and why the absence of one is itself a form of danger
| Key Points |
|
A Gap at the Heart of Global Policy
There is a question that sits, largely unasked, at the centre of every climate negotiation, every infrastructure tender and every government budget.
It is not how warm will the world get, but what will that warmth actually do to us?
The distinction sounds subtle.
It is not.
In late February 2026, a group of leading climate scientists published a commentary in the journal Nature making an assertion that is, on reflection, astonishing: despite decades of climate science, trillions of dollars in research funding and an entire global institution, the Intergovernmental Panel on Climate Change, dedicated to synthesising our understanding of a changing planet, the world has never conducted an internationally mandated assessment of climate-change risk.
Not a science assessment.
A risk assessment.
The difference is the difference between a meteorologist telling you there is a cyclone forming in the Coral Sea and an emergency coordinator telling you which hospitals are below the storm surge line, which roads will be cut, which communities cannot self-evacuate and how many people are likely to die if the levee fails.
The first is knowledge.
The second is the thing that saves lives.
Professor Rowan Sutton, Director of the UK's Met Office Hadley Centre and one of the paper's two lead authors, put it plainly in a statement released alongside the Nature comment: "Despite clear scientific evidence and repeated warnings, the world remains unprepared for the scale and complexity of these challenges."1
The paper, co-authored by nine scientists and co-signed by a broader group of researchers, calls for a global assessment of avoidable climate-change risks, designed to tell policymakers and citizens not just what is happening to the climate but precisely what the consequences are, how severe they might be and, critically, which of those consequences can still be prevented.
What the IPCC Does, and What It Cannot
To understand the gap these scientists are trying to fill, it helps to understand what the IPCC (Intergovernmental Panel on Climate Change) actually does.
Established in 1988, the IPCC is not a research body.
It does not conduct experiments or gather primary data.
It synthesises published scientific literature into Assessment Reports, produced on roughly six-to-seven-year cycles, that summarise the state of knowledge about the physical climate system, the impacts of change and the options for mitigation and adaptation.
Its reports are authoritative, comprehensive and, when you read the underlying chapters rather than the carefully worded Summary for Policymakers, frequently alarming.
But the IPCC's mandate is to be a science assessment, not a risk assessment, and the two are not the same thing.2
As the Nature paper explains, a scientific assessment asks first what is expected to happen and then considers how it might affect society.
A risk assessment inverts that logic.
It asks first what outcomes would be catastrophic for society and then works backwards to determine how likely those outcomes are and what the range of possible trajectories looks like.
This is the methodology used in public health, defence, engineering and finance.
It is the framework that underpins nuclear safety regulation, pandemic preparedness and the design standards of bridges.
It has never, the authors note, been applied comprehensively to climate change at a global, internationally mandated level.
The proposed assessment would run in parallel with the IPCC process, not replace it.
Its purpose is distinct: where the IPCC documents what science knows with the greatest confidence, a global risk assessment would focus on what society most needs to know, including outcomes that are low-probability but high-impact, cascading failures that cross sector boundaries and consequences that existing national-level assessments consistently underestimate.1
The Failures Already Visible
The case for a new approach is not theoretical.
Europe sweltered through a series of heatwaves in recent years that killed thousands of people and triggered wildfires that burnt more than 380,000 hectares of land in Spain alone.4
Heat killed tens of thousands across the continent in the 2003 European heatwave, yet mortality remained underestimated in national emergency plans for years afterward.
Governments that had flood defences did not necessarily have plans for what would happen to their water supply systems when the power grid failed during a concurrent extreme heat event.
They had not planned, in other words, for cascades.
Professor Tom Oliver, from the University of Reading and a co-author of the paper, described the problem this way: "Climate risks are deeply interconnected and can trigger each other in ways that aren't always obvious. Extreme heat harms human health directly, but it also drives drought and crop failure, which can lead to food shortages and civil unrest."5
The Nature paper cites the risk to food systems as one of the clearest examples of this interconnection.
Agricultural yields are threatened by changing precipitation patterns, shifting pest ranges, heat stress to crops, soil degradation and pollinators in decline.
These are not independent risks.
They compound one another, and a simultaneous failure across multiple breadbasket regions, something that existing national food-security assessments rarely model, could trigger cascading humanitarian and economic consequences across the globe.
Similarly, the authors point to the risk of mass casualties from extreme heat in cities such as Belém, in Brazil's Amazon basin, where modelling suggests that wet-bulb temperatures exceeding the limits of human physiological tolerance could arrive within the lifetime of children born today, under high-emissions scenarios.
Policymakers, the paper argues, may be broadly aware that heat deaths will increase.
They are not necessarily aware of the threshold at which conditions become unsurvivable without air-conditioning, the urban populations most exposed, the grid resilience required to sustain that cooling and the public health systems that would be overwhelmed in its absence.
That is the difference between knowing that risk exists and understanding what it looks like at the moment it materialises.3
Cascading Risks and the Limits of Siloed Thinking
The scientists argue that one of the central failures of current preparedness is that risk is assessed in silos.
The health ministry models excess deaths from heat.
The transport department models road damage from flooding.
