Hide table of contents

Past societies often found their end amid a challenging climate. Environmental changes can put considerable strain on a society and there are many cases in history where it was a major contributor to their collapse, especially when the change was sudden. Today we are also living through a significant shift in global climate, which makes it likely that we reach temperatures last seen on Earth 20-50 million years ago. While this change is massive, so far it has been relatively slow and predictable. This could change if we start pushing the Earth system more and more out of its old equilibrium and possibly trigger its tipping points.

What are tipping points?

The origin of the idea of tipping points in the climate system is from a 2008 paper by Lenton et al.. Complex systems have a natural ability to bounce back to their original state, so long as they are not pushed too far. A tipping point is a point in the system where any additional pressure will make it impossible to go back to how it was before (Figure 1). Before the tipping points feedbacks in the system tend to bring it back to its original state. Once the tipping point is crossed it won’t.

Tipping Schematic

Figure 1: Schematic presentation of a tipping point. When a system moves away from its previous stable state, it gradually approaches a critical threshold called the tipping point. Once this tipping point is crossed, the system undergoes a significant shift without requiring further external influence and moves towards a new stable state.

Applied to our climate system tipping points can be found in large scale elements of the Earth system like the Amazon rainforest, the Greenland Ice Sheet, or the Atlantic Meridional Overturning Circulation. These are called tipping elements. If these tipping elements are pushed beyond their tipping points, this would change Earth permanently. For instance, let’s consider the Greenland Ice Sheet. Currently, it is melting, but we’re unsure if it has reached a critical point of no return. This ice sheet has a significant impact on the local climate through a continuous cycle. It is composed of layers of ice, piling up to heights of 2000-3000 meters. As a result, the air at the top is cooler. Moreover, due to its high albedo, the snow cover of the ice sheet reflects a substantial amount of solar radiation, leading to further cooling. However, when the overall global climate becomes excessively warm, the snow cover starts to melt. This melting process triggers two effects: the sheet’s height decreases, and it becomes darker as the snow cover melts and the exposed ice is darker than the snow. Both of these changes contribute to a rise in local temperature, causing even more ice to melt. Additionally, as the ice sheet melts, the resulting water flows down to the base of the glacier. This water acts as a lubricant, making the glacier move more rapidly. Consequently, these faster-moving glaciers reach the ocean quicker, where they break off as icebergs. In combination, a feedback loop is initiated that cannot be halted until all the ice has melted. In the case of the Greenland Ice Sheet, this scenario could result in an additional sea level increase of approximately 7 meters within the next few thousand years.

The Earth system as a whole likely does not have a single tipping point. Instead it has a complex interaction of separate large tipping elements, who each have individual tipping points. This complex interaction is hard to predict, but entails the risk of a so-called tipping cascade. This means that the tipping elements get in a positive feedback loop and start to push each other outside the safe territory and over their tipping points. The process and problems of tipping cascades are nicely described by Armstrong McKay et al. (2022). The process is similar to the feedback loops for a single tipping element and is mediated by influence on the global temperature and other processes in the climate system. This means if one tipping element like the Greenland Ice Sheet is tipped, it changes the global temperature, while also influencing processes like the salinity of the water in the North Atlantic. This increased temperature and change in salinity in turn could trigger another tipping element, in the case of Greenland Ice Sheet most likely the Atlantic Meridional Overturning Circulation, as it strongly depends on salty water in the North Atlantic to keep the circulation going. This self-reinforcing process could theoretically continue until all tipping elements in the Earth system are beyond their tipping point. Armstrong McKay et al. show that such a cascade might even get started in the temperature range of 1.5-2°C. Indeed, some tipping elements could already be beyond their point of no return.

Tipping element interactions

The original study by Lenton et al. identified 15 tipping elements, which they deemed policy relevant. A more recent study by Ripple et al. (2023) used a literature review to identify 41 tipping elements and feedback loops. However, not all of those tipping points are the same magnitude and many of the tipping points identified by Ripple et al. (2023) likely only have little effect on the overall trajectory of the Earth system. Therefore, most papers only focus on a subset.

One interesting study here is by Wunderling et al. (2021). They selected the four tipping points that have been identified as being the most likely to start a tipping cascade: Greenland Ice Sheet, the West Antarctic Ice Sheet, the Atlantic Meridional Overturning Circulation (AMOC), the El Niño Southern Oscillation (ENSO) and the Amazon rainforest. Wunderling and coauthors represent all of the tipping points as a collection of separate differential equations that are all coupled with each other. This coupled model has a collection of parameters that account for the timescale and sensitivity of tipping for each tipping element and the strength of interaction between them. They use a Monte Carlo approach to run the model and show that even in a world with only 2°C warming, we have a 61 % chance that at least one of the major tipping elements will be triggered and in 39 % of the simulations a cascade of two or more tipping elements is started.

