A target, it turns out, can be two entirely different things. It can be the finish line (the point at which you stop and declare success). Or it can be a waypoint, a necessary passage on a longer road whose end you cannot yet see clearly. Climate science has spent decades insisting that 1.5 degrees of warming is the number to hold, the ceiling above which things get genuinely bad. Two new studies from the International Institute for Applied Systems Analysis suggest, with some force, that 1.5°C is a waypoint. Possibly one we’ll miss. And either way, not the end.
The problem the researchers are circling is a bit unglamorous, as climate problems go. Not dramatic wildfires or flooded capitals, but the quieter physics of systems that respond slowly. Sea levels don’t rise the moment temperatures spike. Permafrost doesn’t thaw on the same schedule as the atmosphere above it. These are time-lagged impacts, the climate’s IOUs, and they keep accruing even when the headline temperature number stops climbing.
What Johannes Bednar and colleagues at IIASA set out to establish, in a paper published in Environmental Research Letters, is what it would actually take to stop those IOUs from growing. Their starting point is an advisory opinion handed down by the International Court of Justice in July 2025, which found that states have binding legal obligations to prevent significant climate harm. Sea-level rise and permafrost thaw, the researchers argue, are precisely the “significant harm” the court had in mind. Coastal infrastructure damage alone is projected to run to hundreds of billions of dollars annually. Permafrost, which blankets roughly 20 to 25 percent of the Northern Hemisphere’s land area, stores nearly twice as much carbon as the entire atmosphere currently holds; its thaw pumps that carbon back out as CO2 and methane, accelerating the very warming driving its melt. The feedback loop is not theoretical.
The conclusion that falls out of Bednar’s modeling is uncomfortable. Net-zero emissions, the stated goal of most national climate commitments, won’t stop sea levels or permafrost thaw from getting worse. To actually halt their growth, global emissions would need to go net-negative and stay there. Not for a decade or two. For centuries.
“They represent significant harm affecting hundreds of millions of people through sea-level rise and infrastructure damage,” Bednar told the IIASA press office. “If we take the Court’s opinion seriously, stabilizing these impacts implies a long-term commitment to carbon removal that goes far beyond today’s targets.”
The paper’s underlying logic is a little counterintuitive at first. Think of sea-level rise as a moving object, not a static measurement. At the point where temperatures peak, that object is already in motion, carrying the momentum of everything that’s happened to the ice sheets and oceans up to that moment. Returning to 1.5°C doesn’t immediately stop it; the object keeps going, because the physics of large ice bodies and ocean heat uptake don’t respond on human-policy timescales. To bring the object to a halt, you need to keep braking: sustained CO2 removal from the atmosphere, pulling temperatures down further until the residual momentum bleeds out. The paper models four scenarios with different technological assumptions about how fast emissions can fall and how much carbon removal capacity we can deploy, and in all four, net-negative emissions extend well past the 23rd century before the time-lagged impacts finally stabilize.
A companion paper, published in Nature Communications and led by Thomas Gasser (also at IIASA), reaches the same destination via a different route. Gasser’s team modelled optimal climate policy under genuine uncertainty about how the Earth system will respond, an uncertainty that mainstream integrated assessment models typically paper over. When that uncertainty is factored in properly, the economically rational response shifts significantly: net-zero targets arrive roughly a decade sooner, near-term carbon prices go up substantially, and the long-run pathway still requires centuries of negative emissions. Not, Gasser emphasizes, as a response to temporary overshoot, but as basic hedging against the risk that Earth system feedbacks turn out worse than central estimates suggest.
“Uncertainty in the climate system calls for precaution,” he told IIASA’s press office, “much like uncertainty in income encourages more careful spending.” It’s a disarmingly domestic analogy for a civilizational problem, but the logic is sound.
What neither paper can resolve, and both acknowledge, is that centuries-long emission removal on the scale required demands institutional machinery that doesn’t yet exist. The emissions cuts we’re currently struggling to implement take decades of political will and infrastructure investment. Bednar’s scenario modeling shows that delays in near-term mitigation compound the removal burden later; the “multiple failures” scenario, where both emission reductions and carbon removal technology fall short, produces not just higher peak sea levels but a much wider spread of possible outcomes. Delayed action doesn’t just make the median case worse; it makes governance harder, because the range of possible futures gets larger. Coauthor Artem Baklanov is fairly direct about what that means: “The Paris temperature goals should be understood as milestones rather than endpoints. Stabilization of delayed impacts requires temperatures to eventually decline below 1.5°C, which in turn requires sustained net-negative emissions well beyond 2100.”