The energy regulator models peak demand during hot summers.
What almost no authority models is the interaction between all three, simultaneously, during a compound extreme event in which the power grid fails, water treatment plants lose pressure, hospitals lose cooling and transport links are disrupted.
This is not a hypothetical sequence.
It is the sequence that has played out, in partial form, in multiple high-impact events in recent decades: during the 2003 European heatwave, during Hurricane Maria in Puerto Rico in 2017 and during the Texas winter storm of 2021, when the failure of a single system, the power grid, precipitated cascading failures across water, heating, healthcare and food supply.
A global climate-risk assessment, the paper argues, would need to map these interdependencies explicitly, drawing on new methodologies that can model interconnected systems and borrow from risk frameworks used in the financial sector, security analysis and public health.2
The goal is not simply an academic exercise.
It is an operational framework: a living risk dashboard that finance ministries, city planners and disaster-management agencies can actually use.
Such a tool would need to be updated more frequently than an IPCC Assessment Report, which takes years to produce and is outdated in parts by the time it reaches publication.
The pace of climate extremes, the rapid evolution of clean energy technology and the emergence of social tipping points all demand a more nimble instrument, one capable of incorporating new data and revised assessments on a timescale relevant to infrastructure planning cycles and budget rounds.
Tail Risks and the Courage to Look
Perhaps the most politically sensitive element of the proposed assessment is its explicit intention to examine tail risks: low-probability scenarios that are nonetheless plausible and whose consequences would be catastrophic.
Multi-breadbasket failure, in which simultaneous crop losses across several of the world's major agricultural regions overwhelm the global food system, is one such scenario.
Regional uninhabitability, in which large parts of the tropics or subtropics become physiologically unsurvivable during summer months without mechanical cooling, is another.
The collapse of the West Antarctic Ice Sheet, which would commit the world to metres of sea-level rise over coming centuries, is a third.
Current national risk assessments, the paper argues, systematically underweight these scenarios, in part because the political incentive to acknowledge them is low and in part because the science, while pointing clearly at their possibility, cannot assign precise probabilities to events that have no historical precedent.
The Nature paper proposes handling this uncertainty through a framework borrowed from risk engineering: rather than asking for a precise probability, identify a severity threshold, such as five metres of sea-level rise, and assess how the likelihood of crossing that threshold changes as a function of time and of human decisions about emissions.1
This is a more honest way of communicating risk under uncertainty.
It does not pretend to precision it does not have.
But it makes visible the shape of the danger and the degree to which human choices can alter that shape.
Equity, Justice and the Global South
The scientists are explicit that the proposed assessment must not replicate the structural inequities that have distorted global climate governance.
The IPCC has long been criticised for producing knowledge dominated by institutions in the Global North, assessed through disciplinary frameworks that can struggle to capture the lived realities of communities most exposed to climate change.
A farmer in the Sahel whose rainfall patterns have become unrecognisable over a single generation holds knowledge that no climate model fully captures.
A low-lying Pacific atoll nation that faces the loss of its entire territory to sea-level rise has a stake in how risk is framed and communicated that far exceeds its representation in existing global processes.
The Nature paper calls for the proposed assessment to actively embed local knowledge, loss-and-damage realities and adaptation limits from highly exposed regions, and to involve communities, civil society and Indigenous knowledge holders in the design, vetting and ground-truthing of the risk mapping exercise.4
David Obura, director of CORDIO East Africa and a co-author, brings to the paper a perspective grounded in the climate realities of the African continent and the Indian Ocean region, where the gap between formal assessments and lived exposure is stark.
The paper stops short of prescribing a specific governance architecture for translating global findings into national and municipal planning, though it implies that the assessment's value will depend entirely on its ability to cascade downward, into the budget cycles, infrastructure codes and emergency plans of the governments that need it most.
The paper also raises, without fully resolving, the tension between a process that names and quantifies vulnerability and the political resistance of governments that may not welcome the scrutiny.
Some form of risk-transparency mechanism, analogous to the peer review of national climate plans under the Paris Agreement, is implied rather than explicitly proposed.
Infrastructure, Finance and the Cost of Delay
One of the paper's most consequential arguments concerns infrastructure.
Roads, bridges, coastal defences, water treatment facilities and energy grids built or renovated today are being designed to engineering standards that reflect historical climate data, not the conditions those structures will face across their operational lifetimes.5
A bridge built in 2026 to a fifty-year flood standard may, by 2076, be operating in a climate where what was once a fifty-year flood occurs every decade.
This is not a hypothetical problem.
It is a systematic underinvestment in resilience that a global risk assessment could make visible and quantifiable.
The Nature paper aligns with growing concern among central banks, financial regulators and insurers about the systemic implications of climate-related financial risk.
The 2026 World Economic Forum Global Risks Report found that economic losses from weather and climate-related extremes have reached approximately $500 billion in Europe alone, with much of that damage remaining uninsured.6
The authors suggest that formal involvement of financial institutions in the risk assessment, through stress tests and risk product development, could help ensure that the findings translate into lending conditions, insurance pricing and investment criteria rather than simply remaining advisory.
