Atlantic circulation collapse? New clues on the fate of a crucial conveyor belt » Yale Climate Connections

Among the many reasons why we ought to cut climate-warming greenhouse gas emissions as quickly and sharply as possible, the weakening of a system of ocean currents known as the Atlantic Meridional Overturning Circulation, or AMOC, ranks high indeed. Several key scientific papers over the last couple of years have put this long-percolating climate concern back on the front burner. It’s hard to overstate how widespread and calamitous the impacts could be if this conveyor belt were to collapse – and it’s a process that could begin in the next several decades, if the new work is on target.

The AMOC is a vast oceanic loop that carries warm water northward through the uppermost Atlantic toward Iceland and Greenland, where it cools and descends before returning southward. It ferries colossal amounts of heat from the tropics toward the polar region, helping balance Earth’s climate machine just as it’s done on and off for millions of years.

“On and off” is a key phrase here. It’s the “off” phase – when the AMOC’s flow is greatly reduced, and landscapes surrounding the subpolar North Atlantic get sharply colder – that concerns many people, including the world’s leading AMOC experts.

As best we can tell from the limited data at hand, together with model-based estimates, the AMOC has already slowed by a few percent since the mid-20th century after roughly 1,600 years of relative stability. We also know that the AMOC has collapsed to a near-halt multiple times over the past 2 million years, most recently from about 12,000 to about 11,000 years ago. These epic “off” phases can have profound impacts on large parts of the globe.

Lately, AMOC’s future has become a still more fraught topic. Newly leveraged statistical techniques and modeling approaches have been looking into potential warning signs of AMOC collapse. If these indicators prove robust, we might get notice years or decades in advance of a tipping point that, in turn, would push the AMOC into a century-long collapse process. And such a tipping point could arrive sooner than conventional wisdom held not so long ago.

Below, we’ll summarize how AMOC works and what recent observations tell us about it. In part two of this two-part post, we’ll dig deeper into new research on when, and whether, an imminent AMOC collapse might become evident through various types of early-warning data.

(Mostly) on and off: The AMOC’s speed settings

At its heart, the AMOC is a heat redistribution device. It takes some of the excess warmth that builds up at Earth’s sun-drenched tropical latitudes and spreads it poleward across the North Atlantic. It’s part of a global conveyor belt that extends throughout the world’s interlinked oceans. This is sometimes called the global thermohaline circulation because of the variations in temperature (“thermo-”) and salinity (“-haline”) that drive the system.

As water flows north across the Atlantic, evaporation from the surface makes it progressively saltier. The current then chills as it flows from the tropics toward the far North Atlantic off Greenland. Since colder, saltier water is denser than warmer, fresher water, the flow sinks and returns southward at depth.

For North America and Europe, the Gulf Stream is by far the most crucial part of the AMOC. Not only does the Gulf Stream’s warmth keep northern Europe far milder than other locations at such high latitudes, but the flow itself serves as a type of protective seawall for North America’s Atlantic coast. Because of the way ocean height and ocean currents interrelate, sea level can be several feet lower toward the U.S. East Coast, on one side of the Gulf Stream, than it is toward the central North Atlantic, on the other side.

Like most weather and climate features, AMOC varies on timescales ranging from the brief (hours and days) to the geologic (thousands of years). But a simple climate model developed more than 60 years ago was the first to show that AMOC also has two favored long-term speed settings: essentially, “on” and “off” (or “strong” and “weak” modes). An AMOC collapse is a shift from the “on” to the “off” mode.

What makes the AMOC fizzle? Meltwater pouring from the Arctic into the far North Atlantic in massive amounts seems to be capable of triggering AMOC collapse. It does so by overtopping and displacing saltier water from the AMOC with fresher meltwater that floats atop the ocean rather than joining AMOC’s descending branch. The result: stopping up the AMOC’s south-to-north flow, just as a stalled car can lead to a miles-long traffic jam upstream.

This basic idea – that the AMOC can hum along for many thousands of years, then collapse and stay weak for hundreds of years before a gradual recovery – has proven durable. It shows up time and again in model experiments that “hose” the subpolar North Atlantic with vast amounts of meltwater.

What’s more, markers of prehistoric climate frozen in Greenland ice cores confirm that the region got drastically colder during AMOC shutdowns.

The playing out of an AMOC collapse – which models and paleoclimate data suggest would take a few decades to a century once a tipping point is reached – would destabilize our planet’s climate and complicate life for many millions if not billions of people. Surface air temperatures over northwestern Europe could plunge far below current readings, even in a world otherwise warmed by human-produced greenhouse gases. Everything from agriculture to migration patterns could be profoundly affected. More modest drops in annual temperature, mainly during winter, could spread as far as the United States as well as encompassing most of Europe. And the average sea level could jump by as much as a meter (3.3 feet) along the U.S. East Coast, on top of the higher water produced by warming oceans and melting ice.

The details and timelines of such impacts are still in flux, with much still to investigate. This includes not only the potential endpoint climate after an AMOC collapse but wild gyrations in weather and climate patterns that could play out during a century of collapse itself. And of course, all this would occur in tandem with fossil-fuel emissions already pushing the globe toward warming-related extremes and instabilities. In short, it’s a climate scientist’s nightmare come true.

How is the heart of AMOC behaving right now?

There’s a stark contrast between the colossal trouble an AMOC collapse might bring and the relatively mellow state of AMOC at present. To be sure, there’s plenty of variability in AMOC from day to day, week to week, and year to year – something that wasn’t recognized until a full-scale sampling effort called RAPID got underway more than two decades ago.

