Penn State researchers report that the link between the North Atlantic “cold blob” (also called the North Atlantic warming hole) and the Atlantic Meridional Overturning Circulation (AMOC) is more complex and phase- dependent than previously assumed.
The cold blob is a roughly 965,000- square- mile region south and southeast of Greenland that has cooled by as much as 1.8°F over the past century while most other ocean waters have warmed.
It has long been linked to a weakening AMOC, the ocean “conveyor belt” that transports warm surface water northward and denser cold water southward in the deep ocean. A sustained AMOC decline (influenced by climate change) is viewed as a key driver of the cooler surface temperatures there.
In a Nature Communications paper, a team led by Laifang Li (assistant professor of meteorology and atmospheric science) and including lead author Yifei Fan (doctoral candidate) used climate model simulations across many scenarios.
These incorporated varying greenhouse- gas levels, past and present climate conditions, ocean heat transfer, ocean– atmosphere heat exchange, and atmospheric radiation (including low -cloud effects). The work builds on their 2025 finding that atmospheric radiation also helps keep the region cooler.
The cooling effect of a weakening AMOC on surface temperatures is not constant:
- When the current weakens from its greatest strength, surface temperatures drop only modestly.
- When it weakens from mid- range strength, the temperature drops roughly triples, and the cold blob expands across the subpolar basin.
- When the current weakens further from an already diminished state, the cooling effect fades; the cold blob shifts northeast toward Iceland and the Norwegian Sea. Variable atmospheric heat radiation and low- cloud reflections contribute to these shifts.
Models project that AMOC will move from its current higher- strength phase toward its lowest- strength phase by 2100. The cold blob’s most significant temperature shift could occur as late as 2067 or could already have begun as early as 2021.
Earlier peaks in temperature change would correspond to an earlier fading of the cool patch. Once AMOC is weak enough, the ocean is expected to absorb more heat from the atmosphere and approach its capacity to buffer climate change, increasing uncertainty in future surface temperatures, currents, weather and climate effects.
The North Atlantic “cold blob” (also called the North Atlantic warming hole) is a region south and southeast of Greenland that has cooled or lagged substantially behind global ocean warming over the past century (by up to ~1–1.8°F / ~0.5– 1°C in some estimates).
It stands out as one of the few areas resisting the broader warming trend.
The Atlantic Meridional Overturning Circulation (AMOC) acts as a major “conveyor belt”: it carries warm, salty surface water northward into the subpolar North Atlantic, where the water cools, becomes denser, sinks, and returns southward at depth. A weakening AMOC reduces northward ocean heat transport (OHT) into this region, which is widely hypothesized as the primary cause of the cold blob.
Recent observational analyses (including full- depth ocean heat content data back to the 1950s and satellite records) support this view.
They show that the cooling extends below the surface, surface heat loss from the region has actually decreased in recent decades, and multidecadal heat-content variations correlate more strongly with horizontal ocean heat transport changes than with atmospheric surface fluxes.
This points to reduced heat delivery by the AMOC rather than increased atmospheric extraction of heat.
Climate models and reconstructions generally link the blob’s emergence to anthropogenic forcing (primarily greenhouse gases, with contributions from aerosols and Greenland meltwater freshening that inhibits deep convection).
Direct AMOC measurements (e.g., RAPID array since ~2004) and indirect indicators also indicate a weakening trend over recent decades, though natural variability remains significant.
A 2026 Nature Communications study (Fan, Chan, Zhang, Li et al., using large ensembles of CESM2 simulations across paleoclimate to future scenarios, plus other CMIP6 models) demonstrates that the AMOC- subpolar SST relationship is not stationary.
It shifts across three regimes defined by AMOC strength (change points near ~20 Sv and ~11.5 Sv):
- Regime 1 (strong AMOC, > ~20 Sv): Low sensitivity (~0.09 °C per Sv). Classic dipole fingerprint—cooling in the central subpolar basin paired with warming along the Gulf Stream extension. The cold blob begins to emerge. Oceanic, clear-sky radiative, and cloud processes contribute roughly equally and modestly.
