• Climate

2°C is not known to be a "point of no return", as Jonathan Franzen claims

Posted on:  2019-09-17

Key takeaway

This statement makes a claim about the climate system that does not reflect our best understanding of how it works. The impacts and risks of climate change increase as warming continues, and limiting warming to the 2°C milestone has been chosen as an international goal, but it is not true that scientists expect runaway warming to begin if this milestone is crossed.

Reviewed content


The consensus among scientists and policy-makers is that we’ll pass this point of no return if the global mean temperature rises by more than two degrees Celsius

Source: The New Yorker, Jonathan Franzen, 2019-09-08

Verdict detail

Misleading: While positive feedbacks exist that amplify temperature changes, scientists have not identified a "point of no return" at 2°C.

Full Claim

Our atmosphere and oceans can absorb only so much heat before climate change, intensified by various feedback loops, spins completely out of control. The consensus among scientists and policy-makers is that we’ll pass this point of no return if the global mean temperature rises by more than two degrees Celsius (maybe a little more, but also maybe a little less)[...] In the long run, it probably makes no difference how badly we overshoot two degrees; once the point of no return is passed, the world will become self-transforming.

UPDATE (25 August 2020): This section of the New Yorker article was edited at an unknown date, and now reads: “Some scientists and policymakers fear that we’re in danger of passing this point of no return if the global mean temperature rises by more than two degrees Celsius (maybe more, but also maybe less).”
A correction note at the bottom of the article states: “A previous version of this article mischaracterized the scientific consensus around a ‘point of no return.'”


Amber Kerr member picture

Amber Kerr

Researcher, Agricultural Sustainability Institute, University of California, Davis

This is not correct. First of all, there is no consensus on what level of warming would be necessary to set off a runaway, exponential greenhouse gas buildup (presumably due to global permafrost melting, methane clathrates disintegrating, and/or continental-scale forest fires). But to the extent that there is a consensus, 2°C is not it. Even a relatively risk-averse assessment of this question, which does recommend trying to limit warming to under 2°C, states that most large-scale destabilizing feedbacks don’t kick in until at least 3°C, and others not until over 5°C1.

Second of all, Franzen seems not to realize that climate models already do include feedback loops as a fundamental aspect of climate dynamics. These include water vapor feedback (the most important and least uncertain), ice-albedo feedback, saturation of terrestrial and oceanic carbon sinks, and acceleration of decomposition. The way Franzen discusses feedback loops in his article—as though he is introducing a new insight by including feedback loops as a multiplier on top of climate model output—suggests that he does not understand the details of how climate models work.

The second portion of the statement is not only wrong but dangerously misleading. Even apart from the increasing risks of threshold-crossing (per above), it makes a profound difference how badly we overshoot two degrees.

Consider the difference between RCP4.5 (an ambitious but realistically achievable scenario in which annual emissions peak in 2040 and go to net zero in 2080) versus RCP8.5 (a do-nothing scenario in which annual emissions are still increasing by 2100). Under RCP4.5, the temperature increase by 2090 is about 2°C, possibly as high as 2.5°C. Under RCP8.5, temperature increase by 2090 is about 4°C, possibly as high as 5°C.

All of the following things are credibly predicted to happen under RCP8.5, whereas they would be avoided by RCP4.5:

– Large swaths of India may become literally uninhabitable by humans, with sustained wet-bulb temperatures over 35°C2.

– Nearly all coral reefs on Earth, including all 29 reefs that are UNESCO World Heritage listed, are likely to experience severe bleaching events annually, leaving them unable to recover3.

– Global declines in staple crop production (maize, wheat, rice, soy), not accounting for CO2 fertilization, could be up to 18%, compared to 9% under RCP4.54.

These are only three illustrative examples; there are many more.

In conclusion, Franzen’s essay shows a lack of understanding of how climate models work. He says that “As a non-scientist, I do my own kind of modelling,” but he seems to be unaware that scientists have already carried out many qualitative and quantitative climate risk assessments, using policy changes and human behavior as variables. His claim that additional warming over 2°C doesn’t matter is scientifically unsound, and his fatalism about human society—though not something that I can assess scientifically—is not a belief that I share.