Building on this model, the same working group published a follow up study (Wunderling et al., 2023) to assess how much leeway we have when it comes to overshooting the temperatures that might cause tipping. The idea here is that the actual tipping point is characterized not only by temperature, but also by the time you stay in this zone of elevated temperatures. Therefore, to decrease the chance of triggering the tipping point you can either make sure that you keep the temperatures low, or if you fail to do so, you can at least try to get below the threshold quickly. This means the danger zone is a function of how much you overstep the threshold temperature and how long you stay above it.

Their results highlight the need to keep to the 1.5°C aim of the Paris Agreement or, better yet, to keep temperatures below 1.0°C of warming above pre-industrial levels, at least in the long term. In their model, if the long term convergence temperature is 2°C, there is a 82% chance of tipping at least one tipping element, even if the maximum global temperature never rises above 2°C as well. However, that’s the long run, and their results also show that we possibly have considerable time to get back to a lower temperature. For many of the simulations there seems to be a time span of 100-200 years to get back to a lower convergence temperature. That said, this only seems to be true so long as the peak temperature is below 2.5°C. If we go above this temperature, the time we have to get back to lower temperature decreases considerably. This is concerning, as currently our world seems to be aimed at a temperature somewhere between 2 and 3°C of warming.

An earlier modeling study by Ritchie et al. (2021) came to similar conclusions as Wunderling et al., but saw a lower chance of tipping cascades. However, Ritchie et al. did not include direct interactions between tipping elements and only mediates all changes via temperature, while the Wunderling et al. model includes additional interactions based on physical knowledge of how the tipping elements influence each other.

There are a lot of studies that look into tipping elements. However, most of those are focussed on the description of the physical processes and are less focussed on the effects on the climate in combination with other tipping elements. Currently, the studies mentioned here are the main path of research on tipping cascades and how interacting tipping points react to a variety of temperatures.

If you know of important studies I’ve missed, please email me and I’ll update this article!

How warm could it get?

Assessing the ultimate temperature impact of tipping points is quite difficult, as the potential ranges of the additional warming caused by them are quite wide. However, we can still make a rough assessment. Armstrong McKay et al. (2022) provides a summary of estimates for the effects of the 15 tipping elements identified by Lenton et al.. If we naively add up all those values we end up with an additional warming of ~1.17°C.

This is quite a lot, but does not tell the whole story, as the tipping elements differ considerably in their likely trigger temperature, whether their contribution to temperatures is positive or negative, and the timescale over which this warming would unfold. If we only add together tipping elements with an estimated trigger temperature below 3°C we end up with ~0.04°C additional warming. If we instead add together tipping elements with an estimated time scale below a hundred years, we have roughly 0.1°C of additional warming. This means that the global effect in the foreseeable future is detectable, but not massive. Still, those changes might be much bigger regionally. For example, the dieback of the Amazon rainforest could happen in as little as 50 years and could increase local temperatures by an additional 2°C .

How bad is our situation?

This is the trillion dollar question and the honest answer might just be: we’re not exactly sure. As hopefully became clear in this post, there is not that much research about what would happen if we started a tipping cascade. However, given paleoclimatic evidence, we do know that the abrupt changes described here can happen. This has been acknowledged in the most recent reports by the IPCC and the overall state of knowledge about this was exhaustively explored in a 2022 literature review by Boers et al.. An example here would be the so-called Dansgaard-Oeschger Events. These events are characterized by a complex interaction of sea ice, atmospheric dynamics and the Atlantic Meridional Overturning Circulation which leads to an abrupt temperature increase of up to 16°C in high northern latitudes in less than ten years.

There have been calls for more research into this area and higher temperature climate change in general (see for example Kemp et al. (2022)) and there are also calls for creating a special report of the IPCC about exactly this topic (Boers et al., 2022). But at the moment, this research has yet to be produced and summarized. This research could be challenging, as tipping points are not considered in most of the major models used to model climate change (Kemp et al., 2022). It’s not even clear whether the global climate models we use are even up to that task (Balaji et al., 2022). Boerst et al. highlight several times that most of our current models struggle to replicate abrupt transitions in our paleoclimatic record, due to a stability bias and insufficient spatial resolution. This means it is hard to trigger extreme events, even if you apply massive forcing. However, when it comes to tipping points, extreme events are exactly the thing we want to study.

Still, my conclusion from writing this is a grim optimism. Don’t get me wrong, triggering any of the tipping points is very bad. I think it would be a monument to our failure if we destroyed the great biodiversity in the Amazon rainforest. However, it seems like tipping points cannot easily contribute an additional amount of warming that would put Earth as a whole on a >3°C trajectory, at least in the next hundred years. Also, many of the biggest impacts will be felt regionally, not globally. Finally, the biggest impacts of the tipping points will be to the general functioning of the biosphere and not to the temperature directly. This means my optimism is more in the form that I now think that tipping points alone are unlikely to cause a collapse of global civilization due to temperature changes. Still, we should be careful, as many of the past abrupt transitions in climate are not well understood and the Earth system might have some unwelcome surprises for us in store.