- Regime 2 (intermediate AMOC, ~11.5– 20 Sv): Sensitivity more than triples (~0.31 °C per Sv). Cooling intensifies and expands across the entire subpolar basin (strongest in the Irminger and Labrador Seas). This is the “maturing” stage of the warming hole. The amplification is driven primarily by atmospheric radiative processes: reduced clear- sky longwave radiation (from cooler and drier troposphere) and increased shortwave reflection by low- level clouds. Ocean heat transport changes are largely offset by compensating turbulent fluxes.
- Regime 3 (weak AMOC, < ~11.5 Sv): Sensitivity drops again (~0.12 °C per Sv). The canonical cold blob decays or shifts northeastward toward Iceland and the Norwegian Sea. Radiative amplifiers weaken and relocate; the subpolar region can begin absorbing more heat from the atmosphere.
These regimes map onto historical-to-future trajectories in models (e.g., CESM2 transitions through them from the late 20th century into the 21st). The year of peak AMOC- SST sensitivity can help predict the shift into the weak regime and associated decay of the classic warming hole. Atmospheric radiative processes dominate the regime shifts; oceanic processes contribute more indirectly via air– sea coupling.
Implications
- The cold blob is a useful but imperfect and non- stationary indicator of AMOC strength. Its presence or intensity does not map linearly to AMOC decline, and its future disappearance would not necessarily signal AMOC recovery.
- Models project continued AMOC weakening under greenhouse- gas forcing, with the system potentially moving from the current stronger and intermediate regimes toward the weak regime by 2100. The timing of major shifts in the blob carries uncertainty (model-dependent estimates range from the early 2020s to the 2060s).
- Broader climate effects of a weaker AMOC (and associated blob) include potential influences on the jet stream, storm tracks, European and North American weather extremes, and regional sea- level patterns. A full AMOC collapse remains a low- probability but high- impact tipping- point risk this century under high- emissions scenarios; most assessments project substantial weakening rather than abrupt shutdown.
- Improved monitoring (satellites, ocean arrays, sustained heat- content observations) and model constraints that account for regime dependence will reduce projection uncertainties.
In short, the cold blob is tightly linked to AMOC weakening via reduced ocean heat transport, but the strength of that coupling, and the resulting SST pattern- depends strongly on the background AMOC state and is mediated importantly by atmospheric radiation and clouds. This adds important nuance to interpreting both past observations and future projections.
Regime shifts of AMOC- sea surface temperature relationship
The Atlantic Meridional Overturning Circulation (AMOC) regulates climate, and subpolar North Atlantic sea- surface temperature (SST) patterns are commonly used as proxies for its strength.
This study shows the AMOC- SST relationship is state- dependent, not stationary.
Using Community Earth System Model (CESM2) simulations across diverse climate states plus a multi- model ensemble, the authors identify three distinct regimes:
- Strong AMOC: typical dipole SST fingerprint.
- Intermediate AMOC: amplified subpolar SST anomalies.
- Weak AMOC: muted North Atlantic signals.
These regimes arise primarily from changes in atmospheric radiative processes (ocean processes contribute more indirectly via air–sea interactions).
The year of peak AMOC- SST sensitivity can predict the transition into the weak- AMOC regime and the associated decay of the North Atlantic warming hole (“cold blob”).
This provides a physically based constraint on model uncertainties in climate projections and implies that SST- based AMOC indicators must account for state dependence.
AMOC is indexed as maximum overturning transport below 500 m at 40°N. The warming- hole SST index (WHSST) is the difference between subpolar North Atlantic mean SST (roughly 45°- 60°N, 20°- 50°W) and the global mean.