Charles Koven member picture

Charles Koven

Staff Scientist, Lawrence Berkeley National Lab

According to our best climate models, which include all of the feedback loops that we can figure out how to put into them, there are not any thresholds beyond which warming will, on its own, spin out of control. It is true that, if the planet reaches higher amounts of warming, then we expect current carbon sinks to weaken and new processes like permafrost thaw to emit greenhouse gases. But at the same time, we also expect that the amount of warming from each additional increment of CO2 that ends up in the atmosphere will weaken as its concentration increases.

The surprising result when we couple all of these processes together in models is that these two sets of effects tend to cancel each other out, so that the total amount of warming is roughly proportional to the cumulative amount of CO2 we have emitted. Every bit of carbon that we emit is an extra bit of warming that the planet will experience.

The levels of warming that have been set as targets in climate negotiations—like 1.5 or 2 degrees C—are not thresholds beyond which the world will end; they are points where we can try to estimate the impacts and then set as goals that will allow us to avoid some of the worst effects of climate change. If we exceed these goals, then we will certainly experience greater impacts, and there could be surprises that we don’t understand, but there is no reason to expect that warming will become self-sustaining if we fail to keep temperatures below these levels.

Patrick Brown member picture

Patrick Brown

Assistant Professor, San Jose State University

There is certainly no consensus among scientists that 2.0°C (3.6°F) global warming above pre-industrial levels represents a “point of no return” for climate change.

The dominant feedback in the climate system is the stabilizing feedback known as the Planck Response1 which makes self-perpetuating, run-away warming exceedingly unlikely (at least within the range of temperatures considered under anthropogenic climate change scenarios). There is some speculation that long-term Earth-system feedbacks may become active near 2°C warming2 which would enhance warming to well beyond 2°C in the long run, even without additional anthropogenic greenhouse gas emissions. Nevertheless, we expect the global temperature to depend mostly on cumulative anthropogenic greenhouse gas emissions over the next century and we do not expect to suddenly lose control of the global temperature if the 2°C limit is passed.

So what does the 2°C limit mean if it is not a “point of no return”?

Two degrees of warming is often thought of as a value after which global warming becomes “dangerous”. The 2°C target was perhaps first made prominent by William Nordhaus in the late 1970s3. He appears to have chosen this amount of warming as it was thought that this might represent the upper boundary of global temperatures that had been experienced during the Holocene Epoch of the past ~10,000 years4.

Over the 1980s and 1990s, the conventional wisdom coalesced around 2°C as an amount of warming that should be avoided but there was never a scientific consensus that 2.0°C represented some well-defined bright line where impacts suddenly became much worse or feedbacks suddenly became completely self-perpetuating5. It is perhaps not surprising that it is very difficult to define a single value for “dangerous” warming since the definition of “dangerous” will inevitably depend on the impact being discussed, the geographic location, the timescale, and the risk tolerance.

Nevertheless, it is useful to have an official objective to organize mitigation policy around. By 1992, the United Nations Framework Convention on Climate Change adopted the official objective of stabilizing global temperature at a level that would “avoid dangerous anthropogenic interference with the climate system”6 and the 2009 Copenhagen Accord defined this limit to be 2°C7. The 2015 Paris Accord affirmed this 2°C goal but also articulated ambitions for limiting global temperature to 1.5°C8,9. This new ambition for limiting global warming to 1.5°C spurred the IPCC to release a report in late 2018 on the impacts associated with global warming of 1.5°C (2.7°F) above pre-industrial levels as well as the technical feasibility (from an energy systems perspective) of limiting global warming to such a level10.

The 2018 IPCC report showed that 1.5°C might be crossed as early as 2030 (12 years after the report was released in 2018) under the current rate of warming. Another important calculation related to 2030 was that in order to avoid 1.5°C in the long run, global CO2 emissions would have to be reduced by 45% by 2030 and reach net zero by 2050.

