The Atlantic Meridional Overturning Circulation (AMOC) stability landscape is commonly investigated with single-realization hysteresis diagrams driven by freshwater input in the North Atlantic Ocean. However, the effect of CO2 forcing on one side and the role of internal climate variability on the timing of tipping and the AMOC hysteresis on the other side remain less explored. Here, we address this gap by running three independent AMOC hysteresis simulations, consisting of a slow ramp-up plus ramp-down in the CO2 concentration (0.2 ppm/year) within the PlaSim-Large-Scale Geostrophic (LSG) intermediate complexity model. We show that the realizations of the CO2-driven hysteresis cycle, and particularly, the timing of the tipping and recovery, are remarkably affected by internal climate variability. In one of the three simulations, we even observe a reversed cycle, where the AMOC recovers at a higher CO2 level than at the collapse point. While statistical Early Warning Signals (EWSs) show some success in detecting the tipping points, we also find that the internal variability in the EWS considerably reduces the predictability of collapse and leads to false positives of an approaching AMOC tipping. We suggest that the AMOC collapse in the presence of internal climate variability may have characteristics that deviate substantially from the behavior seen in simple models and that caution is needed when interpreting results from a single-experiment realization. Our findings highlight the need for a probabilistic approach in defining a “safe operating space” for AMOC stability, since it might not be possible to define a single critical CO2 threshold to prevent AMOC collapse.
Funding
This work has received funding from the Italian Ministry of Education, University and Research (MIUR) through the JPI Oceans and JPI Climate “Next Generation Climate Science in Europe for Oceans”—ROADMAP project (D. M. 593/2016), from the European Union's Horizon 2020 research and innovation program under Grant Agreement No. 820970 (TiPES), and from the European Union's Horizon Europe research and innovation program under Grant Agreement No. 101081193 (OptimESM project)
History
Author affiliation
College of Science & Engineering
Comp' & Math' Sciences
Version
VoR (Version of Record)
Published in
Chaos: An Interdisciplinary Journal of Nonlinear Science