Study: Massive Abrupt Warming During Past Low-CO₂ Era

When policymakers are told:

• Rapid warming is unprecedented.
• Climate instability is uniquely modern.
• Current rates are outside natural bounds.

They are not being given the full paleoclimate context. This paper has it.

A newly published paper in Climate of the Past provides something climate science could use more of: large-scale, data-driven reconstructions grounded in proxy evidence rather than projections. The study, “A global analysis of pollen-based reconstructions of land climate changes during Dansgaard–Oeschger events” by Liu, Prentice, and Harrison (2026), examines terrestrial pollen records spanning 50,000 to 30,000 years ago, a period covering Marine Isotope Stage 3 (MIS 3), when Dansgaard–Oeschger (D–O) events were most frequent.

The authors state plainly: “Although D–O events are found throughout the last glacial period, the largest number and the most regular patterning occurred during Marine Isotope Stage 3 (MIS 3; 57 to 29 ka) when there were 11 separate events (D–O 15 to D–O 5)” . That’s eleven abrupt warming events between roughly 57,000 and 29,000 years ago.

And here is the key point that should give pause to anyone who believes that rapid warming necessarily requires rising atmospheric CO₂: “The surface air temperature in Greenland increased by 10–16 °C during the warming phases; these warming events occurred over an interval of between 50 and 200 years” . Ten to sixteen degrees Celsius, in as little as 50 years, and not just once, but repeatedly. During MIS 3, atmospheric CO₂ hovered around roughly 190–210 ppm. The study notes that past CO₂ values are taken from Bereiter et al. (2015) , and their modeling experiments were initialized at 207.5 ppm CO₂ . There was no industrial-era surge, no 100-ppm jump, no massive upward trend comparable to the modern increase. CO₂ was low by modern standards and relatively stable compared to today’s trajectory.

Yet Greenland experienced repeated abrupt warmings of up to 16°C.

If CO₂ were the sole or dominant control knob for abrupt warming magnitude and rate, these events should not appear in the paleoclimate record. But they do.

A common response is that Greenland is unique, and of course it is. But these events were not isolated Arctic curiosities.

As the authors make clear: “The D–O signals are not just seen in Greenland – they are registered globally…” . The paper synthesizes pollen data across Eurasia, North America, the tropics, and the Southern Hemisphere. From the abstract: “These reconstructions show that the largest warming occurred in northern extratropics, especially Eurasia, while western North America and the southern extratropics were characterised by cooling” .

They further report: “The change in winter temperature was significantly larger than the change in summer temperature in the northern extratropics and the tropics…” . So we are looking at large-scale reorganizations of hemispheric climate, not a localized anomaly. The authors also note antiphasing between hemispheres: “Both oceanic and ice-core records indicate that temperature changes are out-of-phase between the northern and southern hemispheres…” . In other words, rapid, large-amplitude, hemispherically structured climate shifts occurred repeatedly, without any corresponding CO₂ surge.

Interestingly, the authors themselves observe that “Dansgaard–Oeschger (D–O) warming events are comparable in magnitude and rate to the anticipated 21st century warming” . They add the appropriate caveat that “both the baseline state and the mechanism inducing this warming differ from anticipated 21st century climate changes” . Fair enough. The glacial world had larger ice sheets and different boundary conditions. But what cannot be ignored is that the climate system has demonstrated the capacity for extremely rapid, double-digit warming without elevated CO₂. That is not speculation; it is proxy-based evidence.

The modeling framework discussed in the paper reinforces this point. The LOVECLIM simulations described were initialized at 207.5 ppm CO₂ and driven by orbital forcing, ice-sheet configuration, and meltwater pulses in the North Atlantic . The mechanism behind D–O events is widely associated with changes in ocean circulation, particularly the Atlantic Meridional Overturning Circulation, along with associated feedbacks. Not CO₂ spikes. The climate system clearly contains powerful internal nonlinear dynamics capable of producing abrupt warming independent of rising atmospheric carbon dioxide.

We are frequently told that modern warming is unprecedented, unmatched in rate, and outside natural variability. Yet here we have eleven events in a roughly 28,000-year window showing 10–16°C warming in 50–200 years, repeatedly, under low and relatively stable atmospheric CO₂ levels.

Of course, Greenland amplification is larger than global mean change, and modern global warming of about 1–1.3°C since the 19th century is a global average. But the broader claim often made concerns rate and systemic instability. D–O events demonstrate that the climate system can warm extremely rapidly, that large reorganizations can occur naturally, and that global-scale signals can arise without rising CO₂. That does not mean CO₂ has no radiative effect; it does. But rapid warming, by itself, is not diagnostic of CO₂ forcing. That distinction is often blurred in public discourse.

The seasonal pattern in the reconstructions also matters. The study finds that winter temperature changes were significantly larger than summer changes in many regions , a hallmark of circulation-driven shifts rather than simple uniform radiative forcing. Meanwhile, moisture responses were mixed and not globally coherent. As the authors note, there is no globally consistent relationship between changes in CO₂-corrected plant-available moisture and summer temperature . These were complex, dynamic reorganizations of the climate system, not monotonic, globally uniform responses to a single forcing agent.

The structure of D–O cycles is also revealing. They are described as involving “an abrupt large warming in matter of decades, followed by a long slow cooling over centuries to millennia, with a terminal phase of fast cooling” . That oscillatory behavior strongly suggests threshold dynamics and internal variability. The Earth system is not a simple linear thermostat controlled by one variable. It is a nonlinear, multi-component system involving ocean circulation, sea ice, ice sheets, atmospheric patterns, moisture transport, and vegetation feedbacks. The new pollen synthesis underscores that complexity.

To be clear, modern warming occurs under different boundary conditions, and today’s climate is not a glacial state. But none of those caveats erase the empirical fact that abrupt, large-magnitude warming events have occurred repeatedly in Earth’s history under low and relatively stable CO₂ levels.

That alone should encourage caution in equating rapid warming exclusively with rising atmospheric carbon dioxide. Policymakers are often told that current warming is entirely unprecedented in rate and character. The MIS 3 record shows multiple abrupt warming events, double-digit temperature shifts, global teleconnections, and no corresponding CO₂ surge. That does not invalidate concerns about anthropogenic influence, but it does argue for humility in claims of uniqueness.

The Liu et al. (2026) paper confirms eleven abrupt warming events between 57 and 29 ka (57,000 to 29,000 years ag), Greenland warming of 10–16°C in 50–200 years , global-scale expressions of these events , and CO₂ levels near ~200 ppm during this interval .

These are peer-reviewed reconstructions based on terrestrial pollen data, not speculative modeling exercises. The broader takeaway is straightforward: the climate system has always been capable of dramatic, rapid changes. CO₂ is one influence among many, but paleoclimate evidence makes it clear that abrupt warming is not uniquely or exclusively tied to rising atmospheric carbon dioxide. That is not a slogan. It is what the data show.

Final Thoughts

The Liu et al. (2026) paper represents serious, quantitative proxy work.

It confirms:

• 11 abrupt warming events between 57–29 ka (57,000 to 29,000 years ago)
• Greenland warming events of 10–16°C in a 50–200 years span
• Global-scale expressions of these events
• CO₂ levels around ~200 ppm during this period

These are not blog claims. They are peer-reviewed reconstructions.

The broader takeaway?

The climate system has always been capable of dramatic, rapid changes. CO₂ is one influence among many, but paleoclimate evidence shows that abrupt warming is not exclusively tied to rising atmospheric carbon dioxide.

That is not denial. That is data.

And in a field that too often drifts toward narrative certainty, data still matter.