The media coverage of the 2018 IPCC report often reported something to the effect of “The IPCC concluded that we have until 2030 (or 12 years) to avoid catastrophic global warming”. This was not the conclusion of the report11. For one thing, the word “catastrophic” did not appear in the IPCC report. This was because the report was not tasked with defining a level of global warming which might be considered to be catastrophic (or dangerous) but rather was tasked with evaluating the impacts of 1.5°C of warming and comparing them to 2.0°C of warming. The report found that impacts are likely to be measurably worse at 2.0°C of warming compared to 1.5°C of warming. However, the report does not identify any bright-line after which impacts suddenly explode in severity.

Most impacts scale with the amount of global warming and the amount of global warming scales with cumulative anthropogenic CO2 emissions. Thus, the severity of impacts can be reduced by reducing anthropogenic CO2 emissions regardless of if/when the 1.5°C or 2.0°C values are passed. The New Yorker article is wrong to assert that there is some bright line at 2.0°C of warming after which we are condemned to catastrophe and human decisions no longer matter.

Alexis Berg member picture

Alexis Berg

Research Associate, Harvard University

This is inaccurate. I think there is some confusion here.

Historically, 2°C has been chosen as some kind of internationally agreed-upon “speed limit” to warming, resulting from a mix of some legacy from earlier scientific discussions about global warming, considerations on the range of past climate variations and analyses of the possible impacts of climate change. For instance see here.

So this target doesn’t mean that 1.9°C is safe and 2.1°C or 2.5°C is guaranteed catastrophe. In particular, it doesn’t imply runaway climate change beyond 2°C.

Now, last year, a paper came out in PNAS, Trajectories of the Earth System in the Anthropocene1, which did suggest that as early as by 2°C warming, some feedbacks could start kicking in in the climate system that could automatically push the Earth towards a “Hothouse climate”, i.e., 4 or 5°C warming (indeed, a global catastrophe)—feedbacks like methane emissions from melting permafrost, carbon emissions from ecosystem collapse, etc. In other words, the authors suggested that unless we stabilize the climate now at 2°C, then beyond that threshold it would run away uncontrollably towards a hothouse climate, and that the intermediate space between 2 and 5°C, in a way, did not actually exist.

It is worth pointing out that, while that possibility can’t be excluded, it does not represent, at least to my knowledge, the consensus amongst scientists. This was a speculative paper intended to highlight, I believe, the high side of the risk distribution.The authors offered, at the time, no real new evidence or climate simulation analysis to substantiate their claims. Climate model simulations, for instance, which do include some of these feedbacks, do not suggest runaway climate change beyond 2°C.

So, although the author here doesn’t cite that paper explicitly, what I think happened is that he took the possibility raised somewhat speculatively by this article and interpreted it as certainty and reflecting scientific consensus. This is, I believe, clearly an exaggeration. Unfortunately, since the author also factors in the fact that, in his view, limiting warming to 2°C won’t be possible (it is fair to recognize that it is becoming extremely challenging, as years go by and CO2 accumulates), it leads him to the conclusion that doom is unavoidable, as we’ll cross into beyond-2°C territory and thus will be automatically pushed towards Hothouse climate. This mistaken conclusion that doom is certain is the basis for much of the discussion in the rest of the article on climate actions and priorities and personal attitudes (e.g., hope).

Marcos Fontela member picture

Marcos Fontela

Postdoctoral researcher, Institute of Marine Research (IIM-CSIC)

The whole essay can be dismantled with a single article: Trajectories of the Earth System in the Anthropocene, published in 2018 Steffen et al1.

Biogeophysical feedbacks have different tipping points. Some are in the range of the 2ºC limit, while others would occur at higher temperature anomalies. For example, a critical transition in the Atlantic Meridional Ocean Circulation (AMOC) is not expected unless beyond 3ºC.

Potential interactions among the tipping elements of the Earth system could generate tipping cascades, but the farther we stay below 2ºC [or a higher level of warming], the less likely will be the occurrence of tipping cascades.

Science Feedback is a non-partisan, non-profit organization dedicated to science education. Our reviews are crowdsourced directly from a community of scientists with relevant expertise. We strive to explain whether and why information is or is not consistent with the science and to help readers know which news to trust.
Please get in touch if you have any comment or think there is an important claim or article that would need to be reviewed.

Published on:


Related Articles