On early-warning of full versus partial Atlantic overturning circulation collapse
Johannes Lohmann · 2025
Read On early-warning of full versus partial Atlantic overturning circulation collapse by Johannes Lohmann (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-17142.txt.
Introduction
Johannes Lohmann asks whether early-warning signals can tell a full collapse of the Atlantic overturning circulation from a partial one. Climate models project a slowdown, and some see an irreversible drop, but the warming threshold is deeply uncertain.
Warning signals read generic statistical shifts as a steady state nears a saddle-node tipping point. Lohmann builds a conceptual model with two regions tipping in sequence, shutting deep water formation one by one. Only very near each tip does the system follow the expected form, so trends differ by observable. That ambiguity makes it hard to say which collapse type comes next, or when, from signals alone.
Catalog record
| CREATOR | Johannes Lohmann |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-17142
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On early-warning of full versus partial Atlantic overturning circulation collapse Johannes Lohmann Abstract Climate models indicate a possible collapse of the Atlantic Meridional Overturning Circulation (AMOC) even for moderate climate change scenarios. There is considerable uncertainty in its likelihood for a given scenario and the critical global warming threshold. An alternative are early-warning signals (EWS) in AMOC fingerprints, which leverage generic statistical properties before destabilization of a steady state (a saddle-node bifurcation). But an AMOC collapse may be a sequence of partial collapses with shutdown of deep water formation in distinct regions. A conceptual model is presented featuring a sequential shutdown in two deep water formation regions. Multiple tipping points are present that do not follow the saddle-node normal form. As a result, the choice of the observable used to monitor EWS dramatically influences the prediction of the collapse via EWS. Various trends in EWS for different observables make it hard to determine what type of collapse (partial or full) will follow. I Introduction The Atlantic Meridional Overturning Circulation (AMOC) is a major pattern in the climate system and a large contributor to the global meridional heat transport, carrying 1.2 PW of heat northward across the subtropics 38 . A collapse of the AMOC would lead to dramatic decreases in temperature especially in Northern Europe, while warming the Southern Hemisphere and producing various cascading impacts 37 . This possibility was pointed out by Stommel in 1961 73 , where a bistability of the AMOC was shown in a conceptual model as a result of the competing positive salt-advection and negative temperature-advection feedbacks. In the model, the strength of the circulation is determined by the opposing influence of atmospheric freshwater and temperature forcing on the meridional density gradient. As the forcing changes, a saddle-node bifurcation can occur where the present-day state loses stability and the circulation enters a weak salinity-driven state. This is a bifurcation-induced tipping point (TP) 6 . Presently, the atmospheric forcing gradients onto the Atlantic are changing due to polar amplification of global warming, as well as increased glacial ice melt and changing precipitation patterns. Both increased warming and freshening of northern Atlantic surface waters could lead to a shutoff of convection and thereby the AMOC’s current downwelling path. Due the likely existence of the salt-advection feedback - consistent with the observation that the Atlantic imports salt 64 ; 3 , whereby an initial AMOC weakening would be amplified in a self-reinforcing way by decreasing surface salinity in downwelling regions - this may happen abruptly and irreversibly. The possibility for AMOC tipping is strongly supported by abrupt North Atlantic climate changes seen in paleoclimate records of the last glacial period 65 ; 32 . I.1 Observational and model evidence for AMOC slowdown and collapse A direct and continuous monitoring of the strength of the AMOC has only been in place for two decades. The RAPID-MOCHA array (at 26.5 ∘ N, since 2004) and the OSNAP array (55-60 ∘ N, since 2014) show considerable variability (Fig. 1 a) such that no significant trend in AMOC strength can be inferred as of now 38 ; 48 . Five hydrographic sections in the period 1957-2004 give snapshot measurements that do indicate an AMOC decline 18 , but it has argued that a large part of this signal can be explained by aliasing and seasonal bias in these snapshots 39 . It has been attempted to empirically reconstruct the AMOC strength further back in time. Exploiting the correlation of RAPID AMOC strength and sea level anomalies 28 an AMOC weakening of about about 1 Sv in the period 1993-2014 was estimated (green line in Fig. 1 a), which is however not statistically significant and does not include the reversal of the trend in the more recent RAPID data. A correlation of boundary density anomalies finds no AMOC decline during 1981-2016 88 , and neither does a model-assisted reconstruction based on air-sea heat fluxes of the past 60 years 75 . Figure 1: a Time series of observed and reconstructed changes in the AMOC circulation strength. Included are direct AMOC strength measurements from the RAPID array (red) and five hydrographic sections (blue dots) 18 ; 39 , as well as a reconstruction based on sea level anomalies (green) 28 . The longest time series (black) is the sea surface temperature fingerprint 63 sometimes used as proxy for AMOC strength. Shown is the anomaly of sea surface temperatures in the North Atlantic subpolar gyre with twice the global mean sea surface temperature anomaly to correct for global warming and polar amplification, as was used in 25 . b Climatology of the mixed layer depth for the month of March from the GLORYS reanalysis (period 1993 to 2024). Much longer AMOC reconstructions based on sea surface temperature (via its sensitivity to AMOC heat transport) and other paleo-oceanographic constraints suggest that the AMOC has been weakening in the industrial era 63 ; 76 ; 21 ; 19 . A “fingerprint” based on subpolar gyre temperatures is shown by the long time series in Fig. 1 a. But it is debated whether all available data support such inferred AMOC trends 53 ; 30 ; 20 , and whether the sea surface temperature signal in relation to the AMOC is significant given natural variability 40 . Climate models show on average no AMOC weakening when simulating the historical period 85 , or even a slight increase 51 that may be dominated by certain models with unrealistic response to aerosol forcing 66 . Model averages remove internal variability, and it has been argued using ensemble simulations with a single model that individual simulations showing an AMOC weakening best match observational constraints, meaning that a weakening may have indeed occurred, but may have been pure centennial internal variability 16 . Most model projections into the future show a significant AMOC decline as a result of accelerating global warming 85 . Due to model biases, which result in a spread in present-day AMOC strength ranging from 9 to 25 Sv 50 , models differ substantially in the magnitude of this decline 15 . However, given the long time scales of ocean dynamics, the decline until 2100 in CMIP6 is still a transient response. It has become clear that models with extended simulations until 2300 or longer can in fact transition to a state where North Atlantic deep water formation ceases and the AMOC becomes very shallow and weak, with downwelling restricted to the subtropics 67 ; 26 . This happens for all models forced with high-emission (SSP585) scenarios, and also some with intermediate (SSP245) and even low emission (SSP126) scenarios 26 . The uncertainty remains high even if all models would agree, as can be seen by ensemble simulations with a single model where some ensemble members in intermediate scenarios reach a weak AMOC state and other do not 67 ; 26 . Thus, from current models one cannot exclude an AMOC shutdown for all emission scenarios due to global warming alone, where the shutdown of convection is likely initiated by the effect of thermal forcing on buoyancy and subsequent salt-advection feedback 26 . The risk is increased by accerelating Greenland melt 8 , which is missing in most projections and