Climate Tipping Points: Taking a Narrative Not Numbers Approach

Climate tipping points have rapidly moved from scientific journals to the centre of global media attention, driven by headlines of accelerating polar ice loss and warnings of an impending Atlantic current collapse. The financial sector is paying close attention. A recent report by J.P. Morgan identified tipping points as an under-modelled "deep uncertainty" that could pose immediate cash loss risks to portfolios and fundamentally challenge standard economic models, a concern that sits well outside the scope of most institutions' current risk frameworks.

Yet, despite this growing alarm, tangible action by institutions remains stalled. Decision-makers find themselves stuck because the data required to drive traditional risk frameworks is limited. Standard hazard and economic models are built to project gradual, continuous warming, making the abrupt, unpredictable nature of a tipping point structurally difficult to quantify.

With that in mind, this blog post explores exactly what climate tipping points are, outlines our current scientific understanding of the major systems at risk, and presents a practical narrative-based approach for risk managers to assess these high-impact and non-linear events.

When we think of climate change, we often picture a slow, steady increase in temperatures and a proportional rise in extreme weather. However, the Earth's climate system is not linear. It contains critical thresholds known as tipping points.

Scientifically, a tipping point is a critical threshold at which a tiny perturbation can qualitatively alter the state or development of a system. Once crossed, these thresholds trigger abrupt, self-amplifying, and often irreversible changes, fundamentally shifting how the Earth operates.

Tipping point

A critical threshold at which a tiny perturbation can qualitatively alter the state or development of a system. Once crossed, these thresholds trigger abrupt, self-amplifying, and often irreversible changes, fundamentally shifting how the Earth operates.

To understand how the climate changes, scientists often look at the stability of a system using a "ball and valley" analogy. In this framework, the climate resides in a "stable state" (the bottom of a valley) where it can withstand minor fluctuations and return to equilibrium.

A tipping point occurs when the system is forced out of its current valley and into an alternative one, fundamentally changing its behaviour. According to the IPCC, there are two distinct ways this transition happens.

The Global Tipping Points Report produced by Professor Tim Lenton in 2023 brought together more than 200 researchers from 90 organisations to provide a synthesis of the Earth System's tipping points. The team identified more than 25 parts of the Earth System with tipping points and highlighted around 15 major tipping points.

The assessment of these tipping points spans:

• The cryosphere (the frozen world)
• The biosphere (our living green planet)
• The ocean-atmosphere system (the currents and winds that regulate global weather).

In the cryosphere, the most significant are the Greenland and West Antarctic ice sheets, whose collapse would drive sea-level rise of several metres over centuries; the collapse of mountain glaciers worldwide; and the thawing of boreal permafrost, which stores vast quantities of carbon that, once released, would further accelerate warming.

In the biosphere, the Amazon rainforest is a clear example. Around 17% of the Amazon has already been deforested and models suggest that somewhere between 20% and 25% of deforestation, combined with regional drying driven by warming, could push the forest past a dieback tipping point, converting large portions from tropical rainforest to savannah. The carbon release from that transition, and the loss of the Amazon's role as a carbon sink, would be globally significant.

In the ocean-atmosphere system, the most discussed is the Atlantic Meridional Overturning Circulation (AMOC), which we return to below. Others include the collapse of low-latitude coral reef systems, now experiencing mass bleaching events at warming levels once considered safe, and shifts in the West African and Indian monsoon systems that affect rainfall patterns for billions of people.

The most pressing question surrounding climate tipping points is exactly when they will be triggered and how fast changes will take effect. According to a definitive 2022 study published in Science, there is no single global trigger; different systems tip at different Global Warming Levels (GWLs).

Global Warming Levels (GWLs) represent the increase in the Earth's average surface temperature relative to the pre-industrial period (1850 to 1900), serving as a standardised metric to assess the severity of climate impacts at specific stages of warming. At the GWL level of approximately 1.3°C to 1.4°C, the Earth is already entering the lower uncertainty range for five critical tipping elements. As warming approaches 1.5°C, these systems including the Greenland Ice Sheet, the West Antarctic Ice Sheet, and low-latitude coral reefs transition into a state where self-perpetuating collapse becomes highly probable.

But the most important takeaway for the tipping points trigger is the level of uncertainty that scientists have in these estimates. For example, while the central estimate for the collapse of the Greenland Ice Sheet is 1.5°C, the scientifically plausible range for this trigger starts as low as 0.8°C and extends up to 3.0°C. Similarly, the Amazon Rainforest dieback has a central estimate of 3.5°C, but the uncertainty range spans from 2.0°C to 6.0°C. Further, this uncertainty spans across multiple Shared Socioeconomic Pathway (SSP) scenarios such that a single tipping point could theoretically be triggered under our most optimistic, low-emissions future, or it might only be reached in a worst-case, high-emissions trajectory.

