including the repeated glacial periods of the Pleistocene (2.58 million – 11,700 years ago) which ended around the time when the Dansgaard-Oeschger event occurred with an influx of freshwater into the Atlantic.
Global temperature reconstruction of the last two millennia with instrumental temperature from 1880 to 2020. (Credit: Efbrazil, Wikimedia)
Of course, much of this historical data is reconstructed from proxies as part of paleoclimatology, so everything has to be taken with a grain of salt. Even so, we can for a large period directly measure aspects such as CO2 concentration in the atmosphere before we have to resort to proxies. This shows us that to get similar atmospheric levels of that gas in Earth’s history we have to look all the way back to ~16 million years ago during the middle Miocene after atmospheric CO2 levels had gradually come down from 1,600 ppm.
Even as some of us are contemplating direct weather modification, it might thus be an idea to consider the impact of anthropogenic greenhouse gases, as they clearly cause a very rapid increase in the global surface temperature. This warming then increases the melting of glaciers and similar, which in turn increases freshwater forcing into the THC, which thus could result in a sudden AMOC collapse.
Of course, the fun thing about such climate models is that they are only a projection based on our current knowledge. Only in hindsight will we know just how far off the mark we were, but when the stakes are this high, it might not be a terrible idea to err on the side of caution.
Featured image: Illustration of the Atlantic Meridional Overturning Circulation (AMOC). Eric S. Taylor, Woods Hole Oceanographic Institution
hackaday.com/2026/05/27/amoc-a…
Ballistic missile submarines can carry enough nuclear weapons to ruin almost everybody’s day, all at once. Thus, there is a great incentive for novel solutions on how to keep track of them. Credit: 
A Chinese research effort has built a gradiometer of great sensitivity, which lead to widespread speculation around its potential military applications. Credit: 
A graph highlighting the challenge of detecting submarines via gravimetry. In 1989, the best gravimeters might have been able to detect a submarine within 30 meters or so—a militarily useless figure. There have been improvements in technology since, but short of an increase in capability of many orders of magnitude, gravitational methods of detection remain difficult to execute at range. Credit: 
thepwnicorn
in reply to Catalin Cimpanu • • •Akarinium likes this.