Climate change is not linear: why we need to update the Paris Agreement

If you’ve heard anything about climate change, you’ve probably heard about tipping points: thresholds in certain geophysical systems that trigger irreversible changes once crossed.

But you might not have heard of a ‘cascade’, in which multiple tipping points are crossed at once and some of the resulting changes eventually cross another cluster of tipping points, and so on.

Researchers have been investigating this possibility for a number of years. What they’ve put forward might completely reframe your sense of urgency around climate action.

How geophysical cascades turn 2ºC into 6ºC

‘In our view, the clearest emergency would be if we were approaching a global cascade of tipping points that led to a new, less habitable, ‘hothouse’ climate state’2

The general picture which emerges from the scientific literature on this topic is that all of the different tipping points fall into roughly ‘three clusters based on their estimated threshold temperature’, with each one acting like a domino for the one before it:3

1ºC-3ºC

These will be the first to go, as we’re already at (roughly) 1.4ºC.

This cluster includes the collapse of the Greenland and West Antarctic ice sheets, a summer-ice-free Arctic, deglaciation of the Alps, death of tropical coral reefs, thawing of the northern permafrost, the breakdown of the Labrador Sea Current and a shift in the West African monsoon.

In 2021, Professor Will Steffen estimated that the tipping point for a summer-ice-free Arctic in the long term has already been crossed, and the same is the case with tropical coral. Otherwise, he estimates that we’ll tip the West Antarctic in the next 5 or so years and tip Greenland halfway through the 2040s.4

3ºC-5ºC

The Northern hemisphere’s boreal forest shift (southern dieback combined with northern expansion), the loss of the Amazon, the breakdown of the Atlantic Meridional Overturning Circulation (AMOC), as well as the complete collapse of northern permafrost, East Antarctic glacier collapse, Barents Sea ice, and most mountain glaciers.

There’s a complication with a few of these, too. Whilst Amazon dieback without extensive deforestation is, indeed, expected to occur between 3ºC and 5ºC, it can start between 1.5ºC and 2ºC, if we get to 20-25% deforestation. Estimates from 2021 suggest that we’re at 17%.5

Similarly, some recent estimates of the temperature range for the collapse of the AMOC collapse have been a quite a lot lower, with a peer-reviewed paper in 2024 suggesting that this crucial component of thermohaline circulation could cross a tipping point between 1.1 and 3.8C.6

5ºC+

Winter-ice-free Arctic and East Antarctic ice sheet collapse.

If you know anything about older epochs of our planet’s climate, this is the moment that an interglacial period reaches its thermal maximum – in other words, Hothouse Earth.

To understand how tipping these systems has the potential to ignite a runaway transition to a completely different climatic epoch, consider these estimates of the warming effect of a selection of relatively close-at-hand tipping points:3

Tipping point	Expected warming (est. range)
Permafrost thawing	0.09ºC (0.04–0.16ºC)
Relative weakening of land and ocean carbon sinks	0.25ºC (0.13–0.37ºC)
Increased bacterial respiration in the ocean	0.02ºC
Amazon forest dieback	0.05ºC (0.03–0.11ºC)
Boreal forest dieback	0.06ºC (0.02–0.10ºC)
Total	0.47ºC (0.24–0.66ºC)

So, in total, tipping these geophysical systems would ensure that even if we met our most stringent obligations under the Paris Agreement and limited warming to 1.5ºC (a highly unlikely prospect, according to lauded climatologist Professor James Hansen), additional warming from tipped systems could bring us up to or over 2ºC, violating that treaty wholesale.⁸

But if we’re taking into account both realistic phase-down pathways and an up-to-date understanding of aerosol cooling,⁹ we’re likely to be heading over 2ºC – that is, unless we replace the aerosols released by burning fossil fuels with our own manufactured alternative.

By the time we’re in the region of 2.5ºC, we’re flirting with the collapse of the AMOC, which has the potential to increase temperatures in pretty much all regions of the Northern Hemisphere besides the UK, Ireland, Scandinavia and Western Europe (which will cool dramatically) by between 0.5 and 1.5ºC.10

This is what it means to think about climate change as a non-linear process, and it could potentially obsolesce much of the geopolitical framework that we’re currently using to deal with it.

Why we need an ‘early warning’ amendment to the Paris Agreement

‘Based on this analysis of tipping cascades and taking a risk-averse approach, we suggest that a potential planetary threshold could occur at a temperature rise as low as ∼2.0 °C above preindustrial’.3

‘…at the international level, there is the potential for a “tail risk treaty”: an agreement or protocol that activates stronger commitments and mechanisms when early-warning indicators of potential abrupt change are triggered.’ 7

One of the most important treaties in human history, the Paris Agreement may have a fatal flaw.

By binding signatories only to limiting warming, we may have given ourselves permission to allow polluters to tip our climate into total breakdown as long as they do so at lower temperatures.

But clearly, it is as important to not cross tipping points as it is to not increase temperatures. So, why not formally recognise this responsibility as part of the Paris Agreement?

Such an amendment would alter obligations in the event that some established threshold of risk had been breached and in the meantime, require signatory-members to build up their capacity to meet these obligations.

As things stand, we are completely unprepared for our near future, limping along in a delusion that temperature is all that matters.

Establishing a global network for monitoring tipping points

To effectively enforce (and in the short term, demonstrate the need for) an ‘early warning’ amendment to Paris, we need to massively expand our monitoring networks.

In this endeavour, fostering international cooperation in methodologies and the free sharing of data will be absolutely critical.

Those in charge of these networks will need to be able to speak with one voice when submitting their assessments to the authorities, just as peer-reviewed research is playing a crucial role in sounding the alarm today.

Whilst it is unlikely to be very expensive, such a network is likely to face accusations of alarmism from within and without the scientific community, let alone political resistance to its strengthening of the Paris Agreement.

But its time has come. If we do not increase our focus on tipping points, we may well win the war on increasing temperatures but lose our climate.