This is the difference between a report that sits on a shelf and one that reprices risk in bond markets.
Communication, Fatigue and the Problem of Yet Another Report
The scientists are aware of the irony.
They are calling for a new global report in a world already saturated with global reports that have not catalysed the action they were designed to motivate.
The challenge of communication is therefore not peripheral to their proposal.
It is central to it.
Professor Peter Stott, the paper's other lead author and a climate scientist at the Met Office and the University of Exeter, argues that the world currently stands at a crossroads in global efforts to reduce emissions: "Bridging the current gap in global risk assessment is an urgent priority. An internationally mandated transparent assessment of avoidable climate change risks is essential to make clear the scale of the risks and the opportunities we have to avoid the worst-case scenarios and safeguard our shared future."1
The framing of the proposed assessment as one focused on avoidable risks, rather than inevitable catastrophe, is deliberate and important.
Risk assessments can be paralysing.
They can also be clarifying.
The difference depends almost entirely on whether they communicate agency: not just what might happen, but what remains within human power to prevent.
The paper insists that a global climate-risk assessment does not provide a counsel of despair.
It gives a clear picture of the outcomes that societies can still choose to avoid.
This framing matters especially in the information environment of 2026, characterised by entrenched political polarisation, AI-amplified misinformation and a public that has absorbed decades of climate warnings without feeling the full weight of what those warnings mean for their own lives, hospitals, water supply and food systems.
A risk assessment that speaks in the language of consequence, rather than the language of atmospheric physics, has a different chance of cutting through.
It also has a different vulnerability to misrepresentation.
Tail risks, by definition, carry wide uncertainty bands, and uncertainty has historically been weaponised by those with an interest in delay.
The paper's proposed framework, which focuses on impact thresholds and the range of outcomes under different emissions trajectories rather than on point predictions, is designed in part to be resistant to this distortion.
Honest uncertainty, communicated clearly, is not the same as ignorance.
The Role of Journalism and Independent Media
The paper's implications for journalists and independent media are significant, if not fully spelled out.
A global climate-risk assessment of the kind proposed would generate a body of data and analysis that is, in principle, more directly translatable into the kind of reporting that connects with lived experience: not global temperature anomalies but local flood probabilities, not radiative forcing but the likelihood that a specific city's water supply fails during a compound drought-heat event.
If such a process were to involve open-access data tools and explicit partnership with science journalists and investigative reporters, it could enable a level of climate-risk reporting that has so far remained aspirational.
Whether that partnership is built into the governance architecture of any future assessment remains to be seen.
Conclusion: The Reckoning Still to Come
What the scientists who published in Nature in February 2026 are really calling for is a reckoning.
Not a new collection of frightening statistics, though some frightening statistics will inevitably be part of it.
Not another synthesis of what is already known, though such a synthesis will be necessary.
What they are calling for is an honest, internationally mandated, interdisciplinary account of what is actually at stake if the world continues on its current trajectory, and what remains within reach if it changes course.
The absence of such an account is not an accident.
It reflects the difficulty of producing it, the political discomfort of confronting its conclusions and the institutional gap between the science of what is happening and the governance of what to do about it.
Professor Sutton captures the stakes with unusual directness: humanity still has the opportunity to shape "a more prosperous, liveable future," but a global assessment of avoidable climate-change risks is what would enable political leaders and citizens to understand what is at stake, and to "seize that opportunity, while we still have it."1
The clock, as every climate scientist knows, does not pause for institutional deliberation.
The question is whether the world's governments will commission a serious assessment of what they stand to lose before the losses become irreversible, or whether history will record that humanity had, and squandered, the information it needed to act.
The reckoning, in other words, is coming.
The only question is whether it arrives on our terms, or on the climate's.
References
- Stott, P. A., Lo, Y. T. E., Marsham, J. H., Obura, D., Oliver, T. H., Palmer, M. D., Ranger, N., Sharpe, S. & Sutton, R. (2026). "We need a global assessment of avoidable climate-change risks." Nature, 650, 826–828. doi: 10.1038/d41586-026-00544-6
- Sutton, R. T. (2019). "Climate science needs to take risk assessment much more seriously." Bulletin of the American Meteorological Society, 100, 1637–1642.
- Oliver, T. H., et al. (2026). Interconnected climate risks framework. Global Sustainability, 9, e4.
- Euronews. (27 February 2026). "The world remains unprepared: Why scientists are calling for a global assessment of climate change."
- University of Reading. (25 February 2026). "Update climate change risks list to avoid worst impacts."
- World Economic Forum. (2026). 2026 Global Risks Report: Environmental risks remain urgent.
- Met Office. (25 February 2026). "Global call to action: addressing the critical gap in climate change risk assessment."
- King, D., Schrag, D., Dadi, Z., Ye, Q. & Ghosh, A. (2015). Climate Change: A Risk Assessment. Cambridge University Centre for Science and Policy.
- Gambhir, A., et al. (2025). Interconnected climate risk modelling. Nature Communications, 16, 7382.
- DIO East Africa. (25 February 2026). "Global Climate Change Risk Assessment."