Armstrong McKay, D. I., Staal, A., Abrams, J. F., Winkelmann, R., Sakschewski, B., Loriani, S., Fetzer, I., Cornell, S. E., Rockström, J., & Lenton, T. M. (2022). Exceeding 1.5°C global warming could trigger multiple climate tipping points. Science, 377(6611), eabn7950. https://doi.org/10.1126/science.abn7950

Balaji, V., Couvreux, F., Deshayes, J., Gautrais, J., Hourdin, F., & Rio, C. (2022). Are general circulation models obsolete? Proceedings of the National Academy of Sciences, 119(47), e2202075119. https://doi.org/10.1073/pnas.2202075119

Boers, N., Ghil, M., & Stocker, T. F. (2022). Theoretical and paleoclimatic evidence for abrupt transitions in the Earth system. Environmental Research Letters, 17(9), 093006. https://doi.org/10.1088/1748-9326/ac8944

Kemp, L., Xu, C., Depledge, J., Ebi, K. L., Gibbins, G., Kohler, T. A., Rockström, J., Scheffer, M., Schellnhuber, H. J., Steffen, W., & Lenton, T. M. (2022). Climate Endgame: Exploring catastrophic climate change scenarios. Proceedings of the National Academy of Sciences, 119(34), e2108146119. https://doi.org/10.1073/pnas.2108146119

Lenton, T. M., Held, H., Kriegler, E., Hall, J. W., Lucht, W., Rahmstorf, S., & Schellnhuber, H. J. (2008). Tipping elements in the Earth’s climate system. Proceedings of the National Academy of Sciences, 105(6), 1786–1793. https://doi.org/10.1073/pnas.0705414105

Ripple, W. J., Wolf, C., Lenton, T. M., Gregg, J. W., Natali, S. M., Duffy, P. B., Rockström, J., & Schellnhuber, H. J. (2023). Many risky feedback loops amplify the need for climate action. One Earth, 6(2), 86–91. https://doi.org/10.1016/j.oneear.2023.01.004

Ritchie, P. D. L., Clarke, J. J., Cox, P. M., & Huntingford, C. (2021). Overshooting tipping point thresholds in a changing climate. Nature, 592(7855), 517–523. https://doi.org/10.1038/s41586-021-03263-2

Wunderling, N., Donges, J. F., Kurths, J., & Winkelmann, R. (2021). Interacting tipping elements increase risk of climate domino effects under global warming. Earth System Dynamics, 12(2), 601–619. https://doi.org/10.5194/esd-12-601-2021

Wunderling, N., Winkelmann, R., Rockström, J., Loriani, S., Armstrong McKay, D. I., Ritchie, P. D. L., Sakschewski, B., & Donges, J. F. (2023). Global warming overshoots increase risks of climate tipping cascades in a network model. Nature Climate Change, 13(1), Article 1. https://doi.org/10.1038/s41558-022-01545-9





More posts like this

Sorted by Click to highlight new comments since: Today at 7:57 PM

Thanks for this! I One study that seems important in this context but is missing is Wang et al 2023 which is a review paper of the tipping point literature.

I've read it now and it was quite interesting. Though it did not really shift my conclusions. The only update I had was that we might even know less about the long term consequences (2100+) than I thought before. 

Thank you for the recommendation! I'll read through it and update the permanent version on Github. 

I'm interested in maybe a bit of expansion of your conclusions here. Your last two sentences say:
"This means my optimism is more in the form that I now think that tipping points alone are unlikely to cause a collapse of global civilization due to temperature changes. Still, we should be careful, as many of the past abrupt transitions in climate are not well understood and the Earth system might have some unwelcome surprises for us in store."

What do you think this means for the overall contribution of tipping elements to global civilisation collapse? Am I to interpret the "alone" and "due to temperature changes" as suggesting that tipping elements, in concert with other factors, may have a contribution, if not alone and just due to temperature changes? Or is this merely that you don't think that this part of the literature review can say anything beyond that narrow scope? If the latter, will you be investigating the other impacts?

I think that tipping elements could make a significant contribution to the destabilization of global civilization, which ultimately could contribute to collapse. This would likely not happen via temperature, but by other disruptive elements like significant sea level rise or destruction of ecosystem. However, the main effects of this are likely beyond 2100. Therefore, I am really unsure how this will ultimately play out. I think to make a good estimate of this we are currently knowing too little. Hopefully, the next special report of the IPCC will be about this. This would make things likely much clearer. Therefore, I'll probably not investigate this much further right now, as things seem to uncertain. 

