Net Zero Is Not the Finish Line: Studies Say Carbon Removal Must Continue for Centuries

Politicians and corporate leaders have cast “net zero by 2050” as the finish line in the race against climate change. New research suggests it is closer to the halfway mark.

Two coordinated studies released in recent weeks conclude that even if the world succeeds in cutting greenhouse gas emissions to net zero and meeting the temperature goals of the Paris Agreement, humanity will still need to pull more carbon dioxide out of the air than it puts in—and keep doing so for centuries—to halt some of the most persistent climate threats.

The work, led by scientists at the International Institute for Applied Systems Analysis (IIASA) in Austria and collaborators in the United Kingdom and elsewhere, uses climate–economy models and legal analysis to examine what it would take not just to stabilize global temperatures, but to stop long-term impacts such as sea-level rise and thawing permafrost from worsening.

“In all four scenarios, net-negative emissions must be sustained far beyond the 23rd century—even under optimistic assumptions,” said Johannes Bednar, a researcher at IIASA and the University of Oxford, in an email interview. “Net zero is not the end point of climate policy; it is only the beginning of a long-term effort to reduce the stock of carbon in the atmosphere.”

One paper, published Jan. 26 in Environmental Research Letters, focuses on what the authors call “time-lagged climate impacts,” mainly global mean sea-level rise and carbon emissions from thawing Arctic permafrost. The second, published online in March in Nature Communications, looks at how uncertainty in the Earth system—from climate sensitivity to permafrost feedbacks—changes what an economically “optimal” climate pathway should look like.

Taken together, the results challenge the way governments and companies frame net-zero pledges and add weight to arguments that current climate plans are incomplete.

Paris targets as milestones, not endpoints

The Environmental Research Letters study, “Stabilizing time-lagged climate impacts requires net-negative emissions for centuries,” asks what global emissions pathway would be needed if countries aim not only to limit temperature rise to 1.5 to 2 degrees Celsius, as in the 2015 Paris Agreement, but also to halt the increase of slower-responding impacts.

Sea-level rise lags behind surface warming by decades to centuries as ice sheets melt and oceans expand. Large areas of permafrost—frozen soils that cover about a quarter of the Northern Hemisphere’s land surface and store almost twice as much carbon as the atmosphere—thaw slowly and can emit carbon dioxide and methane for centuries.

“Even if we stopped emissions tomorrow, sea level would continue to rise for a very long time,” Bednar said. “If we take seriously our duty to avoid further significant harm, that means we must consider drawing down atmospheric CO₂, not just stopping emissions.”

To explore what that would entail, the team coupled an integrated assessment model—which links economic activity and emissions to climate outcomes—with a simplified Earth system model. They tested four stylized global mitigation scenarios, ranging from a stringent “Full Paris” path with strong near-term cuts and large-scale carbon removal, to a “Multi-failure” path with weaker efforts and limited removal capacity.

In their most optimistic case, global temperatures peak at around 1.7 degrees Celsius above pre-industrial levels this century and then gradually decline toward about 0.5 degrees Celsius as net-negative emissions continue. In less ambitious cases, peak warming is higher and long-term sea-level rise and permafrost emissions are larger.

Across all cases, however, one conclusion held: stabilizing sea level and permafrost-related emissions at the lowest attainable levels required maintaining global net-negative CO₂ emissions—meaning more CO₂ is removed from the atmosphere than emitted—for more than 200 years.

The study uses a simple analogy: peaking temperature is like hitting the brakes on a heavy vehicle rolling downhill. Net-negative emissions are the braking force. How far the vehicle travels before stopping—how much more sea level rises, how much more permafrost carbon is released—depends on how hard and how long the brakes are applied.

“Just reaching net zero slows the vehicle; it does not stop it,” Bednar said. “To actually bring long-term impacts to a halt, you have to keep pushing below zero for centuries.”

Law and “significant harm”

The paper places this scientific finding in the context of a landmark advisory opinion issued on July 23, 2025, by the International Court of Justice (ICJ) in The Hague. Responding to a request from the United Nations General Assembly, the ICJ concluded that states have binding obligations under international law to prevent “significant harm” from climate change, including where there is a risk of future significant harm, and to use “all means at their disposal” to do so.

Bednar and his co-authors, including legal scholar Justin Macinante of the University of Edinburgh, argue that rising seas and permafrost thaw clearly qualify as risks of future significant harm. If so, they say, the ICJ’s language could be interpreted as implying not only a duty to cut emissions rapidly, but also a duty to support sustained net-negative emissions to stabilize those impacts.

