For most of the climate era, the central instruction has been straightforward: stop putting so much carbon dioxide into the atmosphere.
COP31 is increasingly confronting a second question.
What if we also have to take enormous amounts back out?
Carbon dioxide removal, or CDR, has moved further into the negotiations in Antalya as governments confront the growing possibility that global temperatures will temporarily exceed the Paris Agreement's 1.5°C threshold.
For the first time, carbon removal and carbon capture and storage have appeared in draft COP31 text under discussion.
That does not mean the world has discovered a technological undo button for climate change.
The numbers suggest something much more difficult.
Carbon removal is not one technology
Carbon removal describes methods that take carbon dioxide already in the atmosphere and store it for a meaningful period.
Some are familiar.
Forests absorb carbon as they grow. Restoring ecosystems and improving soil management can increase the amount of carbon held by natural systems.
Others are technological.
Direct air capture uses machines and chemical processes to extract CO₂ directly from ambient air before storing it underground or using it in products.
Enhanced rock weathering attempts to accelerate natural chemical processes through which certain minerals react with carbon dioxide.
These approaches vary enormously in cost, permanence, land requirements and technological maturity.
They also need to be distinguished from carbon capture and storage, or CCS.
CCS generally captures carbon dioxide from power plants or industrial facilities before it reaches the atmosphere.
CDR attempts to remove carbon that is already there.
That difference matters.
The scale today is nowhere near the scale being discussed
Human activity already removes substantial quantities of carbon through conventional land-based methods, particularly forests.
But newer engineered removal technologies remain tiny.
Current novel carbon-removal methods remove only around two million tonnes of CO₂ each year, according to estimates highlighted around COP31.
Compare that with what could eventually be required.
UN climate analysis suggests that bringing temperatures back towards 1.5°C after an overshoot could require removing and permanently storing around 320 billion tonnes of CO₂ over this century, alongside very deep reductions in ongoing emissions.
The distance between those numbers explains why carbon removal is simultaneously attracting investment and scepticism.
The world may need it.
But needing a technology in a climate model does not guarantee that it can be deployed cheaply, safely or quickly enough in the real world.
Why would we need removal at all?
The reason lies in how carbon dioxide behaves.
CO₂ accumulates in the atmosphere. As long as humanity continues adding more than natural and engineered systems remove, atmospheric concentrations keep increasing.
Reaching net zero means bringing those additions and removals broadly into balance.
But some emissions will be exceptionally difficult to eliminate completely.
Agriculture, aviation, cement production and some industrial processes may continue producing residual emissions even in a deeply decarbonised economy.
Carbon removal could compensate for some of those emissions.
Its role becomes larger if the world overshoots 1.5°C and later attempts to reduce global temperatures.
At that point, reaching net zero would no longer be enough.
The world would need sustained net-negative emissions: removing more greenhouse gases than it adds.
That is a much more demanding proposition.
The danger is treating tomorrow's removal as permission for today's emissions
This is where the argument becomes political.
If governments and industries assume enormous amounts of carbon can be removed later, they may feel less pressure to reduce fossil-fuel emissions quickly today.
That creates a risk.
Future societies could inherit responsibility for building an enormous carbon-removal industry because earlier generations chose slower emissions reductions.
There are also practical limits.
Large-scale tree planting competes for land needed for food and ecosystems.
Direct air capture requires energy and remains expensive.
Permanent underground storage requires suitable geology, infrastructure, monitoring and public acceptance.
Other emerging techniques raise questions about environmental consequences that are still being studied.
None of this means carbon removal should be rejected.
It means its role needs to be understood accurately.
Removal and emissions cuts solve different parts of the problem
The easiest tonne of carbon to deal with is often the tonne that never enters the atmosphere.
Replacing a coal power plant with clean electricity prevents future emissions.
Capturing carbon from the air after fossil fuels have already been burned requires additional infrastructure, energy and money.
That is why scientists generally treat carbon removal as a complement to deep emissions reductions rather than a substitute for them.
It may be essential for balancing genuinely difficult residual emissions.
It could also become necessary if temperatures overshoot 1.5°C.
But the more slowly emissions fall now, the larger the removal challenge future generations may inherit.
COP31 is beginning a much bigger conversation
The presence of carbon removal in COP31 discussions matters because climate policy is gradually entering unfamiliar territory.
The world spent decades debating how to prevent warming.
It is now increasingly discussing how to manage an overshoot and, potentially, reverse part of it.
That brings new technologies, industries and financial interests into climate negotiations.
It also creates a new question for governments.
How much should climate policy rely on technologies that may become essential but do not yet exist at anything close to the scale future pathways could demand?
There is no simple answer.
Carbon removal may become an important part of the climate response.
But there is a dangerous difference between preparing to remove unavoidable emissions and assuming future technology will clean up emissions we could avoid today.
COP31 is beginning to confront that distinction.
The atmosphere will eventually reveal whether we understood it.
Was this article helpful?
Your feedback helps us improve our journalism.