Failure to track a stable AMOC state under rapid climate change

The Atlantic Meridional Overturning Circulation (AMOC) is a tipping element of the climate system. The current estimate of the global warming threshold for the onset of an AMOC collapse is +4C. However, such a threshold may not be meaningful because AMOC stability depends on the rate of radiative forcing and background climate state. Here, we identify an AMOC stabilising mechanism that operates on timescales longer than present-day radiative forcing increase. Slow forcing permits coherent adjustment of surface and interior ocean properties, supported by enhanced evaporation and reduced sea-ice extent, counteracting destabilising feedbacks. This mechanism is explicitly demonstrated in a slow CO2 increase experiment (+0.5 ppm/yr), in which the AMOC remains stable up to +5.5C of global warming. By contrast, under intermediate- and high-emission scenarios, the AMOC collapses at substantially lower warming levels (+2.2C and +2.8C, respectively). Our findings demonstrate the strong radiative forcing path dependence of AMOC tipping and imply that limiting the rate of radiative forcing is critical for reducing the near-term risk of an AMOC collapse.

💡 Research Summary

The Atlantic Meridional Overturning Circulation (AMOC) is widely recognised as a potential climate‑system tipping element, and many studies have quoted a global‑mean surface temperature (GMST) threshold of roughly +4 °C above pre‑industrial levels for the onset of an AMOC collapse. This paper challenges the usefulness of a single temperature threshold by demonstrating that AMOC stability is strongly dependent on the rate of radiative forcing and on the background climate state.

Using the Community Earth System Model (CESM) version 1.0.5 and a suite of CMIP6 models, the authors conduct four complementary experiments. First, a quasi‑equilibrium “hosing” simulation imposes a prescribed fresh‑water flux (F_H) over the North Atlantic while conserving global salinity. Two statistically steady states are identified for F_H = 0.18 Sv (weaker AMOC) and F_H = 0.45 Sv (stronger AMOC). The stronger state lies closer to the salt‑advection feedback regime and is therefore more susceptible to perturbations.

Second, the historical forcing (1850‑2005) is followed by three Representative Concentration Pathways (RCP2.6, RCP4.5, RCP8.5) that are continued with fixed greenhouse‑gas concentrations after 2100 out to the year 2500. In these extensions, the AMOC recovers under RCP2.6 but collapses under RCP4.5 and RCP8.5. The collapse occurs when the GMST anomaly reaches +2.2 °C (RCP4.5) and +2.8 °C (RCP8.5), respectively, providing a lower bound for a temperature‑based tipping threshold. Crucially, the same AMOC state (F_H = 0.45 Sv) remains stable under RCP2.6, showing that the threshold is not universal but depends on the forcing pathway.

Third, the authors design a “CO₂ ramp” experiment in which atmospheric CO₂ is increased linearly at a very slow rate of +0.5 ppm yr⁻¹ (≈0.176 % yr⁻¹) for 1 750 model years, while all other forcings stay at pre‑industrial levels and F_H remains at 0.45 Sv. This slow ramp drives the GMST anomaly to +5.5 °C—well beyond the RCP8.5 warming—yet the AMOC stays robust throughout the entire simulation. The result directly contradicts the notion that a specific temperature anomaly inevitably triggers a collapse.

To understand why the slow ramp preserves AMOC, the authors analyse oceanic structure in depth and density coordinates, meridional heat transport, and water‑mass transformation (WMT). In the early 300 years, AMOC weakening is driven primarily by reduced surface heat loss (thermal buoyancy flux, Ψ_T). However, as the simulation proceeds, enhanced surface evaporation and reduced sea‑ice extent increase surface salinity, generating a compensating haline buoyancy flux (Ψ_S). The combined effect raises the density of the upper North Atlantic, offsetting the warming‑induced density reduction.

A key diagnostic is Ψ_NADW, the surface‑forced estimate of the NADW (North Atlantic Deep Water) transport derived from surface buoyancy fluxes between 40° N and 65° N. Ψ_NADW tracks the actual AMOC strength closely and can be decomposed into thermal (Ψ_T) and haline (Ψ_S) components. In the CO₂ ramp, Ψ_T initially declines, but Ψ_S grows sufficiently to bring Ψ_NADW back to its pre‑ramp magnitude. Simultaneously, the density level of maximum AMOC strength (σ_max²) and the depth of the surface sinking region shift together, preserving the adiabatic overturning pathway. This coordinated adjustment is absent in the rapid‑forcing RCP runs, where the surface sinking region lightens faster than σ_max², leading to a breakdown of the adiabatic pathway and eventual collapse.

The study therefore demonstrates a clear “forcing‑path dependence” of AMOC tipping: rapid radiative forcing overwhelms the ocean’s ability to adjust surface buoyancy fluxes, amplifying the salt‑advection feedback and precipitating collapse at relatively modest warming. In contrast, a slow, gradual increase allows the ocean‑atmosphere system to re‑equilibrate, strengthening the haline feedback and maintaining a stable overturning even under high absolute warming.

Implications are profound for climate‑policy risk assessment. Relying solely on temperature thresholds underestimates the risk posed by high‑emission trajectories, because the same temperature can be reached via very different forcing histories with dramatically different AMOC outcomes. Conversely, limiting the rate of CO₂ emissions—effectively flattening the radiative‑forcing trajectory—offers a mitigation lever that is independent of, but complementary to, temperature‑target strategies. The authors conclude that emission‑rate constraints should be explicitly incorporated into climate‑risk frameworks to reduce near‑term AMOC collapse probability.