New AMOC Model Finds Collapse Risk Depends on Speed of Warming, Not Just Final Temperature

In a climate simulation running at Utrecht University, researchers watched a digital Atlantic Ocean survive 5 degrees Celsius of global warming without its massive circulation system collapsing—so long as the heat was added at a crawl. But when they accelerated the warming to match today's pace, the system broke down at just 2 degrees.[2]

The findings, published in Nature Climate Change, introduce a critical nuance to how scientists understand the Atlantic Meridional Overturning Circulation (AMOC). For years, the focus has been on finding the exact temperature threshold—often estimated between 2.5 and 4 degrees Celsius—that would push the current past a point of no return. The new data suggests that a single, fixed temperature threshold may not exist.[1][2][3]

Instead, the system is vulnerable to "rate-induced tipping." The speed at which the climate changes appears to be just as dangerous as the absolute heat it reaches.[1][2]

To understand why speed matters, it helps to look at the mechanism driving the AMOC. The system acts as a global conveyor belt. It pulls warm, salty surface water from the tropics up toward Greenland and the Arctic. There, the water cools, becomes denser, and sinks to the deep ocean, flowing back south and pulling more warm water north behind it.[3][4]

Global warming disrupts this engine in two ways. First, it heats the surface water, making it lighter. Second, it melts polar ice, dumping fresh water into the North Atlantic. Because fresh water is less dense than salty water, it resists sinking. If the water doesn't sink, the conveyor belt stalls.[3][4]

The Utrecht University team found that if these changes happen slowly, the ocean has time to adapt. "Under slow warming, the entire ocean, from the surface down to its deepest layers, has time to gradually reorganize and adapt to the changing conditions," noted Henk Dijkstra, a professor of dynamical oceanography and co-author of the study.[2]

But when the surface warms rapidly, the deep ocean cannot keep up. The surface and deep layers fail to adjust together, pushing the circulation into a persistently weak state even if a stable flow is still physically possible at that temperature.[1][2]

The researchers tested this by running their model twice. In the first run, they increased atmospheric carbon dioxide slowly, at 0.5 parts per million (ppm) per year. The AMOC remained stable well past 4 degrees of warming, surviving even at 5 degrees. In the second run, they increased CO2 at 2.5 ppm per year—a rate closely mirroring the actual 2.6 ppm annual increase observed between 2015 and 2025. Under this fast-warming scenario, the AMOC collapsed at around 2 degrees of warming.[1][2]

Lead author René van Westen compared the dynamic to driving a car. "If you're driving toward a wall, it makes sense to steer around it. To do that, you need to brake, otherwise you fly off the road," he said. "When it comes to global warming, the world is still pressing extra hard on the accelerator right now."[2]

However, the researchers are explicit about the limits of the evidence. The simulation was deliberately placed in an unusually vulnerable initial state using a large artificial freshwater input to isolate the rate-of-warming effect. The experiment proves that rate-induced tipping is a physical mechanism, but it is not a prediction that the real-world AMOC will definitively collapse at 2 degrees.[1][4]

Furthermore, the observational record remains thin. Continuous direct measurements of the AMOC at 26.5 degrees North only began in 2004. While those instruments have recorded a weakening of about one Sverdrup per decade, the timeframe is too short to separate human-caused decline from natural decadal variability. There is currently no direct observational proof that the AMOC has crossed a tipping point.[1][4]

If rate-induced tipping does apply to the real ocean, it profoundly challenges current climate governance. Many policy frameworks, including aspects of the Paris Agreement, rely on "overshoot pathways." These models assume humanity can safely allow temperatures to temporarily exceed targets, provided we deploy technology to pull carbon out of the atmosphere and cool the planet later.[2][4]

The Utrecht findings suggest that the speed of the overshoot itself could break the system. If the AMOC collapses, the consequences would be severe and effectively permanent on human timescales: Western Europe would face drastic cooling and drying, tropical rainfall belts would shift and disrupt global agriculture, and sea levels would rise faster along the North Atlantic coasts.[2][3]

The data indicates that stabilizing the climate is not just about the final destination. The ocean's survival depends heavily on how hard we hit the brakes along the way.[2][4]