The research, led by Ingo Richter of the Japan Agency for Marine-Earth Science and Technology together with colleagues from institutions in Japan, the United States, Norway, and Brazil, rests on an unusually broad foundation of evidence. The team compared simulations of a pre-industrial climate with projections of a high-emissions future, known as scenario ssp585, drawn from 23 state-of-the-art climate models participating in the Coupled Model Intercomparison Project Phase 6, the same modeling framework that underpins the assessments of the Intergovernmental Panel on Climate Change. By contrasting these two worlds, one without human interference and one heated by sustained emissions, the researchers could isolate how the machinery of Atlantic variability itself changes when the planet warms.

The headline finding confirms a trend that earlier studies had already hinted at: the variability of sea-surface temperatures along the equatorial Atlantic, the signature of the Atlantic Zonal Mode, weakens under global warming. In plain terms, the Atlantic Niño becomes less intense, swinging less dramatically between warm and cold phases. Off the equator, by contrast, the picture is different, with sea-surface temperature variability tending to increase slightly in the flanking regions of the tropical Atlantic. That split result matters, because the equatorial mode is the one most tightly linked to the seasonal rains that farmers and water managers from Senegal to Angola depend upon, and its weakening has been flagged in previous work as a potential source of forecast uncertainty.

What makes the new analysis distinctive is its dissection of why the weakening occurs. The Atlantic Niño, like its Pacific counterpart, is sustained by the Bjerknes feedback, a self-reinforcing loop in which a warm anomaly in the eastern equatorial ocean weakens the prevailing easterly trade winds, which in turn suppresses the upwelling of cold subsurface water, further warming the surface. The study finds that this feedback is losing its grip. Contrary to some earlier projections, the team found that the mean-state equatorial Atlantic thermocline, the sharp vertical boundary between warm surface water and cold deep water, actually shoals slightly under radiative forcing, a change that should, on its own, strengthen the mode by making the surface more sensitive to subsurface dynamics. The decline of the Atlantic Niño therefore cannot be blamed on a deepening thermocline. Instead, the culprits are a weakening of the mean upwelling that supplies cold water to the surface and a reduced sensitivity of surface winds to sea-surface temperature anomalies, both of which blunt the Bjerknes feedback at its most critical links.

The second major discovery concerns the changing character of the mode’s forcing. In the pre-industrial simulations, the Atlantic Zonal Mode is driven primarily by dynamic forcing, meaning anomalies in surface wind stress that stir the ocean and rearrange its heat content. In the high-emissions future, the composites of Atlantic Niño events show that this dynamic pathway diminishes, while thermodynamic forcing takes on a more prominent role. Thermodynamic forcing operates through the exchange of heat at the ocean surface, particularly through latent heat flux, the energy carried away by evaporation, and through shortwave radiation, the sunlight that warms the upper ocean. In a warmer world, the Atlantic Niño becomes less a story of winds pushing water around and more a story of clouds, evaporation, and radiant heat reshaping the sea surface.

That shift in mechanism is tied to a striking change in the Atlantic’s relationship with the Pacific. The analysis reveals that the influence of the El Niño-Southern Oscillation on the Atlantic Zonal Mode strengthens under global warming, to the point that a positive correlation emerges between the two phenomena, with Pacific events preceding their Atlantic counterparts by roughly half a year. In today’s climate, the connection between El Niño and the Atlantic Niño has long been described as inconsistent and fragile, appearing in some decades and vanishing in others, a puzzle that has occupied tropical climate scientists for years. The new projections suggest that the greenhouse-warmed atmosphere will knit the two basins together more tightly, transmitting Pacific signals across Central America and the tropical atmosphere with greater reliability.

Two processes appear to explain this tightening bond. First, El Niño itself grows stronger in the high-emissions simulations, and a more powerful Pacific oscillator naturally broadcasts a louder signal into neighboring basins. Second, and perhaps more intriguingly, the weakening of the coupled air-sea feedbacks within the equatorial Atlantic leaves that ocean more susceptible to external forcing. A system whose internal feedbacks have gone quiet is one that listens more attentively to remote voices. When the Atlantic’s own Bjerknes feedback can no longer dominate its variability, the Pacific’s influence, arriving through atmospheric bridges such as shifts in the Walker circulation and changes in tropical tropospheric temperature, finds less resistance and leaves a clearer imprint on Atlantic sea-surface temperatures.

Paradoxically, this foreign domination may carry a silver lining for forecasters. A simple linear analysis performed by the team indicates that, in some of the models, the association of the Atlantic Zonal Mode with El Niño makes the Atlantic mode more predictable, even as its amplitude fades. The logic is straightforward: if a large fraction of Atlantic variability can be traced back to a Pacific precursor that emerges about six months in advance, then forecast systems that skillfully predict El Niño gain, for free, a measure of skill in predicting the Atlantic Niño. Seasonal prediction centers in Africa, Europe, and the Americas could eventually exploit this teleconnection to extend the useful lead time of rainfall outlooks for the Sahel, the Guinea Coast, and northeastern Brazil, regions where the Atlantic mode’s influence on the West African monsoon and coastal precipitation is well documented.

The study also underscores how much remains uncertain. The 23 models do not speak with one voice; the emergence of the ENSO-Atlantic correlation is clear in the ensemble but varies in strength from model to model, and tropical Atlantic simulations remain haunted by persistent mean-state biases that have plagued coupled models for decades. The researchers cross-checked their projections against observational and reanalysis products, including the ERA5 atmospheric reanalysis, the ORAS5 ocean reanalysis, and the HadISST sea-surface temperature dataset, all of which are publicly available, but the fundamental limits of simulating a basin as small and as seasonally locked as the equatorial Atlantic still apply. Whether the real ocean will follow the models’ script is a question that only the coming decades of observation can answer.

Still, the implications are hard to ignore. A tropical Atlantic that dances increasingly to the Pacific’s tune would alter the risk landscape for coastal fisheries, which are disrupted when the cold tongue fails to deliver its usual nutrient-rich upwelling, and for agricultural planners who rely on the statistical rhythms of Atlantic variability. It would also complicate attribution studies, since an Atlantic event that once looked like a local fluke may in the future be the distant echo of an El Niño half a world away. The research, published in Climate Dynamics as volume 64, article 450, offers both a warning and a tool: the Atlantic Niño of the future may be weaker and less self-reliant, but by borrowing predictability from the Pacific, it may also become a signal that humanity can see coming with more confidence than ever before.

Subject of Research: Projected changes in tropical Atlantic sea-surface temperature variability and its coupling to ENSO under global warming

Article Title: Strengthened tropical Pacific influence on tropical Atlantic variability in global warming projections

Article References: Richter, I., Chang, P., Kataoka, T., Kido, S., Keenlyside, N., Kosaka, Y., Okumura, Y., Tokinaga, H., Tozuka, T., & Vilela, I. (2026). Strengthened tropical Pacific influence on tropical Atlantic variability in global warming projections. Climate Dynamics, 64(11), Article 450. https://doi.org/10.1007/s00382-026-08390-y

Image Credits: AI Generated

DOI: 10.1007/s00382-026-08390-y

Keywords: Atlantic Niño, Atlantic Zonal Mode, ENSO, CMIP6, Bjerknes feedback, tropical Atlantic, sea-surface temperature, climate change, thermocline, seasonal prediction, Walker circulation, Climate Dynamics