“Our results suggest that, in light of the ICJ opinion, states’ obligations cannot reasonably be limited to achieving net zero,” the authors write. “They extend to the long-term reduction of atmospheric CO₂ concentrations.”

The paper notes that most national climate pledges under the Paris Agreement—known as nationally determined contributions—rarely include explicit goals for gross carbon dioxide removal, focusing instead on net balances up to mid-century. It highlights what the authors call a “governance gap” for managing a net-negative carbon economy that would have to operate across multiple centuries.

Negative emissions as insurance, not a free pass

The companion study in Nature Communications, led by IIASA scientist Thomas Gasser, tackles a different but related question: How does uncertainty about the Earth system alter what a rational, risk-averse climate strategy should look like?

“Most integrated assessment models treat key Earth system properties—like climate sensitivity or the strength of carbon sinks—as if they were known with certainty,” Gasser said in a statement issued by IIASA. “But in reality, we face deep uncertainty. The prudent approach is to design policy up front to manage that risk.”

To do that, the team embedded a simplified climate model that includes ocean and land carbon sinks and permafrost emissions into an economic framework, and used statistical techniques to represent uncertainty in climate parameters based on Earth system models and observations.

They then compared two types of strategies. In an “ex-post” approach, common in the existing literature, policymakers choose an emissions pathway based on best-estimate parameters and later evaluate how it performs under uncertainty. In an “ex-ante” approach, they instead optimize policy over the full distribution of possible Earth responses, explicitly aiming for, say, a 95% chance of staying below 2 degrees Celsius.

The ex-ante strategies turned out to be significantly more ambitious in the near term. In one cost–benefit case, the ex-post solution reached net-zero CO₂ around 2064, with a global carbon price of about $250 per ton in 2030. The ex-ante solution, designed under uncertainty, reached net zero around 2041 and implied a 2030 carbon price closer to $425 per ton.

In a scenario targeting at least a 95% probability of keeping warming below 2 degrees Celsius, the ex-ante strategy achieved net zero around 2059 and then deployed roughly 980 billion tons of net-negative CO₂ over the long term—nearly 25 years of today’s global emissions.

“Negative emissions should be seen as a safety net against unfavorable outcomes of the Earth system,” the authors write, “and not as an excuse to delay mitigation efforts and overshoot climate targets.”

The difference between ex-ante and ex-post carbon prices, they argue, can be interpreted as a “precautionary premium” that societies must pay because they do not know exactly how sensitive the planet is or how strong feedbacks like permafrost thaw will be.

A multi-century governance problem

Both studies converge on a similar picture: a world that succeeds in cutting emissions to net zero by mid-century, as many governments have pledged, is not done with climate policy. Instead, it enters a “net-negative era” in which active carbon removal—through methods such as reforestation, bioenergy with carbon capture, direct air capture, enhanced weathering, or other technologies—becomes a permanent component of managing the climate.

That raises practical questions far beyond engineering.

Running a global carbon-removal system for centuries implies long-lived financial instruments and institutions that can guarantee, monitor and maintain storage sites; mechanisms to handle the risk that CO₂ put into forests or soils might leak back out; and agreed rules for how responsibilities are shared across countries and generations.

Under United Nations climate rules adopted in recent years, projects that generate carbon credits from removals must already maintain buffer reserves and monitoring for decades to manage “reversal risk.” Extending that logic to centuries is a larger challenge.

Earlier work by some of the same authors has proposed “carbon removal obligations,” in which fossil fuel producers or emitters would incur a legal duty to finance the later removal of a quantity of CO₂ equivalent to what they put into the air. Without such tools, the IIASA team warns, the cost of removing historical emissions could fall largely on future taxpayers, turning carbon removal into what they describe as a public waste-management task imposed on generations yet to be born.

Outside experts have long cautioned against overreliance on negative emissions to compensate for weak near-term cuts. A 2014 analysis led by Sabine Fuss at the Mercator Research Institute on Global Commons and Climate Change in Berlin, for example, warned that many scenarios for limiting warming to 2 degrees Celsius depended on large-scale deployment of bioenergy with carbon capture and storage, a technology still in its infancy and with significant land-use implications.

The new work does not dispute those concerns. Instead, it suggests that some level of large-scale, long-lasting carbon removal is now a physical necessity if governments want not only to slow global warming, but to stabilize the planet’s long-term response.

For policymakers and the public, that reframes what success looks like. Reaching net zero by 2050—if it happens—would mark a crucial turning point. But in light of the latest science, it would not be the end of the climate story. It would be the beginning of a new, longer chapter in which societies must decide how much carbon to pull back out of the sky, for how long, and who will bear that burden.