As global interest in climate change surged in the 1990s, scientists realized there was a pressing need to monitor AMOC in enough detail to keep an eye on short-term ups and downs and eventually to capture long-term trends. The United Kingdom’s National Oceanography Centre teamed up with the University of Miami and NOAA to launch what became the RAPID program. The team settled on 26 to 27 degrees north as a workable latitude for building cross-sections through the south-to-north upper flow and the north-to-south return flow.

On both sides of the North Atlantic, west and east, RAPID maintains vertical moorings – floating, anchored observing systems – that measure temperature and salinity each hour down to thousands of feet below the sea surface. These data are combined with surface wind measurements to estimate wind-driven oceanic flow. Additional measurements are collected in the Florida Current, the crucial segment of the Gulf Stream between Florida and the Bahamas that represents the vast majority of AMOC transport at this latitude (see Fig. 3).

The strength of the Florida Current has been inferred since 1982 from an undersea telecommunications cable that stretches from Florida to the Bahamas (cyan line in Fig. 3 above). As the Florida Current flows past the cable, Earth’s magnetic field leads to measurable variations in voltage. These variations can be calibrated against other instruments and employed to measure the strength of the current.

Since around 2000, a weakening trend in the cable voltage has been observed, as noted in the annual State of the Climate reports from the Bulletin of the American Meteorological Society. By the last several years, the record had become long enough to make the slight weakening statistically significant. However, the weakening has been at odds with an absence of significant change from other observing tools, such as instrument packages dropped from the sea surface and altimeter readings.

The apparent contradiction has been resolved, as reported in a September 2024 paper in Nature Communications. The gist is in the title: Florida Current transport observations reveal four decades of steady state. As it turns out, the cable data hadn’t been adjusted since 2000 to account for a gradual shift of Earth’s magnetic pole away from the North Pole that has steadily weakened the geomagnetic field affecting the submarine cable.

Once corrected and recalibrated, the cable data were in much closer agreement with the other platforms, showing no significant change in the Florida Current during the past 40 years of measurements. And because the Florida Current is such a large part of AMOC, the corrected data cuts the observed decline from 2004 to 2022 in total AMOC transport almost in half, from an original drop of about 14% to a revised drop of about 8%.

The authors – led by Denis Volkov at NOAA/University of Miami Cooperative Institute for Marine and Atmospheric Studies – found the revised 8% drop to be only marginally significant. “We have not heard yet from modelers, but I do not think our revised result strongly affects their conclusions, because they deal with much longer timescales compared to the period of observations,” Volkov said in an email.

Even though the new data would seem to diminish the gravity of recent AMOC trends in the North Atlantic, a decrease is still a decrease. And Volkov and colleagues stressed the importance of maintaining programs such as RAPID:

… the likelihood of a future AMOC slowdown and the importance of both the [Florida Current] and the AMOC in the regional and global climate variability emphasizes the value of sustained observations in the Florida Straits and in the subtropical North Atlantic at ~26.5°N. The existing records are just starting to resolve decadal-scale signals relevant to climate variability.

AMOC data at 26N: updated to February 2023 are now freely available on our website at https://t.co/wUULdzQP1I @NERCscience
@NSF
@NOAA
@NOCnews
@UKRI_News pic.twitter.com/yeTPtcln2N

— RAPID (@RAPID_AMOC) September 20, 2024

Observationalists weigh in

The observations-oriented scientists who assess AMOC in the wild and confront the raw data agree that AMOC will weaken as the planet warms, but they tend to be cautious about putting out timelines for AMOC collapse.

Ben Moat, the co-chief scientist of RAPID, posted these views in September 2024 in a news release from the UK National Centre for Oceanography:

Prior to starting the RAPID project in 2004, changes in large-scale ocean circulation were thought to happen very slowly, perhaps on timescales of 100 years. Observations made within the first year of the RAPID array showed the ocean circulation changed on hourly and daily timescales. The results made fundamental changes in the way we understand how the ocean circulates heat around the planet. While we have made some revolutionary steps already, the big unanswered questions are about the predicted weakening of the AMOC. We think it will weaken, but by how much and when is still uncertain.

Eleanor Frajka-Williams, head of experimental oceanography at the University of Hamburg, was among the organizers of a 2023 workshop on AMOC observation needs in a changing climate. Summarizing the mood of the meeting, she posted on X: “From the observational (rather than proxy) record, a majority agreed that we don’t yet know how the AMOC will respond to future anthropogenic [human-caused] change.”

A group of experts from the European Union’s EPOC (Explaining and Predicting the Ocean Conveyor) project also weighed in this year, aiming to put recent papers in perspective:

Over the years the pendulum has swung from a marked AMOC decline/shutdown being considered likely to this being an unlikely scenario. Currently, the prevalent view is somewhere between the two. … The AMOC’s likely future fate remains an important question, though one that we cannot yet answer based on our current level of understanding.

So in summary, we have at least some reassurance from the North Atlantic data that a full-on AMOC collapse hasn’t begun. And it’s unlikely that any future collapse would reach its end point any sooner than the early to mid-2100s. Yet there’s also legitimate concern – stoked by recent work from climate modelers and statisticians – that a tipping point toward eventual collapse could arrive as soon as the next several decades, especially if fossil-fuel emissions aren’t cut sharply. In part two of this two-part post, we’ll look at some of the new research on early-warning signs of AMOC collapse, what those scientists and other assessments are telling us about the threat, and how we can help limit the odds of an AMOC collapse happening in the first place.

Jeff Masters contributed to this post.