A new study published in Science (DOI: 10.1126/science.aed4270) on November 5, 2026, finds that abrupt climate shifts during the last ice age, specifically Dansgaard- Oeschger (DO) events linked to the Atlantic Meridional Overturning Circulation (AMOC), took substantially longer than previously inferred from Greenland ice- core records.
Researchers from Xi’an Jiaotong University’s Institute of Global Environmental Change (led by first author Xiyu Dong, with contributions from Hai Cheng, Xu Zhang, and others) analyzed high- resolution oxygen isotope (δ¹⁸O) records from annually laminated speleothems.
These came from Cherrapunji Cave (northeast India) and Shuixi Cave (southwest China), combined with published data from Zhangjia Cave. The records cover DO events 1 through 8.
Key findings:
- Transitions in the Asian monsoon speleothem records consistently lasted longer than 100 years and averaged about 180 years.
- This is far slower than the decadal- scale shifts seen in Greenland ice -core δ¹⁸O records, which had been used as the main proxy for the speed of AMOC strengthening.
The annual laminae in the speleothems provided precise relative timing of transition durations (complementing absolute U– Th ages).
Modeling indicates a two- stage process: rapid retreat of sea ice in the northeastern North Atlantic quickly adjusted regional atmospheric circulation, producing the sharp Greenland δ¹⁸O rise.
Full AMOC strengthening and further northwestern North Atlantic sea- ice retreat followed more gradually, without major additional changes in atmospheric circulation modes or moisture pathways that would further alter Greenland δ¹⁸O.
Asian monsoon speleothem records capture the complete AMOC strengthening process.
Implications: Natural AMOC transitions were not as rapid as ice- core data alone suggested. When discussing potential future AMOC tipping risks, the duration of any transition is as important as the climatic impacts.
The study highlights the value of combining high- quality, annually resolved paleoclimate records with process- based (isotope- enabled) Earth system models.
The authors note that additional high- quality speleothem records of DO events, better North Atlantic sea- ice reconstructions, and models that incorporate ice- sheet processes will further improve understanding.
This revises the view of how quickly these past high- latitude Northern Hemisphere warmings (5- 16 °C) and associated AMOC changes unfolded, offering a more measured perspective for interpreting both past and potential future abrupt climate shifts.
Abrupt climate change refers to large- scale shifts in the climate system that occur much faster than the gradual external forcing (such as orbital changes or rising greenhouse gases) that triggers them.
These transitions happen on timescales of years to a few decades, persist for decades or longer, and can cause major disruptions to ecosystems and human societies.
Two complementary definitions are commonly used:
- Physical: The climate system crosses a threshold or tipping point and switches to a new state at a rate set by internal dynamics (e.g., feedbacks involving ocean circulation, sea ice, or ice sheets), faster than the forcing itself.
- Impacts- based: A change so rapid and unexpected that natural or human systems struggle to adapt.
The best- documented cases come from the last glacial period (~120,000–11,700 years ago):
- Dansgaard- Oeschger (DO) events: Repeated abrupt warmings of 5- 16 °C over Greenland (and the North Atlantic region), often within decades according to ice- core records, followed by gradual cooling. These occurred roughly every 1,000- 1,500 years. They are strongly linked to strengthening of the Atlantic Meridional Overturning Circulation (AMOC), which transports heat northward.
- Heinrich events: Massive iceberg discharges into the North Atlantic that disrupted ocean circulation, often associated with the coldest phases of DO cycles.
- Younger Dryas (~12,900- 11,700 years ago): A sudden return to near- glacial conditions in the Northern Hemisphere during the last deglaciation, ending with rapid warming.
Other examples include the end of the Carboniferous rainforest collapse and possibly the Paleocene- Eocene Thermal Maximum.
Abrupt changes typically involve nonlinear feedbacks and thresholds:
- Changes in AMOC strength (weakening or collapse reduces heat transport to the North Atlantic).
- Sea -ice and ice- sheet dynamics (ice- albedo feedback amplifies warming or cooling).
- Freshwater input from melting ice that stratifies the ocean and suppresses deep- water formation.
- Atmospheric reorganization (shifts in monsoons, storm tracks, and the Intertropical Convergence Zone).
A recent Science study (November 2026) used annually laminated speleothems from Asia to show that the full AMOC strengthening during DO events 1- 8 actually took longer than previously thought, consistently over 100 years and averaging ~180 years, rather than the decadal scale suggested by Greenland ice- core δ¹⁸O alone.
Modeling indicates a two-stage process: rapid northeastern North Atlantic sea- ice retreat drives the sharp Greenland signal, while the complete AMOC recovery and further sea- ice loss occur more gradually. Asian monsoon records capture the fuller transition.
This revises the view of how quickly these past “abrupt” ocean- circulation shifts unfolded and underscores that transition duration is as important as the magnitude of change when assessing future risks.
Modern climate models and observations raise concerns about possible abrupt shifts under anthropogenic warming, including:
- Weakening or collapse of the AMOC.