The speed of onset following these triggers varies significantly by system. Certain tipping elements respond on decadal timescales; for instance, once the threshold for coral reef mortality is reached, the transition can be completed within 10 to 50 years. In contrast, "slow" tipping elements like the polar ice sheets have immense physical inertia. While the commitment to their collapse may be triggered at 1.5°C GWL, the full transition into a new state may unfold over centuries or millennia. This creates a "commitment lag," where human activity today determines the irreversible physical state of the planet for thousands of years to come.

Recent reports of cooling off the coast of Greenland (the cold blob) and accelerating ice melt in Greenland have sparked a surge of media interest in the stability of AMOC.

The Atlantic Meridional Overturning Circulation (AMOC) is a large-scale ocean current system that acts as a global heat conveyor, transporting warm water northward from the tropics toward Northwestern Europe and the Arctic.

However, as the Greenland Ice Sheet melts, it dumps massive volumes of fresh, buoyant water into this system, jamming the sinking mechanism. This is a classic example of a bifurcation tipping point: as the Earth warms, the "valley" maintaining the AMOC's stability grows shallower, making the system increasingly vulnerable.

Observational data from the RAPID array confirms the circulation is already slowing. The scientific challenge lies in determining exactly when this slowdown crosses into irreversible collapse. The 2022 Armstrong McKay assessment places the central estimate for an AMOC tipping point at a Global Warming Level of 4.0°C. Yet the uncertainty range spans from 1.4°C to 8.0°C. The IPCC AR6 similarly maintains that while a full collapse in the 21st century is very unlikely, the risk cannot be eliminated.

Because this wide temperature uncertainty spans nearly every Shared Socioeconomic Pathway (SSP) scenario, it presents a fundamental problem for financial risk modelling. Under the SSP trajectories commonly used in bank stress testing, which project median warming between 2.7°C and 3.6°C by 2100, an AMOC collapse is neither explicitly modelled nor safely excluded.

The deep uncertainties surrounding climate tipping points create a fundamental problem for financial institutions and risk managers.

Because we do not know precisely where most tipping point thresholds sit, or how fast the transitions occur, we cannot reliably translate these events into standard quantitative metrics. Attaching precise numbers such as expected annual losses or flood depths to a system like the AMOC would require a level of quantitative confidence that current science simply does not provide. Doing so would offer false precision, burying a deep, structural uncertainty beneath a layer of manufactured mathematics.

The appropriate response to this lack of quantitative certainty is not to ignore the risk, nor is it to force the science into a spreadsheet. Instead, institutions must build structured narrative scenarios.

A narrative scenario is a scientifically grounded, qualitative description of a specific tipping point and its systemic consequences. Rather than attaching a spurious probability to when an event will happen, a narrative explores what happens when it does. This approach, currently utilised by bodies like the IPCC and the Network for Greening the Financial System (NGFS), allows organisations to examine portfolio exposure to scientifically credible risks without embedding numbers that cannot be justified.

To understand this approach, consider a narrative scenario based on recent research from the University of Exeter detailing the impact of an AMOC collapse on UK agriculture.

The research indicates that an AMOC collapse would plunge the UK into a drastically colder and intensely drier climate. Under these conditions, the vast majority of the land currently used for arable farming simply ceases to be viable. The traditional "breadbaskets" of the country would no longer support staple crop production, forcing a rapid, widespread abandonment of current farming practices.

But even to construct their narrative and estimates, the authors have to make assumptions that are not scientifically grounded in data, therefore they assume AMOC collapse occurs over the period 2030 to 2050. This does not represent a weakness of the study, but it highlights that to construct these case studies, strong assumptions have to be made as, we do not yet have the scientific precision on when and how quickly tipping points could occur.

A narrative approach shifts the focus from "when will this happen?" to "how exposed are we?" For a financial institution, the UK farming case study becomes a practical tool to identify real vulnerabilities. It prompts essential questions about the concentration of agricultural debt in specific regions, or the resilience of portfolios to sudden systemic food inflation, and the degree to which existing stress tests model the right kind of discontinuity.

Ultimately, the window between what is currently considered "unmodellable" and what becomes a regulatory requirement is closing. Acknowledging the limits of scientific precision is not a failure of a risk framework; it is its foundation. By building the conceptual infrastructure to assess these narrative risks today, institutions ensure they are not starting from zero when market forces and regulatory standards eventually demand quantitative certainty.