Strategic Climate Risks Initiative

@climate-risk.bsky.social

SCRI is a think-and-do tank for navigating the next chapter of the climate crisis, producing insights and tools for navigating non-linear and cascading climate risk. Find out more: https://www.scri.org.uk/

The infrastructure that 1.5°C overshoot pathways depend on has no political constituency, and the people most motivated to see it built are currently its most effective opponents. Political economy remains the biggest and least acknowledged challenge to CO2 removal. Read more in the comments.

Image of 3 pipelines in a forest.

Climate risk can erode the social contract. New evidence from sub-Saharan Africa shows droughts and extreme heat lower tax morale via inequality and trust erosion. For flooding, more visible responses reduced this effect. Protecting state legitimacy may be a key policy challenge in the coming years.

Diagram showing how climate disasters can impact tax morale through inequality and trust.

What should we do when we pass 1.5°C of warming? Double down with stubborn optimism? Pivot to national security? Accelerate adaptation? Pile investment into carbon dioxide removal and solar geoengineering? Or just close our eyes very tight and hope AI will solve it all? luma.com/xbjxedgd

Overshoot: What's the climate strategy above 1.5°C? · Luma

The world missed its goal for tackling climate change. What happens next? Overshooting 1.5C creates profound challenges, including what a new goal should be…

luma.com

Lieutenant General (retd) Richard Nugee CB CVO CBE spent 36 years in the British Army. He oversaw the MOD's landmark Climate Change and Sustainability Review and he is unequivocal: climate change is the most insidious security threat we face, and we are wholly unprepared for it.

The evidence base for derailment dynamics is growing. A new report highlights "epistemic stress" of crises in the Indo-Pacific; partisan disputes, blame and denial narratives corroded emergency coordination during major fires and floods. Worsening impacts eroding the capacity to respond.

Front cover of Toda Peace Institute report showing a forest fire. Title is "Trust, governance and climate disasters in the Indo-Pacific" by Sohail Akhtar.

Planning for a super El Niño as a temporary deviation from normal misses the real risk. A 2025 study found these events can persistently shift climate baselines. If one forms this year, the impacts could extend far beyond short-term disruption, however extreme. Link to the paper in thread.

Graph showing spatial patterns of climate regime shifts and their increases in probability after super El Nino events.

Nordic discussions have cited stable regional ocean circulation as evidence against AMOC collapse. New research reframes stability as a precursor to coupled failure. Nonlinear dynamics create security blind spots; we need strategies and early warning systems that account for them. Paper in thread.

Graph showing AMOC and NOC circulation strength, showing stability in NOC while AMOC declines up to a threshold, before abrupt decline in both.

IR scholars have called for risk research around SRM to move beyond purely comparing geophysical risks, to also factoring in conflictual geopolitics. This is the only way to ensure research can support effective governance discussions around this world shaping technology. Paper linked in thread.

Two planes, one marked US one marked China, flying above the earth creating aerosol plumes.

Governments are planning for war. Meanwhile, two new studies show AMOC collapse could be the bigger threat. Weakening of over 50% by 2100 is expected, with an irreversible tipping point as early as mid-century. As we noted in the Security Blind Spot report, security infrastructure isn't prepared.

A new study by PIK shows AMOC collapse would flip the Southern Ocean from carbon sink to carbon source, adding ~0.2°C of warming. That's on top of wherever temperatures already are when it happens. Tipping points risks cascade in ways that are often under-appreciated. Link in thread.

Picture of blue ocean waves.