The Tipping Points That Get the Least Attention
A PNAS analysis of 20,736 studies finds climate research clustered on Greenland and the AMOC, while boreal permafrost thaw and the subpolar gyre draw less attention despite lower warming thresholds.
An Uneven Map of Concern
Climate tipping points began as a thought experiment. A decade or two later they sit inside risk assessments, adaptation planning and arguments about how fast the world has to move. That shift makes an awkward question unavoidable: is the science looking where the risk actually is?
A study in the Proceedings of the National Academy of Sciences set out to measure that directly, comparing the volume of research attention given to different climate tipping elements against the risk-relevant characteristics of each one (paper).
Reading 20,736 Records
Heming Wang, Bingyu Chen, Fengmei Ma, Nico Wunderling, Timothy M. Lenton and Asaf Tzachor assembled 20,736 records specific to climate tipping elements, published between 2000 and 2025 and covering 14 tipping elements. The paper appeared online on 28 September 2026. The result is a map of where a quarter-century of scientific effort has pooled.
Two names dominate that map: the Greenland Ice Sheet and the Atlantic Meridional Overturning Circulation.
The Quiet Elements
At the other end sit boreal permafrost — specifically abrupt thaw — and the North Atlantic Subpolar Gyre. Both received comparatively limited attention, despite relatively low estimated warming thresholds.
That comparison is the point. A lower threshold means an element is estimated to be in play at less warming, not more. The elements attracting less study are not obviously the safer ones.
A Missing Vocabulary
The authors also searched for explicit engagement with the dynamics that make tipping points worrying in the first place: abrupt transitions, hysteresis and irreversibility. That engagement was limited across the literature.
Hysteresis is what makes a tipping element different from a dial. The trip back is not the reverse of the trip out: a system can stay locked in a changed state after the driver returns to its old level. A literature that describes a risk without that language is describing a different object.
Why Attention Is Not Neutral
Research attention is not a scoreboard of importance. It shapes what can be done. Monitoring networks, model development, impact studies and policy advice all follow the literature that already exists. An element with a thin record is harder to assess, harder to prioritise and harder to govern.
The authors frame that as a possibility rather than a proven causal chain. Their conclusion is carefully worded: these patterns "may influence which tipping risks can be most readily assessed, prioritized, and governed."
What This Does Not Show
The study measures alignment between publication attention and risk-relevant characteristics. It is not a ranking of elements by danger, and record counts measure how much has been written, not the quality of what was written. Estimated thresholds carry wide uncertainty.
Nor does the finding say Greenland and the AMOC are overstudied. Both are enormous risks, and both deserve the attention they get. The question is what the concentration leaves comparatively unexamined.
The Practical Question
If attention keeps concentrating where the literature already is, the elements that are hardest to characterise may stay that way — and those are the ones whose thresholds matter most for how quickly the risk picture changes.
The paper's contribution is smaller and more useful than a warning: it makes the imbalance measurable, and therefore arguable, at the moment when tipping-point risk is moving from a research question into a budget line.