"Armstrong McKay et al. (2022) provides a summary of estimates for the effects of the 15 tipping elements identified by Lenton et al.. If we naively add up all those values we end up with an additional warming of ~1.17°C.

This is quite a lot, but does not tell the whole story, as the tipping elements differ considerably in their likely trigger temperature, whether their contribution to temperatures is positive or negative, and the timescale over which this warming would unfold. If we only add together tipping elements with an estimated trigger temperature below 3°C we end up with ~0.04°C additional warming."

From this, would you agree it is most important to avoid a >3C world given that below that tipping points seem small?

This seems quite at odds with how this paper has been discussed in the media.

Yes, that is how I would interpret their Table S4, which seems like the main summary of their findings. 

What was your impression of how the media represented their findings? It feels to me like the media often represents tipping points as happening instantaneous, while most of them are rather in the time scale of centuries. 

However, you could make an argument that staying much below that is also sensible, as the tipping points are not only triggered by temperature, but also by physical processes like the dilution of salt concentrations in sea water. 

Also, for writing this section I used the estimated values. If you use the minimal values for triggering the tipping points the picture becomes more grim. 

Just did a quick calculation. If you assume the minimal value as the trigger, you get ~0.61°C additional warming at 3°C warming. 

Also, a lot more of the points are triggered at lower warming than this. 

How did you do those calculations? I get different results from looking at the paper Table 1 (e.g. under 3C we would expect a cooling on central estimates?)

I used table S4, which includes a longer list of possible tipping points. 

Thank you for the article and the comments! While I agree with the finding that additional warming  of tipping points seems limited in this century based on the current scientific understanding, I'd be  quite hesitant to conclude that avoiding warming >3°C only is most important. 

Even without strong direct, near- to medium-term impacts on global warming, tipping elements will have significant (regional) implications on ecosystems, human welfare etc. (Or, as Wang et al 2023 put it: "Overall, even considering remaining scientific uncertainties, tipping elements will influence future climate change and may involve major impacts on ecosystems, climate patterns, and the carbon cycle starting later this century. Aggressive efforts to stabilize climate change could significantly reduce such impacts.")

In combination with the significant remaining uncertainty around tipping elements / tipping points, applying the precautionary principle and avoiding as much warming as possible appear to be good strategies, as also called for by many authors of the underlying studies/papers themselves.  

I don't think anyone disagrees with it being important to reduce emissions as much as possible, the question is about relative importance.

Thanks. Yes, the marginal damage of missing the 3°C threshold by 0.1°C is higher than the marginal damage of missing the 1.5°C threshold by the same amount - and this gap is widening if we include tipping points. 

However, benefits are cumulative - i.e. staying below e.g., 2°C reduces the risk of damages and tipping elements at that temperature threshold and those of higher temperatures. A ton of CO2 we avoid today contributes to both goals. So even in terms of relative importance, I would still disagree. 

Thanks! Again it's unclear what/who you are disagreeing with as no one is disputing that a ton permanently avoided now helps in all futures.

To clarify: The point that Florian and I were making was simply what you state that you agree with -- namely that tipping points make the nonlinearity of expected climate damage worse rather than, as often argued, flattening it because nearby tipping points could easily catapult us in really high-warming futures.

This is an important point for impact-oriented philanthropists because we are, de facto, choosing between solutions with different expected performance in different futures.

Hi ! Adressing tipping points is an interesting topic. It's good to know that they alone won't cause extreme warming (even though yes, we ought to be cautious).

Have you looked into tipping points for the topic of biodiversity collapse, or for societal collapse ? (I adress societal collapse a bit in post 2 of this series)


As an aside, may I suggest improving the formatting a bit? I felt quite threatened by the size of the paragraphs. I think the post could benefit from shorter paragraphs, using bolding, bullet points. This post provides some advice: https://forum.effectivealtruism.org/posts/dHHuEYdbMqBf2deyj/using-the-executive-summary-style-writing-that-respects-your

Hi Corentin. Thanks for the comments. I plan to also look more into biodiversity and societal tipping points, but I haven't yet found the time. 

Concerning the reformating, maybe it's just me, but I have a much harder time reading those executive summary style posts and therefore I would rather leave it the way it is. 

Concerning the reformating, maybe it's just me, but I have a much harder time reading those executive summary style posts and therefore I would rather leave it the way it is. 

Wow, that's surprising. Does this include even adding line breaks ? 

Or some bolding ?

Hmm I feel like this is already a lot of line breaks. Most of the paragraphs are only ~ 5 sentences or less.

And at least for me bolding breaks the reading flow. 

More from FJehn
Curated and popular this week
Relevant opportunities