Objective change-point detection on CESM2 simulations (paleoclimate to future scenarios) yields two change points (~20.0 Sv and ~11.5 Sv), defining three regimes:
| Regime | AMOC strength | WHSST sensitivity to AMOC | Spatial SST pattern with weakening AMOC |
|---|---|---|---|
| 1 (strong) | > ~20 Sv | Low (~0.09 °C Sv⁻¹) | Classic dipole: cooling in central subpolar basin and warming along Gulf Stream extension |
| 2 (intermediate) | ~11.5- 20 Sv | High (~0.31 °C Sv⁻¹; >3 times Regime 1) | Intensified cooling across the entire subpolar basin (strongest in Irminger and Labrador Seas) |
| 3 (weak) | < ~11.5 Sv | Low again (~0.12 °C Sv⁻¹) | Cooling shifts northeast into the Norwegian Sea and Iceland basin; canonical warming hole decays |
The pattern is robust across alternative SST and AMOC metrics and appears in at least four additional CMIP6 models.
These regimes correspond to stages of the North Atlantic warming hole: emerging (Regime 1), maturing (Regime 2), and decaying (Regime 3).
Historical and 21st- century CESM2 trajectories cross these regimes between ~1985 and 2100.
A partial temperature-change decomposition attributes WHSST anomalies mainly to:
- Oceanic processes (ocean heat transport convergence and local heat storage, largely balanced by turbulent heat fluxes).
- Surface downward clear- sky radiation (dominated by longwave).
- Cloud radiative effects.
- Regime 1: All three process groups contribute roughly equally (modest net cooling).
- Regime 2: Sensitivity tripling is radiatively driven. Clear- sky longwave (linked to tropospheric cooling and reduced water vapor) supplies ~2/3 of the increase; enhanced shortwave reflection by low- level clouds supply most of the rest. Oceanic terms stay nearly neutral.
- Regime 3: Radiative amplifiers (clear- sky longwave and cloud shortwave) weaken and shift northeast, so the canonical subpolar warming hole fades even as AMOC continues to decline. Oceanic contributions remain secondary.
Ocean heat transport changes scale with AMOC but are largely offset by compensating turbulent fluxes, so they do not drive the regime shifts directly.
Implications
- SST fingerprints of AMOC are non- stationary; interpreting the observed cold blob (or its future evolution) as a simple AMOC proxy requires regime awareness.
- The timing of peak AMOC- SST sensitivity offers a useful predictor for when the system enters the weak- AMOC regime, and the warming hole begins to decay.
- This framework helps explain and constrain inter- model spread in 21st-century North Atlantic SST projections.
The paper builds directly on the authors’ prior work linking atmospheric radiation to the cold blob and aligns with the Penn State news summary discussed previously.
Full text, figures, methods, and supplementary material are available at the Nature Communications page and PDF.
Journal: Nature Communications, volume 17, Article number: 9416 (2026)
DOI: 10.1038/s41467-026-76149-4, Open access
Published: 05 August 2026
Provided: Pennsylvania State University
Authors: Yifei Fan,
Duo Chan,
Gokhan Danabasoglu,
Who M. Kim,
Pengfei Zhang&
Laifang Li (corresponding), among others
Abstract
The Atlantic Meridional Overturning Circulation (AMOC) plays critical roles in regulating climate, and subpolar North Atlantic sea-surface temperature (SST) patterns are widely used to infer changes in its strength. Yet the stationarity of their relationship remains unclear. Here, using Community Earth System Model simulations spanning various climate states and a multi-model ensemble, we show that this relationship is state-dependent. We identify three distinct regimes: strong AMOC with the typical dipole fingerprint; intermediate AMOC with amplified subpolar SST anomalies; and weak AMOC with muted North Atlantic signals. These regimes arise primarily from changes in atmospheric radiative processes, while ocean processes contribute indirectly through air–sea interactions. The year of peak AMOC–SST sensitivity emerges as a predictor of the transition into the weak-AMOC regime and associated decay of the North Atlantic warming hole, offering a physically-based constraint on model uncertainties in climate projections. Our findings also imply that SST-based AMOC indicators must account for state dependence.
You must be logged in to post a comment.