- Rapid ice- sheet mass loss (Greenland, West Antarctica) and associated sea-level rise.
- Permafrost thaw releasing additional greenhouse gases.
- Amazon rainforest dieback or other ecosystem tipping points.
While the current warming is primarily gradual and forced by greenhouse gases, the climate system can still exhibit nonlinear, abrupt responses once thresholds are crossed.
Paleoclimate records remain the primary evidence for such behavior, informing risk assessments for the coming decades to centuries.
In short, abrupt climate change is a well- established feature of Earth’s past climate system, driven by internal thresholds and feedback, and remains a key concern for understanding both historical variability and potential future surprises.
Rapid Natural Warming Spikes Occurred 25 Times — Last Ice Age Edition – Climate- Science.press
The linked Science paper is titled “Glacial abrupt transitions in ocean circulation were much slower than ice-core records imply” (DOI: 10.1126/science.aed4270), published 5 November 2026.
It is a peer- reviewed research article from researchers including Xiyu Dong (first author), Xu Zhang, Hai Cheng, and co-authors from Xi’an Jiaotong University and collaborating institutions. The full text is behind Science’s paywall and subscriber access, but detailed summaries, the dataset, and press materials are publicly available.
Dansgaard– Oeschger (DO) events involved rapid high- latitude Northern Hemisphere warmings of 5- 16 °C, tightly linked to strengthening of the Atlantic Meridional Overturning Circulation (AMOC).
Prior estimates of AMOC transition speeds (often within decades) relied heavily on Greenland ice-core δ¹⁸O records, which reflect both temperature and moisture- source changes.
Using high- resolution δ¹⁸O records from annually laminated speleothems (cave deposits) that more directly track AMOC- driven changes in the Asian monsoon:
- Two new records from Cherrapunji Cave (northeast India) and Shuixi Cave (southwest China), plus published data from Zhangjia Cave.
- Coverage of DO events 1- 8.
Results: The transitions consistently lasted >100 years and averaged ~180 years, substantially longer than the decadal- scale shifts in Greenland ice- core δ¹⁸O.
Annual laminae enabled precise relative dating of transition durations (complementing absolute U– Th ages).
Modeling supports a two- stage process:
- Rapid sea- ice retreat in the northeastern North Atlantic quickly adjusted regional atmospheric circulation, producing the sharp Greenland δ¹⁸O rise.
- Further AMOC strengthening and northwestern North Atlantic sea- ice retreat occurred more gradually, without major additional shifts in atmospheric circulation or moisture pathways that would further change Greenland δ¹⁸O.
Asian monsoon speleothem δ¹⁸O records capture the full AMOC strengthening process.
Natural AMOC transitions were slower than previously inferred from ice cores alone. The authors note that discussions of future AMOC tipping risks should emphasize transition duration alongside potential impacts.
Supporting data (U- Th dates, δ¹⁸O time series, annual lamina thicknesses, anchored ice- core records, stacked series, and lamina- counting code) are archived on Dryad (doi:10.5061/dryad.xgxd254wx).
The study underscores the value of combining annually resolved speleothem records with isotope- enabled Earth- system modeling for quantifying rates of past abrupt climate change.
Journal: Science, Vol 394, Issue 6824
DOI: 10.1126/science.aed4270.
Provided: Xi’an Jiaotong University
Authors: Xiyu Dong, Xu Zhang, Yuchen Sun, Gayatri Kathayat, Youwei Li, Sune O. Rasmussen, Anders Svensson, Christo Buizert, Ashish Sinha , Stephen Barker, Carlos Pérez-Mejías, Gregor Knorr, Rishui Chen , Kan Zhao, Wenjing Du, Haiwei Zhang, Hanying Li, Andrea Columbu, Liangcheng Tan , Yuao Zhang, Youfeng Ning, Martin Werner, Yongjin Wang, R. Lawrence Edwards and Hai Cheng
Abstract
Dansgaard-Oeschger (DO) events during the last ice age are thought to reflect rapid shifts in the Atlantic Meridional Overturning Circulation (AMOC). Greenland ice cores suggest warming transitions within only a few decades—or even just over 10 years. Using precisely dated speleothem oxygen-isotope records from the Asian monsoon region, we have shown that the same DO warmings unfolded over more than a century. Isotope modeling revealed that Greenland records primarily capture a rapid atmospheric reorganization triggered by the initial sea-ice retreat in the northeastern North Atlantic, whereas the parallel AMOC recovery and hemispheric thermal adjustment proceeded much more slowly. These findings reveal that natural AMOC transitions evolved more slowly than the rapid North Greenland oxygen-isotope shifts would imply, thereby refining our understanding of “abruptness” during DO events.
Potential Catastrophe: The Impact of AMOC Collapse on Climate
North Atlantic Cold Blob Reveals AMOC Weakening Is Complex
You must be logged in to post a comment.