The governance challenge is arguably the harder part. Robust institutions capable of persisting across political cycles, allocating carbon removal obligations fairly between wealthy and vulnerable nations, linking present emissions to future removal commitments across generations: these are not technical problems. They’re political and legal ones, and the researchers note, with some exasperation, that they remain largely absent from current climate negotiations.
There’s a particular sting in the permafrost numbers. Even in the most optimistic scenario (full technology deployment, Paris-consistent trajectories) cumulative CO2 emissions from permafrost thaw are projected to account for around five percent of the total carbon removal burden by the time stabilization is achieved. In the worst-case scenario, that figure climbs above 30 percent. Permafrost isn’t merely a victim of climate change; at scale, it becomes a driver, requiring removal efforts to chase a target that’s still moving.
What the two studies together amount to is a fairly systematic dismantling of the idea that climate targets are endpoints. The legal framing of the IIASA paper and the economic welfare framing of the Nature Communications paper start from different premises and use different tools, but both land on centuries of net-negative emissions as the requirement, not the ambition. Whether the political and institutional architecture to sustain that over such timescales can be built, and built fairly, across nations at different stages of development and different exposure to sea-level risk, is a question that 2025’s climate negotiations barely touched. It may be the defining question of the century after this one.
DOI / Source: https://doi.org/10.1088/1748-9326/ae34ca
Frequently Asked Questions
Net-negative emissions means that humanity is removing more CO2 from the atmosphere than it is adding. Getting to net zero means emissions and removals are balanced; going net negative means the balance tips so that the total atmospheric concentration is actually falling. Technologies involved include direct air capture, bioenergy with carbon capture and storage, and enhanced weathering, alongside natural sinks like forests and soils.
The 1.5°C target addresses the peak temperature, the ceiling we’re trying not to breach. But some climate impacts, particularly sea-level rise and permafrost thaw, don’t respond to temperature on human timescales. They carry momentum from past warming and keep getting worse even after temperatures stabilize. These studies argue that to stop those particular impacts from worsening, temperatures eventually need to come back down below 1.5°C, which requires going net negative for a very long time after the peak.
Permafrost is frozen ground that covers roughly 20 to 25 percent of the Northern Hemisphere’s land surface, including large parts of Siberia, Alaska, and northern Canada. It stores nearly twice as much carbon as is currently in the atmosphere. As it thaws due to warming, microbes decompose the organic material, releasing CO2 and methane. This creates a feedback loop where warming causes thaw, which releases more greenhouse gases, which cause more warming, and under the worst-case scenarios modeled here, permafrost thaw could account for more than 30 percent of the total carbon removal burden we’d need to achieve.
There’s significant uncertainty in the precise timescales, which the studies are candid about. The range of phase-out dates for net-negative emissions spans at least two centuries within each scenario, reflecting both physical uncertainties about how the climate system will respond and technological uncertainties about how much carbon removal capacity can realistically be deployed. What the studies find is that across all scenarios modeled, optimistic and pessimistic alike, net-negative emissions extend well past the 23rd century. The uncertainty is in when, not whether.
The researchers don’t have a blueprint, and honestly admit as much. What they argue is that current climate negotiations have barely started thinking about this. Such institutions would need to persist across political cycles (governments typically plan over years to decades, not centuries), allocate removal obligations fairly between rich and poor nations, and create legal or financial mechanisms that tie present emissions to future removal commitments across generations. Analogies might include long-term stewardship institutions developed for nuclear waste sites, or international bodies for managing shared ocean resources: imperfect parallels, but the closest existing models we have.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.
Key Takeaways
- Research suggests that 1.5°C is just a waypoint, not an endpoint for climate action.
- To halt sea-level rise and permafrost thaw, global emissions must go net-negative for centuries.
- Current climate commitments won’t stop delayed impacts, leading to significant harm and economic costs.
- Long-term political and institutional frameworks are essential for managing net-negative emissions.
- Both studies argue that climate targets should be seen as milestones, necessitating sustained carbon removal beyond 2100.